Code_project_1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "<h1>\n",
        "Classification of Social Behavior 🐁🐀 - Elías Aburto Camacllanqui\n",
        "</h1>\n",
        "This code is adapted from a notebook created by Dipam Chakraborty at AIcrowd for the <a href=https://www.aicrowd.com/challenges/multi-agent-behavior-representation-modeling-measurement-and-applications>Multi-Agent Behavior Challenge</a>."
      ],
      "metadata": {
        "id": "283zrrSAjtb-"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6SgPerxMNnas"
      },
      "outputs": [],
      "source": [
        "# import some libraries\n",
        "import os\n",
        "import requests\n",
        "import zipfile\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.layers import LSTM, Dense, Dropout, Input"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Function to download some archives\n",
        "def download_file(url, fname):\n",
        "    if not os.path.isfile(fname):\n",
        "        r = requests.get(url)\n",
        "        if r.status_code == 200:\n",
        "            with open(fname, 'wb') as f:\n",
        "                f.write(r.content)\n",
        "        else:\n",
        "            print(f\"Failed to download file: {fname}. Status code: {r.status_code}\")\n",
        "\n",
        "def download_and_unzip(url, fname):\n",
        "    download_file(url, fname)\n",
        "    if fname.endswith('.zip'):\n",
        "        try:\n",
        "            with zipfile.ZipFile(fname, 'r') as zip_ref:\n",
        "                zip_ref.extractall('.')\n",
        "        except zipfile.BadZipFile:\n",
        "            print(f\"Error: The file {fname} is not a zip file.\")"
      ],
      "metadata": {
        "id": "L1PGXx4uNpe6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Download archives\n",
        "download_and_unzip(\"https://data.caltech.edu/records/s0vdx-0k302/files/task1_classic_classification.zip?download=1\", \"task1_classic_classification.zip\")\n",
        "download_file(\"https://data.caltech.edu/records/s0vdx-0k302/files/calms21_convert_to_npy.py?download=1\", \"calms21_convert_to_npy.py\")\n",
        "\n",
        "# Excecute the script to converte data\n",
        "!python calms21_convert_to_npy.py --input_directory '.' --output_directory '.'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qmQsDUBfNtcv",
        "outputId": "0c8c4783-522c-43fb-c06c-e0641bce7056"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving ./calms21_task1_train\n",
            "Saving ./calms21_task1_test\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load data\n",
        "def load_data(file_path):\n",
        "    data_dict = np.load(file_path, allow_pickle=True).item()\n",
        "    dataset = next(iter(data_dict.values()))\n",
        "    sequence_id = next(iter(dataset.keys()))\n",
        "    vocabulary = dataset[sequence_id]['metadata']['vocab']\n",
        "    return dataset, vocabulary\n",
        "\n",
        "training_data, vocab = load_data('calms21_task1_train.npy')\n",
        "test_data, _ = load_data('calms21_task1_test.npy')"
      ],
      "metadata": {
        "id": "DPBXdVWTNtwj"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Function to Preprocessing\n",
        "\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "def prepare_data(data):\n",
        "    X = [data[seq]['keypoints'].reshape(len(data[seq]['annotations']), -1) for seq in data]\n",
        "    y = [data[seq]['annotations'] for seq in data]\n",
        "\n",
        "    # Padding to ensure all sequences have the same length\n",
        "    X_padded = pad_sequences(X, padding='post', dtype='float32')\n",
        "    y_padded = pad_sequences(y, padding='post', value=3)\n",
        "\n",
        "    return X_padded, y_padded"
      ],
      "metadata": {
        "id": "jGr3uM7xOSuF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Preprocessing data\n",
        "\n",
        "X_train, y_train = prepare_data(training_data)\n",
        "X_test, y_test = prepare_data(test_data)"
      ],
      "metadata": {
        "id": "0mAtkEe2OS0_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Customize a function to ignore class 3 (padding)\n",
        "import tensorflow as tf\n",
        "\n",
        "def custom_loss(y_true, y_pred):\n",
        "    # Create a mask to ignore class 3\n",
        "    mask = tf.cast(tf.not_equal(y_true, 3), dtype=tf.float32)\n",
        "\n",
        "    # Calculate sparse categorical cross-entropy loss\n",
        "    loss = tf.keras.losses.sparse_categorical_crossentropy(y_true, y_pred)\n",
        "\n",
        "    # Apply the mask to ignore loss where the class is 3\n",
        "    loss = loss * mask\n",
        "\n",
        "    # Normalize by the number of valid samples (to avoid bias from ignored 3s)\n",
        "    return tf.reduce_sum(loss) / tf.reduce_sum(mask)"
      ],
      "metadata": {
        "id": "oAkLUTwHY8Z-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Define and compile the model using the input layer.\n",
        "input_layer = Input(shape=(X_train.shape[1], X_train.shape[2]))\n",
        "x = LSTM(64, return_sequences=True)(input_layer)\n",
        "x = LSTM(64, return_sequences=True)(x)\n",
        "x = Dropout(0.5)(x)\n",
        "x = Dense(64, activation='relu')(x)\n",
        "output_layer = Dense(len(vocab), activation='softmax')(x)\n",
        "\n",
        "model = Model(inputs=input_layer, outputs=output_layer)\n",
        "model.compile(optimizer='adam', loss=custom_loss, metrics=['accuracy'])\n",
        "\n",
        "# loss='sparse_categorical_crossentropy'"
      ],
      "metadata": {
        "id": "TXlWUDzCOiA2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Entrenar el modelo\n",
        "history = model.fit(X_train, y_train, validation_split=0.2, epochs=500, batch_size=32)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5gxgKvmPPwn1",
        "outputId": "a9e257de-230c-487e-dea6-2f1f17c60a83"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0698 - loss: 1.4239 - val_accuracy: 0.1563 - val_loss: 1.0794\n",
            "Epoch 2/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0817 - loss: 1.1814 - val_accuracy: 0.1547 - val_loss: 0.8067\n",
            "Epoch 3/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0881 - loss: 1.0158 - val_accuracy: 0.1547 - val_loss: 0.6203\n",
            "Epoch 4/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0860 - loss: 0.8778 - val_accuracy: 0.1547 - val_loss: 0.5198\n",
            "Epoch 5/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0803 - loss: 0.8749 - val_accuracy: 0.1547 - val_loss: 0.4792\n",
            "Epoch 6/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0857 - loss: 0.8622 - val_accuracy: 0.1547 - val_loss: 0.4740\n",
            "Epoch 7/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0789 - loss: 0.8571 - val_accuracy: 0.1547 - val_loss: 0.4849\n",
            "Epoch 8/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0836 - loss: 0.8217 - val_accuracy: 0.1547 - val_loss: 0.5131\n",
            "Epoch 9/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0811 - loss: 0.8295 - val_accuracy: 0.1547 - val_loss: 0.5485\n",
            "Epoch 10/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0785 - loss: 0.8020 - val_accuracy: 0.1547 - val_loss: 0.5697\n",
            "Epoch 11/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0810 - loss: 0.8065 - val_accuracy: 0.1547 - val_loss: 0.5684\n",
            "Epoch 12/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0808 - loss: 0.7234 - val_accuracy: 0.1547 - val_loss: 0.5375\n",
            "Epoch 13/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0888 - loss: 0.7566 - val_accuracy: 0.1547 - val_loss: 0.5196\n",
            "Epoch 14/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0893 - loss: 0.7326 - val_accuracy: 0.1547 - val_loss: 0.5122\n",
            "Epoch 15/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0895 - loss: 0.7492 - val_accuracy: 0.1547 - val_loss: 0.5221\n",
            "Epoch 16/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0859 - loss: 0.7352 - val_accuracy: 0.1547 - val_loss: 0.5238\n",
            "Epoch 17/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0916 - loss: 0.7312 - val_accuracy: 0.1547 - val_loss: 0.5180\n",
            "Epoch 18/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0865 - loss: 0.7405 - val_accuracy: 0.1547 - val_loss: 0.5144\n",
            "Epoch 19/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0888 - loss: 0.7077 - val_accuracy: 0.1547 - val_loss: 0.4922\n",
            "Epoch 20/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0923 - loss: 0.7000 - val_accuracy: 0.1540 - val_loss: 0.4866\n",
            "Epoch 21/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0794 - loss: 0.7364 - val_accuracy: 0.1534 - val_loss: 0.4993\n",
            "Epoch 22/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0873 - loss: 0.6644 - val_accuracy: 0.1537 - val_loss: 0.4563\n",
            "Epoch 23/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0900 - loss: 0.6617 - val_accuracy: 0.1528 - val_loss: 0.4666\n",
            "Epoch 24/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0811 - loss: 0.7093 - val_accuracy: 0.1490 - val_loss: 0.5000\n",
            "Epoch 25/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0861 - loss: 0.6746 - val_accuracy: 0.1505 - val_loss: 0.4681\n",
            "Epoch 26/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0857 - loss: 0.6750 - val_accuracy: 0.1506 - val_loss: 0.4577\n",
            "Epoch 27/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0914 - loss: 0.6168 - val_accuracy: 0.1483 - val_loss: 0.4820\n",
            "Epoch 28/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0932 - loss: 0.6468 - val_accuracy: 0.1495 - val_loss: 0.4751\n",
            "Epoch 29/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0935 - loss: 0.6164 - val_accuracy: 0.1505 - val_loss: 0.4582\n",
            "Epoch 30/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0936 - loss: 0.5951 - val_accuracy: 0.1475 - val_loss: 0.4754\n",
            "Epoch 31/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0853 - loss: 0.6446 - val_accuracy: 0.1474 - val_loss: 0.4658\n",
            "Epoch 32/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0906 - loss: 0.5904 - val_accuracy: 0.1518 - val_loss: 0.4277\n",
            "Epoch 33/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0957 - loss: 0.5949 - val_accuracy: 0.1361 - val_loss: 0.5469\n",
            "Epoch 34/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0873 - loss: 0.6090 - val_accuracy: 0.1445 - val_loss: 0.4820\n",
            "Epoch 35/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0949 - loss: 0.5658 - val_accuracy: 0.1521 - val_loss: 0.4185\n",
            "Epoch 36/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0968 - loss: 0.5681 - val_accuracy: 0.1398 - val_loss: 0.5135\n",
            "Epoch 37/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0888 - loss: 0.5653 - val_accuracy: 0.1477 - val_loss: 0.4441\n",
            "Epoch 38/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0936 - loss: 0.5577 - val_accuracy: 0.1455 - val_loss: 0.4588\n",
            "Epoch 39/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0864 - loss: 0.5639 - val_accuracy: 0.1406 - val_loss: 0.5000\n",
            "Epoch 40/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0928 - loss: 0.5676 - val_accuracy: 0.1503 - val_loss: 0.4250\n",
            "Epoch 41/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0963 - loss: 0.5346 - val_accuracy: 0.1441 - val_loss: 0.4690\n",
            "Epoch 42/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0978 - loss: 0.5503 - val_accuracy: 0.1505 - val_loss: 0.4182\n",
            "Epoch 43/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0938 - loss: 0.5482 - val_accuracy: 0.1525 - val_loss: 0.3927\n",
            "Epoch 44/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0941 - loss: 0.5693 - val_accuracy: 0.1435 - val_loss: 0.4839\n",
            "Epoch 45/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0933 - loss: 0.5577 - val_accuracy: 0.1472 - val_loss: 0.4443\n",
            "Epoch 46/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0931 - loss: 0.5459 - val_accuracy: 0.1525 - val_loss: 0.3918\n",
            "Epoch 47/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0937 - loss: 0.5483 - val_accuracy: 0.1502 - val_loss: 0.4206\n",
            "Epoch 48/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0990 - loss: 0.5408 - val_accuracy: 0.1415 - val_loss: 0.5140\n",
            "Epoch 49/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3s/step - accuracy: 0.0871 - loss: 0.5920 - val_accuracy: 0.1515 - val_loss: 0.3973\n",
            "Epoch 50/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0974 - loss: 0.5329 - val_accuracy: 0.1513 - val_loss: 0.3989\n",
            "Epoch 51/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0923 - loss: 0.5577 - val_accuracy: 0.1447 - val_loss: 0.4595\n",
            "Epoch 52/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0991 - loss: 0.5119 - val_accuracy: 0.1516 - val_loss: 0.3913\n",
            "Epoch 53/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0973 - loss: 0.4843 - val_accuracy: 0.1525 - val_loss: 0.3826\n",
            "Epoch 54/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1038 - loss: 0.4877 - val_accuracy: 0.1478 - val_loss: 0.4426\n",
            "Epoch 55/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0920 - loss: 0.4885 - val_accuracy: 0.1489 - val_loss: 0.4275\n",
            "Epoch 56/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0940 - loss: 0.4932 - val_accuracy: 0.1519 - val_loss: 0.4006\n",
            "Epoch 57/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0906 - loss: 0.5181 - val_accuracy: 0.1492 - val_loss: 0.4374\n",
            "Epoch 58/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1019 - loss: 0.4647 - val_accuracy: 0.1499 - val_loss: 0.4200\n",
            "Epoch 59/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0958 - loss: 0.4525 - val_accuracy: 0.1507 - val_loss: 0.3942\n",
            "Epoch 60/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0916 - loss: 0.4906 - val_accuracy: 0.1472 - val_loss: 0.4312\n",
            "Epoch 61/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1028 - loss: 0.4456 - val_accuracy: 0.1516 - val_loss: 0.3892\n",
            "Epoch 62/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0972 - loss: 0.4804 - val_accuracy: 0.1496 - val_loss: 0.4134\n",
            "Epoch 63/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0946 - loss: 0.4633 - val_accuracy: 0.1485 - val_loss: 0.4205\n",
            "Epoch 64/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0980 - loss: 0.4367 - val_accuracy: 0.1527 - val_loss: 0.3786\n",
            "Epoch 65/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0933 - loss: 0.4702 - val_accuracy: 0.1507 - val_loss: 0.4049\n",
            "Epoch 66/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1041 - loss: 0.4332 - val_accuracy: 0.1488 - val_loss: 0.4213\n",
            "Epoch 67/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1046 - loss: 0.4305 - val_accuracy: 0.1499 - val_loss: 0.4022\n",
            "Epoch 68/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0944 - loss: 0.4467 - val_accuracy: 0.1499 - val_loss: 0.4084\n",
            "Epoch 69/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0978 - loss: 0.4403 - val_accuracy: 0.1500 - val_loss: 0.4091\n",
            "Epoch 70/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0980 - loss: 0.4021 - val_accuracy: 0.1509 - val_loss: 0.3998\n",
            "Epoch 71/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1024 - loss: 0.4327 - val_accuracy: 0.1491 - val_loss: 0.4185\n",
            "Epoch 72/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0964 - loss: 0.4382 - val_accuracy: 0.1520 - val_loss: 0.3755\n",
            "Epoch 73/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1070 - loss: 0.4318 - val_accuracy: 0.1513 - val_loss: 0.3878\n",
            "Epoch 74/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0983 - loss: 0.4135 - val_accuracy: 0.1506 - val_loss: 0.4010\n",
            "Epoch 75/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0981 - loss: 0.4439 - val_accuracy: 0.1523 - val_loss: 0.3809\n",
            "Epoch 76/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0994 - loss: 0.4225 - val_accuracy: 0.1508 - val_loss: 0.3835\n",
            "Epoch 77/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0995 - loss: 0.4231 - val_accuracy: 0.1507 - val_loss: 0.3827\n",
            "Epoch 78/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0992 - loss: 0.4077 - val_accuracy: 0.1524 - val_loss: 0.3698\n",
            "Epoch 79/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1067 - loss: 0.4048 - val_accuracy: 0.1476 - val_loss: 0.4225\n",
            "Epoch 80/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0976 - loss: 0.4253 - val_accuracy: 0.1513 - val_loss: 0.3839\n",
            "Epoch 81/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0943 - loss: 0.4129 - val_accuracy: 0.1524 - val_loss: 0.3767\n",
            "Epoch 82/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1005 - loss: 0.4012 - val_accuracy: 0.1492 - val_loss: 0.4155\n",
            "Epoch 83/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1059 - loss: 0.4042 - val_accuracy: 0.1532 - val_loss: 0.3692\n",
            "Epoch 84/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1036 - loss: 0.4002 - val_accuracy: 0.1481 - val_loss: 0.4132\n",
            "Epoch 85/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0919 - loss: 0.4519 - val_accuracy: 0.1486 - val_loss: 0.4046\n",
            "Epoch 86/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0925 - loss: 0.4399 - val_accuracy: 0.1525 - val_loss: 0.3634\n",
            "Epoch 87/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1020 - loss: 0.4085 - val_accuracy: 0.1488 - val_loss: 0.3964\n",
            "Epoch 88/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1000 - loss: 0.3900 - val_accuracy: 0.1479 - val_loss: 0.4106\n",
            "Epoch 89/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0981 - loss: 0.4107 - val_accuracy: 0.1508 - val_loss: 0.3938\n",
            "Epoch 90/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1011 - loss: 0.4595 - val_accuracy: 0.1444 - val_loss: 0.4553\n",
            "Epoch 91/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0893 - loss: 0.4585 - val_accuracy: 0.1485 - val_loss: 0.4209\n",
            "Epoch 92/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1088 - loss: 0.4497 - val_accuracy: 0.1523 - val_loss: 0.3888\n",
            "Epoch 93/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1049 - loss: 0.4482 - val_accuracy: 0.1440 - val_loss: 0.4645\n",
            "Epoch 94/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0954 - loss: 0.4651 - val_accuracy: 0.1422 - val_loss: 0.4992\n",
            "Epoch 95/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0952 - loss: 0.4343 - val_accuracy: 0.1516 - val_loss: 0.3978\n",
            "Epoch 96/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0984 - loss: 0.4487 - val_accuracy: 0.1514 - val_loss: 0.4028\n",
            "Epoch 97/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1002 - loss: 0.4320 - val_accuracy: 0.1483 - val_loss: 0.4351\n",
            "Epoch 98/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1031 - loss: 0.4303 - val_accuracy: 0.1525 - val_loss: 0.3859\n",
            "Epoch 99/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0977 - loss: 0.4539 - val_accuracy: 0.1494 - val_loss: 0.4160\n",
            "Epoch 100/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0954 - loss: 0.4558 - val_accuracy: 0.1473 - val_loss: 0.4366\n",
            "Epoch 101/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0963 - loss: 0.4264 - val_accuracy: 0.1510 - val_loss: 0.4037\n",
            "Epoch 102/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0998 - loss: 0.4172 - val_accuracy: 0.1477 - val_loss: 0.4293\n",
            "Epoch 103/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0959 - loss: 0.4198 - val_accuracy: 0.1477 - val_loss: 0.4332\n",
            "Epoch 104/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1007 - loss: 0.4154 - val_accuracy: 0.1514 - val_loss: 0.3983\n",
            "Epoch 105/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1011 - loss: 0.3949 - val_accuracy: 0.1498 - val_loss: 0.4127\n",
            "Epoch 106/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0977 - loss: 0.4264 - val_accuracy: 0.1450 - val_loss: 0.4779\n",
            "Epoch 107/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0974 - loss: 0.4139 - val_accuracy: 0.1522 - val_loss: 0.3966\n",
            "Epoch 108/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1015 - loss: 0.4188 - val_accuracy: 0.1523 - val_loss: 0.3867\n",
            "Epoch 109/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0998 - loss: 0.4072 - val_accuracy: 0.1486 - val_loss: 0.4122\n",
            "Epoch 110/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1059 - loss: 0.4047 - val_accuracy: 0.1506 - val_loss: 0.3908\n",
            "Epoch 111/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1028 - loss: 0.4193 - val_accuracy: 0.1522 - val_loss: 0.3789\n",
            "Epoch 112/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0905 - loss: 0.4636 - val_accuracy: 0.1526 - val_loss: 0.3850\n",
            "Epoch 113/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0981 - loss: 0.4200 - val_accuracy: 0.1528 - val_loss: 0.3810\n",
            "Epoch 114/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1012 - loss: 0.4041 - val_accuracy: 0.1529 - val_loss: 0.3761\n",
            "Epoch 115/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1049 - loss: 0.3932 - val_accuracy: 0.1521 - val_loss: 0.3812\n",
            "Epoch 116/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0960 - loss: 0.4175 - val_accuracy: 0.1499 - val_loss: 0.4194\n",
            "Epoch 117/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0993 - loss: 0.3810 - val_accuracy: 0.1517 - val_loss: 0.3985\n",
            "Epoch 118/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1057 - loss: 0.3678 - val_accuracy: 0.1496 - val_loss: 0.4142\n",
            "Epoch 119/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1020 - loss: 0.3743 - val_accuracy: 0.1520 - val_loss: 0.3765\n",
            "Epoch 120/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0949 - loss: 0.3832 - val_accuracy: 0.1463 - val_loss: 0.4359\n",
            "Epoch 121/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0956 - loss: 0.3799 - val_accuracy: 0.1467 - val_loss: 0.4300\n",
            "Epoch 122/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0994 - loss: 0.3620 - val_accuracy: 0.1484 - val_loss: 0.4185\n",
            "Epoch 123/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1078 - loss: 0.3582 - val_accuracy: 0.1394 - val_loss: 0.4896\n",
            "Epoch 124/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0901 - loss: 0.4252 - val_accuracy: 0.1442 - val_loss: 0.4592\n",
            "Epoch 125/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1003 - loss: 0.3764 - val_accuracy: 0.1486 - val_loss: 0.4058\n",
            "Epoch 126/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0936 - loss: 0.4056 - val_accuracy: 0.1492 - val_loss: 0.3966\n",
            "Epoch 127/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0973 - loss: 0.4131 - val_accuracy: 0.1493 - val_loss: 0.4117\n",
            "Epoch 128/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0985 - loss: 0.4022 - val_accuracy: 0.1486 - val_loss: 0.4037\n",
            "Epoch 129/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1016 - loss: 0.4153 - val_accuracy: 0.1483 - val_loss: 0.3995\n",
            "Epoch 130/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1006 - loss: 0.4002 - val_accuracy: 0.1417 - val_loss: 0.4798\n",
            "Epoch 131/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0945 - loss: 0.4541 - val_accuracy: 0.1505 - val_loss: 0.3977\n",
            "Epoch 132/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0885 - loss: 0.5165 - val_accuracy: 0.1505 - val_loss: 0.3919\n",
            "Epoch 133/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0986 - loss: 0.4308 - val_accuracy: 0.1481 - val_loss: 0.4012\n",
            "Epoch 134/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0980 - loss: 0.4220 - val_accuracy: 0.1507 - val_loss: 0.3849\n",
            "Epoch 135/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0937 - loss: 0.4415 - val_accuracy: 0.1521 - val_loss: 0.3703\n",
            "Epoch 136/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0955 - loss: 0.4349 - val_accuracy: 0.1527 - val_loss: 0.3724\n",
            "Epoch 137/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0973 - loss: 0.4173 - val_accuracy: 0.1500 - val_loss: 0.4070\n",
            "Epoch 138/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0995 - loss: 0.4204 - val_accuracy: 0.1483 - val_loss: 0.4065\n",
            "Epoch 139/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0957 - loss: 0.4032 - val_accuracy: 0.1479 - val_loss: 0.3928\n",
            "Epoch 140/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1004 - loss: 0.3901 - val_accuracy: 0.1418 - val_loss: 0.4595\n",
            "Epoch 141/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0977 - loss: 0.3806 - val_accuracy: 0.1418 - val_loss: 0.4477\n",
            "Epoch 142/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0981 - loss: 0.4039 - val_accuracy: 0.1454 - val_loss: 0.4170\n",
            "Epoch 143/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0966 - loss: 0.4103 - val_accuracy: 0.1333 - val_loss: 0.5248\n",
            "Epoch 144/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0994 - loss: 0.4099 - val_accuracy: 0.1413 - val_loss: 0.4896\n",
            "Epoch 145/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0962 - loss: 0.4049 - val_accuracy: 0.1521 - val_loss: 0.4082\n",
            "Epoch 146/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1069 - loss: 0.4169 - val_accuracy: 0.1471 - val_loss: 0.4264\n",
            "Epoch 147/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0970 - loss: 0.4109 - val_accuracy: 0.1484 - val_loss: 0.4211\n",
            "Epoch 148/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1042 - loss: 0.3740 - val_accuracy: 0.1470 - val_loss: 0.4385\n",
            "Epoch 149/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0957 - loss: 0.4051 - val_accuracy: 0.1463 - val_loss: 0.4232\n",
            "Epoch 150/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0950 - loss: 0.3858 - val_accuracy: 0.1487 - val_loss: 0.4049\n",
            "Epoch 151/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1000 - loss: 0.3516 - val_accuracy: 0.1515 - val_loss: 0.3953\n",
            "Epoch 152/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1049 - loss: 0.3822 - val_accuracy: 0.1487 - val_loss: 0.4327\n",
            "Epoch 153/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1033 - loss: 0.3825 - val_accuracy: 0.1489 - val_loss: 0.4286\n",
            "Epoch 154/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1001 - loss: 0.3830 - val_accuracy: 0.1502 - val_loss: 0.3966\n",
            "Epoch 155/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0888 - loss: 0.4308 - val_accuracy: 0.1490 - val_loss: 0.4254\n",
            "Epoch 156/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0948 - loss: 0.4033 - val_accuracy: 0.1482 - val_loss: 0.4347\n",
            "Epoch 157/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0935 - loss: 0.3787 - val_accuracy: 0.1520 - val_loss: 0.3933\n",
            "Epoch 158/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0952 - loss: 0.4044 - val_accuracy: 0.1488 - val_loss: 0.4115\n",
            "Epoch 159/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0965 - loss: 0.3861 - val_accuracy: 0.1426 - val_loss: 0.4847\n",
            "Epoch 160/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0994 - loss: 0.4156 - val_accuracy: 0.1533 - val_loss: 0.4065\n",
            "Epoch 161/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0982 - loss: 0.5131 - val_accuracy: 0.1394 - val_loss: 0.5257\n",
            "Epoch 162/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0937 - loss: 0.5369 - val_accuracy: 0.1417 - val_loss: 0.4469\n",
            "Epoch 163/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0933 - loss: 0.4744 - val_accuracy: 0.1238 - val_loss: 0.5290\n",
            "Epoch 164/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0898 - loss: 0.4932 - val_accuracy: 0.1339 - val_loss: 0.5135\n",
            "Epoch 165/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1036 - loss: 0.4324 - val_accuracy: 0.1492 - val_loss: 0.4139\n",
            "Epoch 166/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1038 - loss: 0.4293 - val_accuracy: 0.1507 - val_loss: 0.3953\n",
            "Epoch 167/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0983 - loss: 0.4271 - val_accuracy: 0.1506 - val_loss: 0.4039\n",
            "Epoch 168/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0926 - loss: 0.4112 - val_accuracy: 0.1507 - val_loss: 0.3973\n",
            "Epoch 169/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1009 - loss: 0.3852 - val_accuracy: 0.1509 - val_loss: 0.3937\n",
            "Epoch 170/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0959 - loss: 0.4093 - val_accuracy: 0.1494 - val_loss: 0.4255\n",
            "Epoch 171/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0992 - loss: 0.4067 - val_accuracy: 0.1526 - val_loss: 0.3890\n",
            "Epoch 172/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0966 - loss: 0.3815 - val_accuracy: 0.1517 - val_loss: 0.4001\n",
            "Epoch 173/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0928 - loss: 0.4060 - val_accuracy: 0.1496 - val_loss: 0.4219\n",
            "Epoch 174/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0982 - loss: 0.3950 - val_accuracy: 0.1515 - val_loss: 0.3879\n",
            "Epoch 175/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0971 - loss: 0.3942 - val_accuracy: 0.1484 - val_loss: 0.4066\n",
            "Epoch 176/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0988 - loss: 0.3653 - val_accuracy: 0.1485 - val_loss: 0.4010\n",
            "Epoch 177/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0918 - loss: 0.3901 - val_accuracy: 0.1520 - val_loss: 0.3739\n",
            "Epoch 178/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0913 - loss: 0.4219 - val_accuracy: 0.1497 - val_loss: 0.3953\n",
            "Epoch 179/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1001 - loss: 0.3679 - val_accuracy: 0.1468 - val_loss: 0.4019\n",
            "Epoch 180/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1020 - loss: 0.3524 - val_accuracy: 0.1495 - val_loss: 0.3962\n",
            "Epoch 181/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1065 - loss: 0.3435 - val_accuracy: 0.1483 - val_loss: 0.4116\n",
            "Epoch 182/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.1009 - loss: 0.3347 - val_accuracy: 0.1443 - val_loss: 0.4487\n",
            "Epoch 183/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1033 - loss: 0.3394 - val_accuracy: 0.1428 - val_loss: 0.4571\n",
            "Epoch 184/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1061 - loss: 0.3302 - val_accuracy: 0.1421 - val_loss: 0.4301\n",
            "Epoch 185/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0921 - loss: 0.3744 - val_accuracy: 0.1453 - val_loss: 0.4206\n",
            "Epoch 186/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1028 - loss: 0.3193 - val_accuracy: 0.1465 - val_loss: 0.4223\n",
            "Epoch 187/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1001 - loss: 0.3177 - val_accuracy: 0.1425 - val_loss: 0.4130\n",
            "Epoch 188/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1080 - loss: 0.3283 - val_accuracy: 0.1415 - val_loss: 0.4340\n",
            "Epoch 189/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1022 - loss: 0.3270 - val_accuracy: 0.1449 - val_loss: 0.4158\n",
            "Epoch 190/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1042 - loss: 0.3440 - val_accuracy: 0.1377 - val_loss: 0.4452\n",
            "Epoch 191/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1006 - loss: 0.3555 - val_accuracy: 0.1449 - val_loss: 0.4217\n",
            "Epoch 192/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1020 - loss: 0.3418 - val_accuracy: 0.1476 - val_loss: 0.4016\n",
            "Epoch 193/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0997 - loss: 0.3445 - val_accuracy: 0.1489 - val_loss: 0.3816\n",
            "Epoch 194/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1029 - loss: 0.3452 - val_accuracy: 0.1368 - val_loss: 0.4650\n",
            "Epoch 195/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0981 - loss: 0.3924 - val_accuracy: 0.1353 - val_loss: 0.4820\n",
            "Epoch 196/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1014 - loss: 0.3818 - val_accuracy: 0.1386 - val_loss: 0.4478\n",
            "Epoch 197/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1003 - loss: 0.3756 - val_accuracy: 0.1472 - val_loss: 0.4395\n",
            "Epoch 198/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0944 - loss: 0.3831 - val_accuracy: 0.1452 - val_loss: 0.4474\n",
            "Epoch 199/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1008 - loss: 0.3752 - val_accuracy: 0.1459 - val_loss: 0.4084\n",
            "Epoch 200/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0960 - loss: 0.4884 - val_accuracy: 0.1390 - val_loss: 0.5419\n",
            "Epoch 201/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1047 - loss: 0.4022 - val_accuracy: 0.1449 - val_loss: 0.4928\n",
            "Epoch 202/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1030 - loss: 0.3912 - val_accuracy: 0.1494 - val_loss: 0.4068\n",
            "Epoch 203/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1051 - loss: 0.3510 - val_accuracy: 0.1471 - val_loss: 0.4001\n",
            "Epoch 204/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1035 - loss: 0.3651 - val_accuracy: 0.1345 - val_loss: 0.4938\n",
            "Epoch 205/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1016 - loss: 0.3831 - val_accuracy: 0.1463 - val_loss: 0.4251\n",
            "Epoch 206/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0984 - loss: 0.3747 - val_accuracy: 0.1535 - val_loss: 0.3573\n",
            "Epoch 207/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0982 - loss: 0.3916 - val_accuracy: 0.1523 - val_loss: 0.3671\n",
            "Epoch 208/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1076 - loss: 0.3644 - val_accuracy: 0.1426 - val_loss: 0.4425\n",
            "Epoch 209/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1007 - loss: 0.3382 - val_accuracy: 0.1353 - val_loss: 0.4823\n",
            "Epoch 210/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0948 - loss: 0.3593 - val_accuracy: 0.1461 - val_loss: 0.4124\n",
            "Epoch 211/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1029 - loss: 0.4012 - val_accuracy: 0.1434 - val_loss: 0.4592\n",
            "Epoch 212/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0969 - loss: 0.4218 - val_accuracy: 0.1462 - val_loss: 0.4147\n",
            "Epoch 213/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0935 - loss: 0.4478 - val_accuracy: 0.1482 - val_loss: 0.3940\n",
            "Epoch 214/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0962 - loss: 0.4315 - val_accuracy: 0.1420 - val_loss: 0.4719\n",
            "Epoch 215/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0969 - loss: 0.3959 - val_accuracy: 0.1463 - val_loss: 0.4567\n",
            "Epoch 216/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1004 - loss: 0.4014 - val_accuracy: 0.1506 - val_loss: 0.4311\n",
            "Epoch 217/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1015 - loss: 0.3983 - val_accuracy: 0.1492 - val_loss: 0.4483\n",
            "Epoch 218/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0951 - loss: 0.4117 - val_accuracy: 0.1487 - val_loss: 0.4574\n",
            "Epoch 219/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0964 - loss: 0.3555 - val_accuracy: 0.1455 - val_loss: 0.4661\n",
            "Epoch 220/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0971 - loss: 0.4206 - val_accuracy: 0.1402 - val_loss: 0.4976\n",
            "Epoch 221/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1030 - loss: 0.4108 - val_accuracy: 0.1409 - val_loss: 0.5095\n",
            "Epoch 222/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1021 - loss: 0.4182 - val_accuracy: 0.1476 - val_loss: 0.4499\n",
            "Epoch 223/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1033 - loss: 0.3808 - val_accuracy: 0.1513 - val_loss: 0.3827\n",
            "Epoch 224/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0957 - loss: 0.3782 - val_accuracy: 0.1501 - val_loss: 0.3833\n",
            "Epoch 225/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1041 - loss: 0.3623 - val_accuracy: 0.1484 - val_loss: 0.3960\n",
            "Epoch 226/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0940 - loss: 0.3942 - val_accuracy: 0.1507 - val_loss: 0.3752\n",
            "Epoch 227/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1045 - loss: 0.3917 - val_accuracy: 0.1471 - val_loss: 0.4486\n",
            "Epoch 228/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0957 - loss: 0.4785 - val_accuracy: 0.1394 - val_loss: 0.5610\n",
            "Epoch 229/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0977 - loss: 0.5404 - val_accuracy: 0.1497 - val_loss: 0.4322\n",
            "Epoch 230/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0977 - loss: 0.5099 - val_accuracy: 0.1514 - val_loss: 0.4077\n",
            "Epoch 231/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0958 - loss: 0.4727 - val_accuracy: 0.1513 - val_loss: 0.4004\n",
            "Epoch 232/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0934 - loss: 0.4878 - val_accuracy: 0.1500 - val_loss: 0.4013\n",
            "Epoch 233/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1021 - loss: 0.4635 - val_accuracy: 0.1503 - val_loss: 0.3935\n",
            "Epoch 234/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1018 - loss: 0.4269 - val_accuracy: 0.1511 - val_loss: 0.3806\n",
            "Epoch 235/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0993 - loss: 0.4203 - val_accuracy: 0.1490 - val_loss: 0.3948\n",
            "Epoch 236/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0996 - loss: 0.4312 - val_accuracy: 0.1491 - val_loss: 0.3929\n",
            "Epoch 237/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0968 - loss: 0.4044 - val_accuracy: 0.1479 - val_loss: 0.4117\n",
            "Epoch 238/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0982 - loss: 0.3799 - val_accuracy: 0.1480 - val_loss: 0.4131\n",
            "Epoch 239/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0994 - loss: 0.3894 - val_accuracy: 0.1476 - val_loss: 0.4270\n",
            "Epoch 240/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0955 - loss: 0.4094 - val_accuracy: 0.1453 - val_loss: 0.4662\n",
            "Epoch 241/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1021 - loss: 0.3779 - val_accuracy: 0.1475 - val_loss: 0.4431\n",
            "Epoch 242/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1066 - loss: 0.3533 - val_accuracy: 0.1508 - val_loss: 0.3907\n",
            "Epoch 243/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1058 - loss: 0.3496 - val_accuracy: 0.1522 - val_loss: 0.3681\n",
            "Epoch 244/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1037 - loss: 0.3427 - val_accuracy: 0.1481 - val_loss: 0.4240\n",
            "Epoch 245/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1065 - loss: 0.3337 - val_accuracy: 0.1475 - val_loss: 0.4266\n",
            "Epoch 246/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0981 - loss: 0.3452 - val_accuracy: 0.1513 - val_loss: 0.3734\n",
            "Epoch 247/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0972 - loss: 0.3507 - val_accuracy: 0.1509 - val_loss: 0.3679\n",
            "Epoch 248/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1030 - loss: 0.3309 - val_accuracy: 0.1443 - val_loss: 0.4085\n",
            "Epoch 249/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1037 - loss: 0.3150 - val_accuracy: 0.1429 - val_loss: 0.4196\n",
            "Epoch 250/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1033 - loss: 0.3262 - val_accuracy: 0.1441 - val_loss: 0.3936\n",
            "Epoch 251/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1013 - loss: 0.3142 - val_accuracy: 0.1536 - val_loss: 0.3260\n",
            "Epoch 252/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1048 - loss: 0.3081 - val_accuracy: 0.1511 - val_loss: 0.3384\n",
            "Epoch 253/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1038 - loss: 0.3092 - val_accuracy: 0.1427 - val_loss: 0.3996\n",
            "Epoch 254/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1022 - loss: 0.2950 - val_accuracy: 0.1494 - val_loss: 0.3455\n",
            "Epoch 255/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1054 - loss: 0.2785 - val_accuracy: 0.1506 - val_loss: 0.3456\n",
            "Epoch 256/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1027 - loss: 0.2804 - val_accuracy: 0.1388 - val_loss: 0.4330\n",
            "Epoch 257/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1044 - loss: 0.2912 - val_accuracy: 0.1460 - val_loss: 0.4566\n",
            "Epoch 258/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0930 - loss: 0.4694 - val_accuracy: 0.1481 - val_loss: 0.4060\n",
            "Epoch 259/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0946 - loss: 0.4170 - val_accuracy: 0.1184 - val_loss: 0.5606\n",
            "Epoch 260/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0827 - loss: 0.5215 - val_accuracy: 0.0962 - val_loss: 0.6893\n",
            "Epoch 261/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0818 - loss: 0.5669 - val_accuracy: 0.1205 - val_loss: 0.5339\n",
            "Epoch 262/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0967 - loss: 0.5152 - val_accuracy: 0.1495 - val_loss: 0.3835\n",
            "Epoch 263/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0944 - loss: 0.4465 - val_accuracy: 0.1468 - val_loss: 0.3908\n",
            "Epoch 264/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1028 - loss: 0.4090 - val_accuracy: 0.1485 - val_loss: 0.4081\n",
            "Epoch 265/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1032 - loss: 0.4131 - val_accuracy: 0.1485 - val_loss: 0.3874\n",
            "Epoch 266/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0964 - loss: 0.4129 - val_accuracy: 0.1470 - val_loss: 0.4193\n",
            "Epoch 267/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0976 - loss: 0.3965 - val_accuracy: 0.1506 - val_loss: 0.3691\n",
            "Epoch 268/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1001 - loss: 0.3354 - val_accuracy: 0.1504 - val_loss: 0.3675\n",
            "Epoch 269/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1013 - loss: 0.3611 - val_accuracy: 0.1486 - val_loss: 0.4060\n",
            "Epoch 270/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1066 - loss: 0.3421 - val_accuracy: 0.1505 - val_loss: 0.3717\n",
            "Epoch 271/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0907 - loss: 0.4132 - val_accuracy: 0.1517 - val_loss: 0.3967\n",
            "Epoch 272/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0924 - loss: 0.3809 - val_accuracy: 0.1498 - val_loss: 0.3810\n",
            "Epoch 273/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0999 - loss: 0.3307 - val_accuracy: 0.1506 - val_loss: 0.3791\n",
            "Epoch 274/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1029 - loss: 0.3189 - val_accuracy: 0.1517 - val_loss: 0.3728\n",
            "Epoch 275/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0978 - loss: 0.3537 - val_accuracy: 0.1488 - val_loss: 0.3685\n",
            "Epoch 276/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0893 - loss: 0.3668 - val_accuracy: 0.1518 - val_loss: 0.3623\n",
            "Epoch 277/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1081 - loss: 0.3257 - val_accuracy: 0.1460 - val_loss: 0.4218\n",
            "Epoch 278/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0947 - loss: 0.3455 - val_accuracy: 0.1452 - val_loss: 0.4307\n",
            "Epoch 279/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1011 - loss: 0.3381 - val_accuracy: 0.1300 - val_loss: 0.4585\n",
            "Epoch 280/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.1065 - loss: 0.3293 - val_accuracy: 0.1321 - val_loss: 0.4675\n",
            "Epoch 281/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1043 - loss: 0.3087 - val_accuracy: 0.1452 - val_loss: 0.4142\n",
            "Epoch 282/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1019 - loss: 0.3149 - val_accuracy: 0.1325 - val_loss: 0.4359\n",
            "Epoch 283/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1010 - loss: 0.3271 - val_accuracy: 0.1110 - val_loss: 0.5347\n",
            "Epoch 284/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0916 - loss: 0.4146 - val_accuracy: 0.1350 - val_loss: 0.4562\n",
            "Epoch 285/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0944 - loss: 0.3462 - val_accuracy: 0.1459 - val_loss: 0.4687\n",
            "Epoch 286/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0909 - loss: 0.4745 - val_accuracy: 0.1363 - val_loss: 0.4708\n",
            "Epoch 287/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1028 - loss: 0.3507 - val_accuracy: 0.1293 - val_loss: 0.5085\n",
            "Epoch 288/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0945 - loss: 0.4195 - val_accuracy: 0.0582 - val_loss: 0.8119\n",
            "Epoch 289/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0763 - loss: 0.7395 - val_accuracy: 0.0836 - val_loss: 0.7413\n",
            "Epoch 290/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0785 - loss: 0.8104 - val_accuracy: 0.1460 - val_loss: 0.5849\n",
            "Epoch 291/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0983 - loss: 0.6240 - val_accuracy: 0.1493 - val_loss: 0.4928\n",
            "Epoch 292/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0962 - loss: 0.5678 - val_accuracy: 0.1518 - val_loss: 0.4206\n",
            "Epoch 293/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0928 - loss: 0.4950 - val_accuracy: 0.1379 - val_loss: 0.5327\n",
            "Epoch 294/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0968 - loss: 0.5277 - val_accuracy: 0.1318 - val_loss: 0.5686\n",
            "Epoch 295/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0976 - loss: 0.5240 - val_accuracy: 0.1488 - val_loss: 0.4271\n",
            "Epoch 296/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0922 - loss: 0.5231 - val_accuracy: 0.1530 - val_loss: 0.4008\n",
            "Epoch 297/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0992 - loss: 0.5313 - val_accuracy: 0.1499 - val_loss: 0.4386\n",
            "Epoch 298/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0949 - loss: 0.5111 - val_accuracy: 0.1470 - val_loss: 0.4760\n",
            "Epoch 299/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0963 - loss: 0.5045 - val_accuracy: 0.1491 - val_loss: 0.4601\n",
            "Epoch 300/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1004 - loss: 0.4950 - val_accuracy: 0.1527 - val_loss: 0.4172\n",
            "Epoch 301/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0997 - loss: 0.4900 - val_accuracy: 0.1530 - val_loss: 0.4063\n",
            "Epoch 302/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0982 - loss: 0.4926 - val_accuracy: 0.1507 - val_loss: 0.4225\n",
            "Epoch 303/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0885 - loss: 0.5228 - val_accuracy: 0.1513 - val_loss: 0.4072\n",
            "Epoch 304/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0978 - loss: 0.4457 - val_accuracy: 0.1543 - val_loss: 0.3749\n",
            "Epoch 305/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1000 - loss: 0.4729 - val_accuracy: 0.1539 - val_loss: 0.3782\n",
            "Epoch 306/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0966 - loss: 0.4701 - val_accuracy: 0.1524 - val_loss: 0.4030\n",
            "Epoch 307/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0967 - loss: 0.4156 - val_accuracy: 0.1497 - val_loss: 0.4290\n",
            "Epoch 308/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0935 - loss: 0.4629 - val_accuracy: 0.1511 - val_loss: 0.4260\n",
            "Epoch 309/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1061 - loss: 0.4181 - val_accuracy: 0.1524 - val_loss: 0.4149\n",
            "Epoch 310/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1054 - loss: 0.4177 - val_accuracy: 0.1516 - val_loss: 0.4269\n",
            "Epoch 311/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1005 - loss: 0.4057 - val_accuracy: 0.1496 - val_loss: 0.4425\n",
            "Epoch 312/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0974 - loss: 0.4203 - val_accuracy: 0.1515 - val_loss: 0.4179\n",
            "Epoch 313/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1017 - loss: 0.4065 - val_accuracy: 0.1523 - val_loss: 0.4037\n",
            "Epoch 314/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1049 - loss: 0.3982 - val_accuracy: 0.1494 - val_loss: 0.4446\n",
            "Epoch 315/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1085 - loss: 0.4028 - val_accuracy: 0.1496 - val_loss: 0.4417\n",
            "Epoch 316/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1023 - loss: 0.4164 - val_accuracy: 0.1494 - val_loss: 0.4395\n",
            "Epoch 317/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0891 - loss: 0.4627 - val_accuracy: 0.1512 - val_loss: 0.4169\n",
            "Epoch 318/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1040 - loss: 0.3909 - val_accuracy: 0.1512 - val_loss: 0.4162\n",
            "Epoch 319/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1001 - loss: 0.3962 - val_accuracy: 0.1482 - val_loss: 0.4478\n",
            "Epoch 320/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1027 - loss: 0.4030 - val_accuracy: 0.1482 - val_loss: 0.4432\n",
            "Epoch 321/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0939 - loss: 0.4317 - val_accuracy: 0.1485 - val_loss: 0.4328\n",
            "Epoch 322/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1028 - loss: 0.3960 - val_accuracy: 0.1493 - val_loss: 0.4286\n",
            "Epoch 323/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1044 - loss: 0.3869 - val_accuracy: 0.1495 - val_loss: 0.4301\n",
            "Epoch 324/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1068 - loss: 0.3962 - val_accuracy: 0.1499 - val_loss: 0.4244\n",
            "Epoch 325/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0882 - loss: 0.4476 - val_accuracy: 0.1505 - val_loss: 0.4206\n",
            "Epoch 326/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1009 - loss: 0.4019 - val_accuracy: 0.1515 - val_loss: 0.4115\n",
            "Epoch 327/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1065 - loss: 0.3877 - val_accuracy: 0.1520 - val_loss: 0.4052\n",
            "Epoch 328/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1068 - loss: 0.3816 - val_accuracy: 0.1508 - val_loss: 0.4219\n",
            "Epoch 329/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0938 - loss: 0.4061 - val_accuracy: 0.1506 - val_loss: 0.4283\n",
            "Epoch 330/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1014 - loss: 0.3807 - val_accuracy: 0.1519 - val_loss: 0.4131\n",
            "Epoch 331/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1024 - loss: 0.3786 - val_accuracy: 0.1518 - val_loss: 0.4158\n",
            "Epoch 332/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1095 - loss: 0.3767 - val_accuracy: 0.1508 - val_loss: 0.4267\n",
            "Epoch 333/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0984 - loss: 0.3908 - val_accuracy: 0.1510 - val_loss: 0.4227\n",
            "Epoch 334/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1054 - loss: 0.3577 - val_accuracy: 0.1516 - val_loss: 0.4137\n",
            "Epoch 335/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0939 - loss: 0.3844 - val_accuracy: 0.1515 - val_loss: 0.4148\n",
            "Epoch 336/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0998 - loss: 0.3568 - val_accuracy: 0.1513 - val_loss: 0.4167\n",
            "Epoch 337/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1036 - loss: 0.3745 - val_accuracy: 0.1503 - val_loss: 0.4278\n",
            "Epoch 338/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1037 - loss: 0.3631 - val_accuracy: 0.1502 - val_loss: 0.4287\n",
            "Epoch 339/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0962 - loss: 0.3943 - val_accuracy: 0.1533 - val_loss: 0.3878\n",
            "Epoch 340/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0999 - loss: 0.3915 - val_accuracy: 0.1520 - val_loss: 0.4010\n",
            "Epoch 341/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0966 - loss: 0.3821 - val_accuracy: 0.1512 - val_loss: 0.4130\n",
            "Epoch 342/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0992 - loss: 0.3869 - val_accuracy: 0.1509 - val_loss: 0.4229\n",
            "Epoch 343/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0936 - loss: 0.4099 - val_accuracy: 0.1525 - val_loss: 0.4029\n",
            "Epoch 344/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1018 - loss: 0.3595 - val_accuracy: 0.1535 - val_loss: 0.3977\n",
            "Epoch 345/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1096 - loss: 0.3671 - val_accuracy: 0.1515 - val_loss: 0.4236\n",
            "Epoch 346/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0996 - loss: 0.3707 - val_accuracy: 0.1523 - val_loss: 0.4168\n",
            "Epoch 347/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1033 - loss: 0.3539 - val_accuracy: 0.1538 - val_loss: 0.4014\n",
            "Epoch 348/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0975 - loss: 0.3742 - val_accuracy: 0.1535 - val_loss: 0.4037\n",
            "Epoch 349/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1052 - loss: 0.3433 - val_accuracy: 0.1525 - val_loss: 0.4133\n",
            "Epoch 350/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0980 - loss: 0.3613 - val_accuracy: 0.1531 - val_loss: 0.4074\n",
            "Epoch 351/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0913 - loss: 0.3939 - val_accuracy: 0.1543 - val_loss: 0.3995\n",
            "Epoch 352/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1042 - loss: 0.3552 - val_accuracy: 0.1540 - val_loss: 0.4074\n",
            "Epoch 353/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0969 - loss: 0.3669 - val_accuracy: 0.1511 - val_loss: 0.4373\n",
            "Epoch 354/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1021 - loss: 0.3475 - val_accuracy: 0.1529 - val_loss: 0.4213\n",
            "Epoch 355/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1029 - loss: 0.3494 - val_accuracy: 0.1537 - val_loss: 0.4123\n",
            "Epoch 356/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0999 - loss: 0.3498 - val_accuracy: 0.1521 - val_loss: 0.4262\n",
            "Epoch 357/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0971 - loss: 0.3409 - val_accuracy: 0.1534 - val_loss: 0.4194\n",
            "Epoch 358/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1084 - loss: 0.3368 - val_accuracy: 0.1532 - val_loss: 0.4181\n",
            "Epoch 359/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0976 - loss: 0.3418 - val_accuracy: 0.1521 - val_loss: 0.4231\n",
            "Epoch 360/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0976 - loss: 0.3359 - val_accuracy: 0.1518 - val_loss: 0.4352\n",
            "Epoch 361/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0966 - loss: 0.3287 - val_accuracy: 0.1512 - val_loss: 0.4422\n",
            "Epoch 362/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1071 - loss: 0.3269 - val_accuracy: 0.1511 - val_loss: 0.4384\n",
            "Epoch 363/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1017 - loss: 0.3195 - val_accuracy: 0.1515 - val_loss: 0.4300\n",
            "Epoch 364/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0949 - loss: 0.3381 - val_accuracy: 0.1507 - val_loss: 0.4289\n",
            "Epoch 365/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1078 - loss: 0.3151 - val_accuracy: 0.1503 - val_loss: 0.4285\n",
            "Epoch 366/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.1053 - loss: 0.3047 - val_accuracy: 0.1469 - val_loss: 0.4405\n",
            "Epoch 367/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1090 - loss: 0.3181 - val_accuracy: 0.1476 - val_loss: 0.4394\n",
            "Epoch 368/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0995 - loss: 0.3206 - val_accuracy: 0.1474 - val_loss: 0.4538\n",
            "Epoch 369/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0939 - loss: 0.3314 - val_accuracy: 0.1460 - val_loss: 0.4630\n",
            "Epoch 370/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1026 - loss: 0.3178 - val_accuracy: 0.1490 - val_loss: 0.4484\n",
            "Epoch 371/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1005 - loss: 0.3209 - val_accuracy: 0.1465 - val_loss: 0.4469\n",
            "Epoch 372/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1088 - loss: 0.2928 - val_accuracy: 0.1460 - val_loss: 0.4481\n",
            "Epoch 373/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0916 - loss: 0.3654 - val_accuracy: 0.1458 - val_loss: 0.4544\n",
            "Epoch 374/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1020 - loss: 0.2977 - val_accuracy: 0.1491 - val_loss: 0.4556\n",
            "Epoch 375/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1028 - loss: 0.3060 - val_accuracy: 0.1408 - val_loss: 0.4605\n",
            "Epoch 376/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1035 - loss: 0.3204 - val_accuracy: 0.1376 - val_loss: 0.4798\n",
            "Epoch 377/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1031 - loss: 0.3129 - val_accuracy: 0.1460 - val_loss: 0.4610\n",
            "Epoch 378/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1047 - loss: 0.3147 - val_accuracy: 0.1387 - val_loss: 0.4794\n",
            "Epoch 379/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0951 - loss: 0.3570 - val_accuracy: 0.1350 - val_loss: 0.4949\n",
            "Epoch 380/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1029 - loss: 0.3403 - val_accuracy: 0.1393 - val_loss: 0.4638\n",
            "Epoch 381/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0966 - loss: 0.3346 - val_accuracy: 0.1399 - val_loss: 0.4508\n",
            "Epoch 382/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1035 - loss: 0.3430 - val_accuracy: 0.1460 - val_loss: 0.4202\n",
            "Epoch 383/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1054 - loss: 0.3319 - val_accuracy: 0.1405 - val_loss: 0.4404\n",
            "Epoch 384/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1007 - loss: 0.3304 - val_accuracy: 0.1402 - val_loss: 0.4392\n",
            "Epoch 385/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1024 - loss: 0.3021 - val_accuracy: 0.1393 - val_loss: 0.4422\n",
            "Epoch 386/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0984 - loss: 0.3373 - val_accuracy: 0.1450 - val_loss: 0.4187\n",
            "Epoch 387/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1012 - loss: 0.3388 - val_accuracy: 0.1438 - val_loss: 0.4268\n",
            "Epoch 388/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0969 - loss: 0.3408 - val_accuracy: 0.1320 - val_loss: 0.4949\n",
            "Epoch 389/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1017 - loss: 0.3436 - val_accuracy: 0.1436 - val_loss: 0.4376\n",
            "Epoch 390/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1012 - loss: 0.3413 - val_accuracy: 0.1457 - val_loss: 0.4304\n",
            "Epoch 391/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0964 - loss: 0.3612 - val_accuracy: 0.1432 - val_loss: 0.4380\n",
            "Epoch 392/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1085 - loss: 0.3326 - val_accuracy: 0.1434 - val_loss: 0.4312\n",
            "Epoch 393/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0986 - loss: 0.3281 - val_accuracy: 0.1370 - val_loss: 0.4552\n",
            "Epoch 394/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.1082 - loss: 0.3415 - val_accuracy: 0.1472 - val_loss: 0.4705\n",
            "Epoch 395/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0988 - loss: 0.4571 - val_accuracy: 0.1397 - val_loss: 0.4612\n",
            "Epoch 396/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0956 - loss: 0.4298 - val_accuracy: 0.1428 - val_loss: 0.5003\n",
            "Epoch 397/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0975 - loss: 0.4260 - val_accuracy: 0.1524 - val_loss: 0.4433\n",
            "Epoch 398/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1039 - loss: 0.5222 - val_accuracy: 0.1513 - val_loss: 0.4158\n",
            "Epoch 399/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0951 - loss: 0.4674 - val_accuracy: 0.1448 - val_loss: 0.4486\n",
            "Epoch 400/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0957 - loss: 0.4729 - val_accuracy: 0.1513 - val_loss: 0.4025\n",
            "Epoch 401/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0886 - loss: 0.5534 - val_accuracy: 0.1513 - val_loss: 0.4064\n",
            "Epoch 402/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.0993 - loss: 0.4671 - val_accuracy: 0.1496 - val_loss: 0.4468\n",
            "Epoch 403/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0954 - loss: 0.4918 - val_accuracy: 0.1528 - val_loss: 0.4356\n",
            "Epoch 404/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0988 - loss: 0.4595 - val_accuracy: 0.1532 - val_loss: 0.4116\n",
            "Epoch 405/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0879 - loss: 0.4978 - val_accuracy: 0.1534 - val_loss: 0.3910\n",
            "Epoch 406/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0946 - loss: 0.4707 - val_accuracy: 0.1527 - val_loss: 0.3757\n",
            "Epoch 407/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0962 - loss: 0.4195 - val_accuracy: 0.1528 - val_loss: 0.3621\n",
            "Epoch 408/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0930 - loss: 0.4519 - val_accuracy: 0.1543 - val_loss: 0.3725\n",
            "Epoch 409/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1039 - loss: 0.4222 - val_accuracy: 0.1542 - val_loss: 0.3841\n",
            "Epoch 410/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1019 - loss: 0.4128 - val_accuracy: 0.1522 - val_loss: 0.3907\n",
            "Epoch 411/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0990 - loss: 0.4095 - val_accuracy: 0.1536 - val_loss: 0.3558\n",
            "Epoch 412/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1036 - loss: 0.3762 - val_accuracy: 0.1531 - val_loss: 0.3651\n",
            "Epoch 413/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0994 - loss: 0.3606 - val_accuracy: 0.1528 - val_loss: 0.3718\n",
            "Epoch 414/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1031 - loss: 0.3582 - val_accuracy: 0.1530 - val_loss: 0.3604\n",
            "Epoch 415/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1009 - loss: 0.3658 - val_accuracy: 0.1520 - val_loss: 0.3751\n",
            "Epoch 416/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1054 - loss: 0.3646 - val_accuracy: 0.1536 - val_loss: 0.3669\n",
            "Epoch 417/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0961 - loss: 0.3870 - val_accuracy: 0.1516 - val_loss: 0.3853\n",
            "Epoch 418/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0998 - loss: 0.3674 - val_accuracy: 0.1500 - val_loss: 0.3904\n",
            "Epoch 419/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0971 - loss: 0.3808 - val_accuracy: 0.1522 - val_loss: 0.3693\n",
            "Epoch 420/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0986 - loss: 0.3596 - val_accuracy: 0.1516 - val_loss: 0.3743\n",
            "Epoch 421/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0960 - loss: 0.3817 - val_accuracy: 0.1495 - val_loss: 0.3877\n",
            "Epoch 422/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0864 - loss: 0.3994 - val_accuracy: 0.1506 - val_loss: 0.3738\n",
            "Epoch 423/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1034 - loss: 0.3190 - val_accuracy: 0.1521 - val_loss: 0.3573\n",
            "Epoch 424/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1062 - loss: 0.3294 - val_accuracy: 0.1518 - val_loss: 0.3526\n",
            "Epoch 425/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.0964 - loss: 0.3444 - val_accuracy: 0.1500 - val_loss: 0.3692\n",
            "Epoch 426/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0974 - loss: 0.3312 - val_accuracy: 0.1502 - val_loss: 0.3716\n",
            "Epoch 427/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0904 - loss: 0.3946 - val_accuracy: 0.1494 - val_loss: 0.3680\n",
            "Epoch 428/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0984 - loss: 0.3321 - val_accuracy: 0.1488 - val_loss: 0.3684\n",
            "Epoch 429/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1045 - loss: 0.2972 - val_accuracy: 0.1470 - val_loss: 0.3887\n",
            "Epoch 430/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0982 - loss: 0.3193 - val_accuracy: 0.1489 - val_loss: 0.3757\n",
            "Epoch 431/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0925 - loss: 0.3298 - val_accuracy: 0.1503 - val_loss: 0.3640\n",
            "Epoch 432/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0985 - loss: 0.3057 - val_accuracy: 0.1505 - val_loss: 0.3679\n",
            "Epoch 433/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1012 - loss: 0.2863 - val_accuracy: 0.1478 - val_loss: 0.3691\n",
            "Epoch 434/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0955 - loss: 0.3292 - val_accuracy: 0.1469 - val_loss: 0.3728\n",
            "Epoch 435/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.1005 - loss: 0.2933 - val_accuracy: 0.1508 - val_loss: 0.3709\n",
            "Epoch 436/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1059 - loss: 0.3123 - val_accuracy: 0.1478 - val_loss: 0.3784\n",
            "Epoch 437/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1049 - loss: 0.3233 - val_accuracy: 0.1460 - val_loss: 0.3802\n",
            "Epoch 438/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1002 - loss: 0.3198 - val_accuracy: 0.1457 - val_loss: 0.3803\n",
            "Epoch 439/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1068 - loss: 0.3174 - val_accuracy: 0.1460 - val_loss: 0.3883\n",
            "Epoch 440/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.1114 - loss: 0.2960 - val_accuracy: 0.1431 - val_loss: 0.4020\n",
            "Epoch 441/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1022 - loss: 0.3294 - val_accuracy: 0.1441 - val_loss: 0.4058\n",
            "Epoch 442/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1015 - loss: 0.3012 - val_accuracy: 0.1484 - val_loss: 0.3875\n",
            "Epoch 443/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1023 - loss: 0.3196 - val_accuracy: 0.1475 - val_loss: 0.4115\n",
            "Epoch 444/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1076 - loss: 0.3113 - val_accuracy: 0.1374 - val_loss: 0.4501\n",
            "Epoch 445/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1061 - loss: 0.2880 - val_accuracy: 0.1435 - val_loss: 0.4348\n",
            "Epoch 446/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0984 - loss: 0.3289 - val_accuracy: 0.1474 - val_loss: 0.4017\n",
            "Epoch 447/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0975 - loss: 0.4111 - val_accuracy: 0.1444 - val_loss: 0.4406\n",
            "Epoch 448/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0928 - loss: 0.4044 - val_accuracy: 0.1473 - val_loss: 0.4071\n",
            "Epoch 449/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0931 - loss: 0.4518 - val_accuracy: 0.1490 - val_loss: 0.3993\n",
            "Epoch 450/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0993 - loss: 0.3821 - val_accuracy: 0.1500 - val_loss: 0.4258\n",
            "Epoch 451/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0897 - loss: 0.4030 - val_accuracy: 0.1349 - val_loss: 0.5157\n",
            "Epoch 452/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0958 - loss: 0.3845 - val_accuracy: 0.1428 - val_loss: 0.4603\n",
            "Epoch 453/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1026 - loss: 0.3559 - val_accuracy: 0.1419 - val_loss: 0.4631\n",
            "Epoch 454/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.1028 - loss: 0.3649 - val_accuracy: 0.1424 - val_loss: 0.4634\n",
            "Epoch 455/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0996 - loss: 0.4016 - val_accuracy: 0.1428 - val_loss: 0.4517\n",
            "Epoch 456/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0992 - loss: 0.3974 - val_accuracy: 0.1229 - val_loss: 0.5621\n",
            "Epoch 457/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0877 - loss: 0.4784 - val_accuracy: 0.1122 - val_loss: 0.6340\n",
            "Epoch 458/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0929 - loss: 0.5239 - val_accuracy: 0.1491 - val_loss: 0.4402\n",
            "Epoch 459/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0913 - loss: 0.5194 - val_accuracy: 0.1444 - val_loss: 0.4693\n",
            "Epoch 460/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0959 - loss: 0.4937 - val_accuracy: 0.1428 - val_loss: 0.4469\n",
            "Epoch 461/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1022 - loss: 0.4722 - val_accuracy: 0.1462 - val_loss: 0.4262\n",
            "Epoch 462/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.0884 - loss: 0.4746 - val_accuracy: 0.1470 - val_loss: 0.4602\n",
            "Epoch 463/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0948 - loss: 0.4709 - val_accuracy: 0.1469 - val_loss: 0.4638\n",
            "Epoch 464/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0954 - loss: 0.4597 - val_accuracy: 0.1474 - val_loss: 0.4458\n",
            "Epoch 465/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0961 - loss: 0.4439 - val_accuracy: 0.1512 - val_loss: 0.3999\n",
            "Epoch 466/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.1014 - loss: 0.4353 - val_accuracy: 0.1503 - val_loss: 0.4079\n",
            "Epoch 467/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0971 - loss: 0.4313 - val_accuracy: 0.1494 - val_loss: 0.4094\n",
            "Epoch 468/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1013 - loss: 0.4105 - val_accuracy: 0.1533 - val_loss: 0.3726\n",
            "Epoch 469/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0954 - loss: 0.4073 - val_accuracy: 0.1532 - val_loss: 0.3740\n",
            "Epoch 470/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.0991 - loss: 0.3954 - val_accuracy: 0.1513 - val_loss: 0.3910\n",
            "Epoch 471/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3s/step - accuracy: 0.1036 - loss: 0.3963 - val_accuracy: 0.1523 - val_loss: 0.3853\n",
            "Epoch 472/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1029 - loss: 0.3741 - val_accuracy: 0.1533 - val_loss: 0.3699\n",
            "Epoch 473/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1006 - loss: 0.3904 - val_accuracy: 0.1528 - val_loss: 0.3729\n",
            "Epoch 474/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.1012 - loss: 0.3768 - val_accuracy: 0.1523 - val_loss: 0.3718\n",
            "Epoch 475/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.0972 - loss: 0.3785 - val_accuracy: 0.1526 - val_loss: 0.3550\n",
            "Epoch 476/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1007 - loss: 0.3768 - val_accuracy: 0.1531 - val_loss: 0.3625\n",
            "Epoch 477/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1040 - loss: 0.3539 - val_accuracy: 0.1520 - val_loss: 0.3653\n",
            "Epoch 478/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0927 - loss: 0.3860 - val_accuracy: 0.1432 - val_loss: 0.4556\n",
            "Epoch 479/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0989 - loss: 0.3840 - val_accuracy: 0.1499 - val_loss: 0.3895\n",
            "Epoch 480/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0963 - loss: 0.3594 - val_accuracy: 0.1514 - val_loss: 0.3830\n",
            "Epoch 481/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1068 - loss: 0.3549 - val_accuracy: 0.1496 - val_loss: 0.3863\n",
            "Epoch 482/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.0983 - loss: 0.3589 - val_accuracy: 0.1441 - val_loss: 0.4280\n",
            "Epoch 483/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1062 - loss: 0.3635 - val_accuracy: 0.1503 - val_loss: 0.3929\n",
            "Epoch 484/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0956 - loss: 0.3775 - val_accuracy: 0.1502 - val_loss: 0.3692\n",
            "Epoch 485/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3s/step - accuracy: 0.1009 - loss: 0.3486 - val_accuracy: 0.1494 - val_loss: 0.3677\n",
            "Epoch 486/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1081 - loss: 0.3257 - val_accuracy: 0.1485 - val_loss: 0.3663\n",
            "Epoch 487/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1062 - loss: 0.3407 - val_accuracy: 0.1489 - val_loss: 0.3701\n",
            "Epoch 488/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1s/step - accuracy: 0.1041 - loss: 0.3408 - val_accuracy: 0.1320 - val_loss: 0.4716\n",
            "Epoch 489/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0993 - loss: 0.3398 - val_accuracy: 0.1346 - val_loss: 0.4495\n",
            "Epoch 490/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.0987 - loss: 0.3495 - val_accuracy: 0.1373 - val_loss: 0.4220\n",
            "Epoch 491/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2s/step - accuracy: 0.0964 - loss: 0.3206 - val_accuracy: 0.1366 - val_loss: 0.4546\n",
            "Epoch 492/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1090 - loss: 0.3190 - val_accuracy: 0.1377 - val_loss: 0.4627\n",
            "Epoch 493/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1035 - loss: 0.3202 - val_accuracy: 0.1337 - val_loss: 0.4788\n",
            "Epoch 494/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2s/step - accuracy: 0.0950 - loss: 0.3362 - val_accuracy: 0.1356 - val_loss: 0.4787\n",
            "Epoch 495/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3s/step - accuracy: 0.1002 - loss: 0.3189 - val_accuracy: 0.1454 - val_loss: 0.4169\n",
            "Epoch 496/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.1046 - loss: 0.3109 - val_accuracy: 0.1253 - val_loss: 0.5203\n",
            "Epoch 497/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 1s/step - accuracy: 0.1027 - loss: 0.3657 - val_accuracy: 0.1291 - val_loss: 0.5043\n",
            "Epoch 498/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 1s/step - accuracy: 0.1012 - loss: 0.3163 - val_accuracy: 0.1494 - val_loss: 0.4016\n",
            "Epoch 499/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2s/step - accuracy: 0.0939 - loss: 0.3463 - val_accuracy: 0.1471 - val_loss: 0.4030\n",
            "Epoch 500/500\n",
            "\u001b[1m2/2\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3s/step - accuracy: 0.1042 - loss: 0.2941 - val_accuracy: 0.1482 - val_loss: 0.3936\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluación del modelo\n",
        "test_loss, test_acc = model.evaluate(X_test, y_test)\n",
        "print(f\"Test Accuracy: {test_acc * 100:.2f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xOLJKuxJOiGd",
        "outputId": "d58f5325-b0e2-4114-9bd3-051585df1047"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 913ms/step - accuracy: 0.1770 - loss: 0.6590\n",
            "Test Accuracy: 17.70%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Visualización del rendimiento\n",
        "import matplotlib.pyplot as plt\n",
        "plt.plot(history.history['accuracy'], label='Accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
        "plt.title('Model Accuracy')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "JADcV952OS32",
        "outputId": "8d47e863-5f6a-435f-f6b1-00571e18e688"
      },
      "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": [],
      "metadata": {
        "id": "ejUfqz6mZOFg"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "plt.plot(history.history['loss'], label='Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.title('Model Loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "0DIpw64kNpiZ",
        "outputId": "0c7d944d-39bc-49d7-b9a0-42b160322e78"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGzCAYAAAD9pBdvAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACbyElEQVR4nO2dd3gU5drG79manpCeQCD03psRC0ikiqKoqCiKgA2OIlZsiJ+KXWzosYEeGwqCCggiXXoL0msglDQI6cnW+f54d9rubJINWwL7/K4rV3ZnZ2femZ1yz1M5nud5EARBEARBBAhNoAdAEARBEERwQ2KEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAivwXEcXn75ZY+/d+LECXAch7lz53p9TARBNHxIjBDEZcbcuXPBcRw4jsM///zj8jnP80hLSwPHcbjhhhsCMML6s2bNGnAch/nz5wd6KARBeBESIwRxmRISEoIffvjBZfratWtx+vRpGI3GAIyKIAjCFRIjBHGZMmzYMPzyyy+wWq2K6T/88AN69uyJ5OTkAI2MIAhCCYkRgrhMufPOO3H+/HmsWLFCnGY2mzF//nzcddddqt+pqKjAE088gbS0NBiNRrRt2xbvvPMOnJt7m0wmPP7440hISEBkZCRuvPFGnD59WnWZZ86cwf3334+kpCQYjUZ07NgRX3/9tfc2VIXjx4/jtttuQ2xsLMLCwnDFFVdgyZIlLvN99NFH6NixI8LCwtCoUSP06tVLYU0qKyvDlClTkJ6eDqPRiMTERFx//fXYuXOnT8dPEMEGiRGCuExJT09HRkYGfvzxR3Han3/+iZKSEtxxxx0u8/M8jxtvvBHvv/8+hgwZgvfeew9t27bFU089halTpyrmnTBhAmbNmoVBgwbhjTfegF6vx/Dhw12WmZ+fjyuuuAJ///03Jk+ejA8++ACtWrXC+PHjMWvWLK9vs7DOK6+8EsuXL8cjjzyC1157DdXV1bjxxhuxcOFCcb4vvvgCjz76KDp06IBZs2ZhxowZ6NatG7Zs2SLO89BDD+HTTz/FqFGjMHv2bDz55JMIDQ3FgQMHfDJ2gghaeIIgLivmzJnDA+C3bdvGf/zxx3xkZCRfWVnJ8zzP33bbbfyAAQN4nuf5Zs2a8cOHDxe/t2jRIh4A/+qrryqWd+utt/Icx/FHjx7leZ7ns7KyeAD8I488opjvrrvu4gHw06dPF6eNHz+eT0lJ4c+dO6eY94477uCjo6PFcWVnZ/MA+Dlz5tS4batXr+YB8L/88ovbeaZMmcID4NevXy9OKysr45s3b86np6fzNpuN53mev+mmm/iOHTvWuL7o6Gh+0qRJNc5DEMTFQ5YRgriMuf3221FVVYXFixejrKwMixcvduuiWbp0KbRaLR599FHF9CeeeAI8z+PPP/8U5wPgMt+UKVMU73mex4IFCzBixAjwPI9z586Jf4MHD0ZJSYlP3B1Lly5Fnz59cNVVV4nTIiIi8MADD+DEiRPYv38/ACAmJganT5/Gtm3b3C4rJiYGW7ZswdmzZ70+ToIgJEiMEMRlTEJCAjIzM/HDDz/g119/hc1mw6233qo678mTJ5GamorIyEjF9Pbt24ufC/81Gg1atmypmK9t27aK94WFhSguLsbnn3+OhIQExd+4ceMAAAUFBV7ZTuftcB6L2nY888wziIiIQJ8+fdC6dWtMmjQJGzZsUHznrbfewt69e5GWloY+ffrg5ZdfxvHjx70+ZoIIdnSBHgBBEL7lrrvuwsSJE5GXl4ehQ4ciJibGL+u12+0AgLvvvhv33nuv6jxdunTxy1jUaN++PQ4dOoTFixdj2bJlWLBgAWbPno2XXnoJM2bMAMAsS1dffTUWLlyIv/76C2+//TbefPNN/Prrrxg6dGjAxk4QlxtkGSGIy5ybb74ZGo0GmzdvduuiAYBmzZrh7NmzKCsrU0w/ePCg+Lnw326349ixY4r5Dh06pHgvZNrYbDZkZmaq/iUmJnpjE122w3ksatsBAOHh4Rg9ejTmzJmDnJwcDB8+XAx4FUhJScEjjzyCRYsWITs7G3FxcXjttde8Pm6CCGZIjBDEZU5ERAQ+/fRTvPzyyxgxYoTb+YYNGwabzYaPP/5YMf39998Hx3GiJUD4/+GHHyrmc86O0Wq1GDVqFBYsWIC9e/e6rK+wsLA+m1Mrw4YNw9atW7Fp0yZxWkVFBT7//HOkp6ejQ4cOAIDz588rvmcwGNChQwfwPA+LxQKbzYaSkhLFPImJiUhNTYXJZPLJ2AkiWCE3DUEEAe7cJHJGjBiBAQMG4Pnnn8eJEyfQtWtX/PXXX/jtt98wZcoUMUakW7duuPPOOzF79myUlJTgyiuvxMqVK3H06FGXZb7xxhtYvXo1+vbti4kTJ6JDhw4oKirCzp078ffff6OoqKhe27NgwQLR0uG8nc8++yx+/PFHDB06FI8++ihiY2PxzTffIDs7GwsWLIBGw57BBg0ahOTkZPTr1w9JSUk4cOAAPv74YwwfPhyRkZEoLi5GkyZNcOutt6Jr166IiIjA33//jW3btuHdd9+t17gJgnBDYJN5CILwNvLU3ppwTu3leZYC+/jjj/Opqam8Xq/nW7duzb/99tu83W5XzFdVVcU/+uijfFxcHB8eHs6PGDGCP3XqlEtqL8/zfH5+Pj9p0iQ+LS2N1+v1fHJyMj9w4ED+888/F+fxNLXX3Z+Qznvs2DH+1ltv5WNiYviQkBC+T58+/OLFixXL+u9//8tfc801fFxcHG80GvmWLVvyTz31FF9SUsLzPM+bTCb+qaee4rt27cpHRkby4eHhfNeuXfnZs2fXOEaCIDyH43mn0ooEQRAEQRB+hGJGCIIgCIIIKCRGCIIgCIIIKCRGCIIgCIIIKCRGCIIgCIIIKCRGCIIgCIIIKCRGCIIgCIIIKJdE0TO73Y6zZ88iMjISHMcFejgEQRAEQdQBnudRVlaG1NRUseCgGpeEGDl79izS0tICPQyCIAiCIOrBqVOn0KRJE7efXxJiRGhpfurUKURFRQV4NARBEARB1IXS0lKkpaWJ93F3XBJiRHDNREVFkRghCIIgiEuM2kIsKICVIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAcknEjBAEQRD1h+d5WK1W2Gy2QA+FuMzQarXQ6XQXXXaDxAhBEMRljNlsRm5uLiorKwM9FOIyJSwsDCkpKTAYDPVeBokRgiCIyxS73Y7s7GxotVqkpqbCYDBQ4UjCa/A8D7PZjMLCQmRnZ6N169Y1FjarCRIjBEEQlylmsxl2ux1paWkICwsL9HCIy5DQ0FDo9XqcPHkSZrMZISEh9VoOBbASBEFc5tT3aZUg6oI3ji86QgmCIAiCCCgei5F169ZhxIgRSE1NBcdxWLRoUZ2/u2HDBuh0OnTr1s3T1RIEQRAEcZnisRipqKhA165d8cknn3j0veLiYowdOxYDBw70dJUEQRAEQVzGeBzAOnToUAwdOtTjFT300EO46667oNVqPbKmEARBEMHHfffdh+LiYrpfBAl+iRmZM2cOjh8/junTp9dpfpPJhNLSUsWfL/hy/XG8/Ps+HMzzzfIJgiAIgqgdn4uRI0eO4Nlnn8V3330Hna5uhpiZM2ciOjpa/EtLS/PJ2JbsycXcjSeQc56KAREEERzwPI9KszUgfzzPe2Ub1q5diz59+sBoNCIlJQXPPvssrFar+Pn8+fPRuXNnhIaGIi4uDpmZmaioqAAArFmzBn369EF4eDhiYmLQr18/nDx50ivjIuqPT+uM2Gw23HXXXZgxYwbatGlT5+9NmzYNU6dOFd+Xlpb6RJBoHcV/bHbvnCAEQRANnSqLDR1eWh6Qde9/ZTDCDBd32zlz5gyGDRuG++67D99++y0OHjyIiRMnIiQkBC+//DJyc3Nx55134q233sLNN9+MsrIyrF+/XiyJP3LkSEycOBE//vgjzGYztm7dSoXgGgA+FSNlZWXYvn07du3ahcmTJwNgFQF5nodOp8Nff/2F6667zuV7RqMRRqPRl0MDAGg1DjHiJbVOEARB+JbZs2cjLS0NH3/8MTiOQ7t27XD27Fk888wzeOmll5Cbmwur1YpbbrkFzZo1AwB07twZAFBUVISSkhLccMMNaNmyJQCgffv2AdsWQsKnYiQqKgp79uxRTJs9ezZWrVqF+fPno3nz5r5cfa2IYoQsIwRBBAmhei32vzI4YOu+WA4cOICMjAyFNaNfv34oLy/H6dOn0bVrVwwcOBCdO3fG4MGDMWjQINx6661o1KgRYmNjcd9992Hw4MG4/vrrkZmZidtvvx0pKSkXPS7i4vA4ZqS8vBxZWVnIysoCAGRnZyMrKws5OTkAmItl7NixbOEaDTp16qT4S0xMREhICDp16oTw8HDvbUk9IDFCEESwwXEcwgy6gPz5wx2i1WqxYsUK/Pnnn+jQoQM++ugjtG3bFtnZ2QBYQsWmTZtw5ZVXYt68eWjTpg02b97s83ERNeOxGNm+fTu6d++O7t27AwCmTp2K7t2746WXXgIA5ObmisKkoSOIESuJEYIgiEuC9u3bY9OmTYpg2A0bNiAyMhJNmjQBwARXv379MGPGDOzatQsGgwELFy4U5+/evTumTZuGjRs3olOnTvjhhx/8vh2EEo/dNP37968xInru3Lk1fv/ll1/Gyy+/7OlqfYIQwGonMUIQBNHgKCkpEa3wAg888ABmzZqF//znP5g8eTIOHTqE6dOnY+rUqdBoNNiyZQtWrlyJQYMGITExEVu2bEFhYSHat2+P7OxsfP7557jxxhuRmpqKQ4cO4ciRI6I1nwgcQd21lwJYCYIgGi5r1qwRrfAC48ePx9KlS/HUU0+ha9euiI2Nxfjx4/HCCy8AYLGK69atw6xZs1BaWopmzZrh3XffxdChQ5Gfn4+DBw/im2++wfnz55GSkoJJkybhwQcfDMTmETJIjIBiRgiCIBoac+fOrdHSvnXrVtXp7du3x7Jly1Q/S0pKUrhriIZDUHftJTFCEARBEIGHxAhIjBAEQRBEICExAhIjBEEQBBFIgluMcBTAShAEQRCBJrjFiGAZsZEYIQiCIIhAQWIEZBkhCIIgiEBCYgQUM0IQBEEQgYTECEiMEARBEEQgCW4xwpEYIQiCIIhAE9xiREtihCAI4nKlf//+mDJlivg+PT0ds2bNqvE7HMdh0aJFF71uby0nWAhuMUKpvQRBEA2OESNGYMiQIaqfrV+/HhzH4d9///V4udu2bcMDDzxwscNT8PLLL6Nbt24u03NzczF06FCvrsuZuXPnIiYmxqfr8BfBLUYoZoQgCKLBMX78eKxYsQKnT592+WzOnDno1asXunTp4vFyExISEBYW5o0h1kpycjKMRqNf1nU5QGIEJEYIgggieB4wVwTmr45W6BtuuAEJCQkujfLKy8vxyy+/YPz48Th//jzuvPNONG7cGGFhYejcuTN+/PHHGpfr7KY5cuQIrrnmGoSEhKBDhw5YsWKFy3eeeeYZtGnTBmFhYWjRogVefPFFWCwWAMwyMWPGDOzevRscx4HjOHHMzm6aPXv24LrrrkNoaCji4uLwwAMPoLy8XPz8vvvuw8iRI/HOO+8gJSUFcXFxmDRpkriu+pCTk4ObbroJERERiIqKwu233478/Hzx8927d2PAgAGIjIxEVFQUevbsie3btwMATp48iREjRqBRo0YIDw9Hx44dsXTp0nqPpTaCu2svBbASBBFsWCqB11MDs+7nzgKG8Fpn0+l0GDt2LObOnYvnn38enONa/csvv8Bms+HOO+9EeXk5evbsiWeeeQZRUVFYsmQJ7rnnHrRs2RJ9+vSpdR12ux233HILkpKSsGXLFpSUlCjiSwQiIyMxd+5cpKamYs+ePZg4cSIiIyPx9NNPY/To0di7dy+WLVuGv//+GwAQHR3tsoyKigoMHjwYGRkZ2LZtGwoKCjBhwgRMnjxZIbhWr16NlJQUrF69GkePHsXo0aPRrVs3TJw4sdbtUds+QYisXbsWVqsVkyZNwujRo7FmzRoAwJgxY9C9e3d8+umn0Gq1yMrKgl6vBwBMmjQJZrMZ69atQ3h4OPbv34+IiAiPx1FXgluMUAArQRBEg+T+++/H22+/jbVr16J///4AmItm1KhRiI6ORnR0NJ588klx/v/85z9Yvnw5fv755zqJkb///hsHDx7E8uXLkZrKxNnrr7/uEufxwgsviK/T09Px5JNP4qeffsLTTz+N0NBQREREQKfTITk52e26fvjhB1RXV+Pbb79FeDgTYx9//DFGjBiBN998E0lJSQCARo0a4eOPP4ZWq0W7du0wfPhwrFy5sl5iZOXKldizZw+ys7ORlpYGAPj222/RsWNHbNu2Db1790ZOTg6eeuoptGvXDgDQunVr8fs5OTkYNWoUOnfuDABo0aKFx2PwhOAWI2QZIQgi2NCHMQtFoNZdR9q1a4crr7wSX3/9Nfr374+jR49i/fr1eOWVVwAANpsNr7/+On7++WecOXMGZrMZJpOpzjEhBw4cQFpamihEACAjI8Nlvnnz5uHDDz/EsWPHUF5eDqvViqioqDpvh7Curl27ikIEAPr16we73Y5Dhw6JYqRjx47QarXiPCkpKdizZ49H65KvMy0tTRQiANChQwfExMTgwIED6N27N6ZOnYoJEybgf//7HzIzM3HbbbehZcuWAIBHH30UDz/8MP766y9kZmZi1KhR9YrTqSsUMwLKpiEIIojgOOYqCcSf4wGwrowfPx4LFixAWVkZ5syZg5YtW+Laa68FALz99tv44IMP8Mwzz2D16tXIysrC4MGDYTabvbarNm3ahDFjxmDYsGFYvHgxdu3aheeff96r65AjuEgEOI6D3W73yboAlgm0b98+DB8+HKtWrUKHDh2wcOFCAMCECRNw/Phx3HPPPdizZw969eqFjz76yGdjITECsowQBEE0RG6//XZoNBr88MMP+Pbbb3H//feL8SMbNmzATTfdhLvvvhtdu3ZFixYtcPjw4Tovu3379jh16hRyc3PFaZs3b1bMs3HjRjRr1gzPP/88evXqhdatW+PkyZOKeQwGA2w2W63r2r17NyoqKsRpGzZsgEajQdu2bes8Zk8Qtu/UqVPitP3796O4uBgdOnQQp7Vp0waPP/44/vrrL9xyyy2YM2eO+FlaWhoeeugh/Prrr3jiiSfwxRdf+GSsAIkRACRGCIIgGiIREREYPXo0pk2bhtzcXNx3333iZ61bt8aKFSuwceNGHDhwAA8++KAiU6Q2MjMz0aZNG9x7773YvXs31q9fj+eff14xT+vWrZGTk4OffvoJx44dw4cffihaDgTS09ORnZ2NrKwsnDt3DiaTyWVdY8aMQUhICO69917s3bsXq1evxn/+8x/cc889ooumvthsNmRlZSn+Dhw4gMzMTHTu3BljxozBzp07sXXrVowdOxbXXnstevXqhaqqKkyePBlr1qzByZMnsWHDBmzbtg3t27cHAEyZMgXLly9HdnY2du7cidWrV4uf+QISIyAxQhAE0VAZP348Lly4gMGDByviO1544QX06NEDgwcPRv/+/ZGcnIyRI0fWebkajQYLFy5EVVUV+vTpgwkTJuC1115TzHPjjTfi8ccfx+TJk9GtWzds3LgRL774omKeUaNGYciQIRgwYAASEhJU04vDwsKwfPlyFBUVoXfv3rj11lsxcOBAfPzxx57tDBXKy8vRvXt3xd+IESPAcRx+++03NGrUCNdccw0yMzPRokULzJs3DwCg1Wpx/vx5jB07Fm3atMHtt9+OoUOHYsaMGQCYyJk0aRLat2+PIUOGoE2bNpg9e/ZFj9cdHM83/ICJ0tJSREdHo6SkxOPAoZr4aWsOnv11Dwa2S8RX9/X22nIJgiAaAtXV1cjOzkbz5s0REhIS6OEQlyk1HWd1vX+TZQQUwEoQBEEQgYTECMhNQxAEQRCBhMQISIwQBEEQRCAhMQISIwRBEAQRSIJajOhIjBAEEQRcAnkKxCWMN46voBYjGo4CWAmCuHwRKnpWVlYGeCTE5YxwfDlXkPWE4O5NQ5YRgiAuY7RaLWJiYlBQUACA1bvgPCzJThDu4HkelZWVKCgoQExMjKKvjqeQGAGJEYIgLl+EbrKCICEIbxMTE1Nj1+K6QGIEJEYIgrh84TgOKSkpSExMhMViCfRwiMsMvV5/URYRARIjIDFCEMTlj1ar9cpNgyB8QVAHsGopgJUgCIIgAk5QixGdliwjBEEQBBFoglqMiKm9JEYIgiAIImAEtRgRYkbsJEYIgiAIImCQGAFgJTFCEARBEAGDxAgAOwWwEgRBEETACGoxoiPLCEEQBEEEnKAWIxTAShAEQRCBJ6jFiE7DNp/ECEEQBEEEjqAWIw4tQmKEIAiCIAJIUIsRCmAlCIIgiMBDYgQUwEoQBEEQgSS4xYgjgJXnqfAZQRAEQQSKoBYjQgArQM3yCIIgCCJQBLUYkWkRCmIlCIIgiAAR1GJEbhmhIFaCIAiCCAxBLUbklhEKYiUIgiCIwOCxGFm3bh1GjBiB1NRUcByHRYsW1Tj/r7/+iuuvvx4JCQmIiopCRkYGli9fXt/xehUhgBWgAFaCIAiCCBQei5GKigp07doVn3zySZ3mX7duHa6//nosXboUO3bswIABAzBixAjs2rXL48F6GyG1FyDLCEEQBEEECp2nXxg6dCiGDh1a5/lnzZqleP/666/jt99+wx9//IHu3bt7unqvwnEcNBxgp9RegiAIgggYHouRi8Vut6OsrAyxsbFu5zGZTDCZTOL70tJSn41Hp9HAbLOTZYQgCIIgAoTfA1jfeecdlJeX4/bbb3c7z8yZMxEdHS3+paWl+Ww81J+GIAiCIAKLX8XIDz/8gBkzZuDnn39GYmKi2/mmTZuGkpIS8e/UqVM+G5OQ3kupvQRBEAQRGPzmpvnpp58wYcIE/PLLL8jMzKxxXqPRCKPR6JdxCTGs5KYhCIIgiMDgF8vIjz/+iHHjxuHHH3/E8OHD/bHKOiN27iUxQhAEQRABwWPLSHl5OY4ePSq+z87ORlZWFmJjY9G0aVNMmzYNZ86cwbfffguAuWbuvfdefPDBB+jbty/y8vIAAKGhoYiOjvbSZtQfrcNNQ5YRgiAIgggMHltGtm/fju7du4tpuVOnTkX37t3x0ksvAQByc3ORk5Mjzv/555/DarVi0qRJSElJEf8ee+wxL23CxaGlAFaCIAiCCCgeW0b69+8PvoZgz7lz5yrer1mzxtNV+BUhgJXECEEQBEEEhqDuTQNIMSPkpiEIgiCIwBD0YkSnZWKELCMEQRAEERhIjAiWEZs9wCMhCIIgiOCExIgjZsRClhGCIAiCCAhBL0b0WrKMEARBEEQgCXoxQgGsBEEQBBFYgl6M6ByFRqw2EiMEQRAEEQiCXoyIbho7uWkIgiAIIhAEvRgRy8GTZYQgCIIgAkLQixG9hiwjBEEQBBFIgl6MCEXPLGQZIQiCIIiAQGJES71pCIIgCCKQkBjRCJYRctMQBEEQRCAgMSIEsJJlhCAIgiACQtCLEarAShAEQRCBJejFCFVgJQiCIIjAEvRiRE8VWAmCIAgioAS9GBEDWKnOCEEQBEEEhKAXI1pHzIiNLCMEQRAEERCCXozoKZuGIAiCIAJK0IsRqQIruWkIgiAIIhAEvRjRUwVWgiAIgggoQS9GtBrqTUMQBEEQgSToxYiOuvYSBEEQREAJejFCdUYIgiAIIrAEvRjRkmWEIAiCIAJK0IsRqTcNWUYIgiAIIhAEvRgRuvZaKJuGIAiCIAKCLtADCCjLpuHq47vRnRsEmz0u0KMhCIIgiKAkuC0jp7chsWAD4rkSSu0lCIIgiAAR3GKE0wIANLDDShVYCYIgCCIgBLcY0TAxooWdKrASBEEQRIAIbjHCsc3Xwk5uGoIgCIIIEMEtRjQyNw3VGSEIgiCIgBDcYsRhGdGApzojBEEQBBEgglyMOGJGODusFDNCEARBEAEhuMWIhrJpCIIgCCLQBLcY4aRsGqrAShAEQRCBIbjFiEbKpqHUXoIgCIIIDMEtRhwBrBx4WMhNQxAEQRABIcjFiOSmoWwagiAIgggMwS1GZBVYqc4IQRAEQQSG4BYj8t40FDNCEARBEAEhuMWIzDLC86AgVoIgCIIIAMEtRjgOAKvACoCCWAmCIAgiAAS5GJHcNABZRgiCIAgiEAS3GJG5aQBQRg1BEARBBIDgFiOCZYRjYsRCGTUEQRAE4XeCW4w4LCM6jllEyE1DEARBEP4nuMWIowKr3iFGzFayjBAEQRCEv/FYjKxbtw4jRoxAamoqOI7DokWLav3OmjVr0KNHDxiNRrRq1Qpz586tx1B9gCBGNEyMmEiMEARBEITf8ViMVFRUoGvXrvjkk0/qNH92djaGDx+OAQMGICsrC1OmTMGECROwfPlyjwfrdRxuGoOGLCMEQRAEESh0nn5h6NChGDp0aJ3n/+yzz9C8eXO8++67AID27dvjn3/+wfvvv4/BgwerfsdkMsFkMonvS0tLPR1m3XAEsEqWEZtv1kMQBEEQhFt8HjOyadMmZGZmKqYNHjwYmzZtcvudmTNnIjo6WvxLS0vzzeA0ghhhb8lNQxAEQRD+x+diJC8vD0lJSYppSUlJKC0tRVVVlep3pk2bhpKSEvHv1KlTvhmcUwAriRGCIAiC8D8eu2n8gdFohNFo9P2KBDeNo86IyUJuGoIgCILwNz63jCQnJyM/P18xLT8/H1FRUQgNDfX16mtGwzZfR9k0BEEQBBEwfC5GMjIysHLlSsW0FStWICMjw9errh3RMkJihCAIgiAChcdipLy8HFlZWcjKygLAUnezsrKQk5MDgMV7jB07Vpz/oYcewvHjx/H000/j4MGDmD17Nn7++Wc8/vjj3tmCi8GpAiul9hIEQRCE//FYjGzfvh3du3dH9+7dAQBTp05F9+7d8dJLLwEAcnNzRWECAM2bN8eSJUuwYsUKdO3aFe+++y6+/PJLt2m9foVzdtNQzAhBEARB+BuPA1j79+8Pnnffw0Wtumr//v2xa9cuT1flexxuGh3ITUMQBEEQgSK4e9OIbhohm4bECEEQBEH4m+AWI4KbRhAj5KYhCIIgCL8T3GLEYRnRcuwtuWkIgiAIwv8EtxghywhBEARBBJwgFyMOywiYGKHUXoIgCILwP8EtRkQ3DWXTEARBEESgCG4x4mQZoWwagiAIgvA/wS1GXCwjFDNCEARBEP4muMUIx9JoNIJlhNw0BEEQBOF3glyMOLlpSIwQBEEQhN8JbjGicRYj5KYhCIIgCH8T3GLEYRnRUGovQRAEQQSM4BYjGqUYITcNQRAEQfif4BYjjgqsGkrtJQiCIIiAQWIEgIanmBGCIAiCCBTBLUYcbhqO3DQEQRAEETCCW4wIAaw8iRGCIAiCCBTBLUacLCM2Ow+LjQQJQRAEQfiT4BYjTgGsAFBWbQ3UaAiCIAgiKAlyMeKwjNhtCDew16VVlkCOiCAIgiCCjuAWIxrH5vN2RIfqAQAlJEYIgiAIwq8EtxhxWEZgtyGKxAhBEARBBITgFiOOAFbwdhIjBEEQBBEggluMcIKbxia6aUqrSYwQBEEQhD8JcjEic9OEkGWEIAiCIAJBcIsR0U1jowBWgiAIgggQwS1GBDeNXcqmKa2iOiMEQRAE4U+CW4zIAlijQ3UAqM4IQRAEQfib4BYjsgBWyqYhCIIgiMAQ5GJECmClmBGCIAiCCAzBLUZUAlgptZcgCIIg/EtwixFOFjMSwmJGLlSYAzgggiAIggg+gluMCJYRAEnRRgBAabUVlWbKqCEIgiAIfxHcYoTjxJeRek7s3JtbUh2oEREEQRBE0BHkYkSyjHC8HSkxoQCA3GISIwRBEAThL4JbjMjcNOBtSIkOAQDkllQFaEAEQRAEEXwEtxiRWUZgl4sRsowQBEEQhL8IbjGisIzYkRLtcNOQGCEIgiAIvxHcYoSTbT5vR2oMuWkIgiAIwt+QGBGw29A4JgwAkHWqGAWlZB0hCIIgCH8Q5GKEU/Sn6dM8Fu1TolBcacFbyw8FdmwEQRAEESQEtxgBFP1pDDoNpmS2BgAcyisL4KAIgiAIInggMaKRSsIDQHIUixs5V24K1IgIgiAIIqggMSJz0wBAQiQrC3+u3ASe5wM1KoIgCIIIGkiMyNw0ABAXYQAAWGw8Sqqogy9BEARB+BoSIxrBMsLcNEadFtGhegBAYRm5agiCIAjC15AYcbKMAEC8wzpSSHEjBEEQBOFzSIw4BbACUtwIWUYIgiAIwveQGHEKYAWA+AghiNUciBERBOFvyguA3x8Fzu4K9EgIIiiplxj55JNPkJ6ejpCQEPTt2xdbt26tcf5Zs2ahbdu2CA0NRVpaGh5//HFUVzeQCqcqbhqyjBBEkLH4cWDnN8Dn/QM9EoIISjwWI/PmzcPUqVMxffp07Ny5E127dsXgwYNRUFCgOv8PP/yAZ599FtOnT8eBAwfw1VdfYd68eXjuuecuevBeQXTTSGIkMZLVGjlWWB6IEREE4W/OHQn0CAgiqPFYjLz33nuYOHEixo0bhw4dOuCzzz5DWFgYvv76a9X5N27ciH79+uGuu+5Ceno6Bg0ahDvvvLNWa4rfEMSIXYoZGdAuAQCw6mABzhRT0zyCuOzRGQM9AoIIajwSI2azGTt27EBmZqa0AI0GmZmZ2LRpk+p3rrzySuzYsUMUH8ePH8fSpUsxbNgwt+sxmUwoLS1V/PkMjY79t1vFSe2So9CvVRxsdh7fbDzhu3UTBNEw0IcGegQEEdR4JEbOnTsHm82GpKQkxfSkpCTk5eWpfueuu+7CK6+8gquuugp6vR4tW7ZE//79a3TTzJw5E9HR0eJfWlqaJ8P0DBUxAgDjr2oOAPhxaw4qTFbnbxEEcTlBlhGCCCg+z6ZZs2YNXn/9dcyePRs7d+7Er7/+iiVLluD//u//3H5n2rRpKCkpEf9OnTrluwG6ESP92ySiRUI4yqqt+HXXGd+tnyCIwKMLCfQICCKo8UiMxMfHQ6vVIj8/XzE9Pz8fycnJqt958cUXcc8992DChAno3Lkzbr75Zrz++uuYOXMm7LI4DTlGoxFRUVGKP5+hcc2mAQCNhsNdfZoCABbuPO279RMEEXjIMkIQAcUjMWIwGNCzZ0+sXLlSnGa327Fy5UpkZGSofqeyshIajXI1Wi0TAA2iEZ1oGXHtQ3Njt1RoOGBnTjF2nLyAnPOVfh4cQRB+QUcxIwQRSHSefmHq1Km499570atXL/Tp0wezZs1CRUUFxo0bBwAYO3YsGjdujJkzZwIARowYgffeew/du3dH3759cfToUbz44osYMWKEKEoCihs3DcBSfHulx2JrdhFGfboRALBsytVol+xDSw1BEP5HbhnheYDjAjcWgghCPBYjo0ePRmFhIV566SXk5eWhW7duWLZsmRjUmpOTo7CEvPDCC+A4Di+88ALOnDmDhIQEjBgxAq+99pr3tuJi0LCmeGpiBADSGoVha3aR+P6d5Yfw5b29/TEygiD8hTybxlIFGMICNxaCCEI8FiMAMHnyZEyePFn1szVr1ihXoNNh+vTpmD59en1W5XvcxIwIJEYpfclrDhWirNqCyBC9r0dGEIS/0Bqk15ZKEiME4WeoN00NbhoASIhQihGrnUfWqWIfD4ogiEDx5h87kF/aQNpVEESQQGKkFjHibBkBgJ0ni304IIIg/I4smH7F7hOY9uueAA6GIIIPEiO1iZFI1/oDO3Iu+HJEBEH4G1lvqjCYcKSgLICDIYjgg8SIGDPixk0TKVlGejVrBADIyrnQMNKSCYLwDrxU8yiMMyFU3wAy/QgiiCAxIlhGbO4sI5IY6ZYWA44DSqutKKow+2N0BEH4A1kAeyhIjBCEvyExUoubJtwoJRyF6LVIjWYpgCeoABpBXD7ILSMwIYTECEH4FRIjtYgROc3jw9EsjqX8jfp0I+ZsyPblyAiC8BfymBGuGqEGEiME4U9IjGhrFyNzx/XGQ9e2xMjujdEsLlycPuOP/b4eHUEQ/kBmGQmFCSE6EiME4U/qVfTsskK0jKgXPQOA/m0T0b9tIgAgPY6KIRHEZYddLkbMMOrpOY0g/AmdcR64aQAgzqkIms1OWTUEccmjcNOY6LwmCD9DYsRDMdK/bQIiZUGt5ytMvhgVQRD+xMlNY7baa5iZIAhvQ2KkljojzsRHGLF7+iDEhrNeFgWlJhzMK4WdnqQI4tJF5qbVwwqLjcQIQfgTEiN1iBlx+YqGQ1IUq8z64P92YMis9fht9xlfjI4gCH8gs4wwMUIPFwThT0iMiGLE4tHXhGJoZ4qrAAC/7iQxQhCXLLKYER1s5KYhCD9DYsTDmBEBeZl4ALDSkxRBXLrI2jvoORvM5KYhCL9CYsTDmBEBg065606cr/DWiAiC8Dd2pWWEYkYIwr+QGNHo2X8PYkYAFsgqJ7ekGlVmz5ZBEEQDQRYzoiU3DUH4HRIj9XTTjL+qOcb0bYrF/7kK0aFM0JB1hCAuUXh5Ng1ZRgjC35AYqacYiQ7V47WbO6NT42ikx7MS8SfOkRghiEsSmWWEuWkoBowg/AmJkXrGjMhp4RAj2WQZIYhLE6c6IyZy0xCEXyExUo86I86kO5rn7TldgvzSam+MiiAIf+JiGSExQhD+hMRIPd00ctLjWfO8P/fmYcA7a5BXQoKEIC4p5GKEIzFCEP6GxIggRmyeFT2T0yI+QnxdabZh7eGCix0VQRD+RFGBlbJpCMLfkBjxomVE4MT5yosZEUEQ/kZRZ8QKq52nflME4UdIjIgBrPWPGYkM0Sve7z9bejEjIgjC3yhiRthri52sIwThL0iMaIWiZ/W3jADAyG6p4uv9uSRGCMJrnM0CVs8ELFW+W4dTbxoAlN5LEH6ExIgX3DQA8OatXbD+6QHgOKCwzISiCrMXBkcQBD6/Flj7BvDP+75bh8Iywq4FFDdCEP6DxIiXxIhRp0VabBjiwlmZ+NwSHz7FEUQwkrfXd8uWuWT0nGAZITFCEP6CxIgXYkbkJEUxMVJQavLK8giCcMBxvlu2U50RgCwjBOFPSIx4yTIikBQVAgDIo+JnBHHpoBIzYibLCEH4DRIjPhIj209cQElV/WuXEAThhJ8sI3pHzAi5aQjCf5AYEcWId4SD4KZZsPM0BryzBrtyLnhluQRB+BC7SjaNlbJpCMJfkBjxesxIiPi6qMKM5xf6MOiOIIIKf8WMsNdmm3euCQRB1A6JEa+7aYyK9/tzS1FcSWm+BHHR+NRNo6zACgBmsowQhN8gMaLxTtEzgcTIEJdp206Qq4YgLh5fihFJeGg5HhzsFDNCEH6ExIiXLSNNGoWKr2/u3hgAsO1EkVeWTRBBjS8tI05uWmqWRxD+RRfoAQQcL8eMxIQZsODhDBh1Wuw7W4KFu87gAJWHJwgv4J+YEYAFsZJlhCD8B4kRL1tGAKBns1gAQGE5K3x2vpxiRgiiXsjcJ/6KGQFY3AjVGSEI/0FuGh+IEYG4cAMAUJ8agqgvCoul/ywj5KYhCP9CYsSHYiRWJkZ4niLzCcJjbDIh78eYER1sMJEYIQi/QWJEECO8XdEsyxsITfPMNjvKTd4XOwRx2SMXI36OGSExQhD+g8SIEMAKeN06EmrQIlTPlk+uGoKoBzY/tVRwFiOcDSYrFT0jCH9BYkQji+H1oavmPIkRgvAcuWXESy0bVFFz01jIMkJ4kT3zga+HAqW5gR5Jg4TEiFYvvfZFEGuEI26EMmoIwnPkYsSXVhKVAFZy0xBeZcF4IGcj8Pf0QI+kQUJixE+WkZyiSqpbQBCeIhcgPjg/RRypvXaexaXoYCU3DeEbTGWBHkGDhMQIJ9sFXip8JkcQI68s3o/ur6zAv6eLvb4Ogrhs8bNlxOwovUSpvYTP0BoCPYIGCYkRjvNpei8nywAoN1mx8dh5r6+DIC5b/BEzwvMyMcLctpRNQ/gMEiOqkBgBJDFi835cxxUtYhXvKauGIDxAbg2x+chNI6sBZBLECEdihPAROhIjalA5eADQGQFrtU/EyM3dGyM9Phxbjp/HO38dxjlHiXiCIOqA3BriM8uI5J41yS0jFooZIXwAWUZUIcsIAOhC2H+r94WCTqtB7/RYJEaxdVCfGoLwAH/EjMgyaUw8uWkIHyCvwC3cbwgF9RIjn3zyCdLT0xESEoK+ffti69atNc5fXFyMSZMmISUlBUajEW3atMHSpUvrNWCfoGWVUn0hRgTiI6hPDUF4jD+yaWSB61IAK2XTEF5ELqrJMqKKx26aefPmYerUqfjss8/Qt29fzJo1C4MHD8ahQ4eQmJjoMr/ZbMb111+PxMREzJ8/H40bN8bJkycRExPjjfF7B50gRqp9tgqhNPx5ctMQRN3xt2VE5qahbBrCa5grpNckRlTxWIy89957mDhxIsaNGwcA+Oyzz7BkyRJ8/fXXePbZZ13m//rrr1FUVISNGzdCr2cnenp6eo3rMJlMMJmkm3Zpaamnw/QMQYzYfCcU5JVYeZ4H58umXwRxueCXbBp5zAg7T8lNQ3gVS6X0mq79qnjkpjGbzdixYwcyMzOlBWg0yMzMxKZNm1S/8/vvvyMjIwOTJk1CUlISOnXqhNdffx02m3sT6MyZMxEdHS3+paWleTJMz9H53k0jVGI1We2oMJP5lyDqhF+yaSTRYealOiMkRgivYamSXvuyeN8ljEdi5Ny5c7DZbEhKSlJMT0pKQl5enup3jh8/jvnz58Nms2Hp0qV48cUX8e677+LVV191u55p06ahpKRE/Dt16pQnw/QcHwawCoQZdAgzsKZ5M37fh5d/3wdeHtREEIQr/rCMyLp1mxWpvfTQQHgJuZuGxIgqPk/ttdvtSExMxOeffw6tVouePXvizJkzePvttzF9unqNfqPRCKPR6OuhSQg+PB+KEYC5airNVfhlx2kAwN1XNEWrxEifrpMgLmn8HDNikQewUqM8wlvI3TQ+qPR9OeCRZSQ+Ph5arRb5+fmK6fn5+UhOTlb9TkpKCtq0aQOtVitOa9++PfLy8mA2N5DMEtEy4rsAVgBIiFQKrFMXqtzMSRAEAKdsGt/GjNjAwQp2ndLCTm4awnsoxAhZRtTwSIwYDAb07NkTK1euFKfZ7XasXLkSGRkZqt/p168fjh49CrvMFHr48GGkpKTAYGggUcVCRTwfFD2T85/rWuG6dlLG0RkSIwRRMwrLiG9jRuy8BhaHGNHBStk0hPcwkxipDY/rjEydOhVffPEFvvnmGxw4cAAPP/wwKioqxOyasWPHYtq0aeL8Dz/8MIqKivDYY4/h8OHDWLJkCV5//XVMmjTJe1txsfjJMnJduyR8fV9v3HdlOgDg5+2nsP+sjzOFCOJSxi8xI46OvdDA7Ch6ZnTUGaG4LsIrUABrrXgcMzJ69GgUFhbipZdeQl5eHrp164Zly5aJQa05OTnQaCSNk5aWhuXLl+Pxxx9Hly5d0LhxYzz22GN45plnvLcVF4ufYkYEGseEAgD+PV2CYR+uR/bMYZTqSxBqKLJpfBszYoNGrDNi5Cyw84DVzkOvpXOTuEgsFMBaG/UKYJ08eTImT56s+tmaNWtcpmVkZGDz5s31WZV/8EM2jZzGjUIV7wvKTEiKcl8ieMfJC0iJDkFqTKjbeQjiskThOuWZFUOjdTt7veAFywgniREw4WOy2qHXUtcM4iIxUwBrbdBZBrivwGoqA46uBKq960pp7CQqDueXuZ33QG4pRn26EVe+scqrYyCISwJna4gvrCMOV4wdGlQ7ip4ZwUSQyWxVpP4SRL0gN02tkBgBZBVYnQJYV70GfHcL8E5r4NwRr62uiZNl5Eh+udt5d5y8IL4m/zURdDiLD1/EjdhllhFHzEiYxgKAR8SPNwKfX0NPs8TFQW6aWiExArgPYD2zXZp+coPXVhcXYcRzw9ohRM92/yuL9+NQHrOOFFWYsf5IIex2JjwMMhMxVW4lgg7nBwSfWEZcY0ZCNVZEoQLGs1uAvD1AmXpRR4KoE5RNUyskRgBZAKvTha8oW3pdXuDVVT5wTUu8OaqL+H7Ml5tRUFaNOz/fjHu+2opl+9jFz2yTTMR7z5TAaiOTMRFATm4CfrkPKDnjn/W5WEZ8cCFXiRkJ5SyI5WTuU44ulcRFQEXPaoXOMEDdMlJdAlSek9774Mkoo0UcokJYDPG5cjM++PsIDjniR/7cy9ZXUiVdjO/4fDPeXn7I6+MgiDozZwiwbyHwm59S8/1oGZHHjIRyFsRCJkZ8XIOIuMyR31tIjKhCYgRQb5Qnt4oAQLmy6qw3SIwKwa6XBuHze3oCAH7ffVb8LESnwbHCclyoUF4E/7vuuNfHQRAecyG79nm8gXOMiE9jRjRizEgIZ0EcJwtcJ9M6cTHI7y10LKni8940lwRiAKtcjDjd9H3kM9ZqOHRrGsNWUS0dpL/sOC32sHHm9IVKNGkU5pPxEETd8FPtDSdLSEl5FaIbeXkdipgRZhlxESO+qnFCBAeK4n0kRtQgywjgxjLiECNxrdh/b1lG8vcDe39VTEqMDEFKtPs6I85sO1HknbEQRH3xV5E+pwv38j0+6ODtECM8z6HaETMSAgtiIRcj5KYJBBUmK6b9ugfrjxQGeigXh8IyQm4aNUiMAIBWRYyUOlwmTXqz/+X5Yj2Ci2Lhg8D8cUDBAcXk9ilRdV7EthMXap+JIHxKYCwjPulP49YyIosZ8VUpeqJGPl1zDD9uzcE9X20N9FAuDrKM1AqJEUA9gLXCocSTOrH/NjNQ5QURUHme/Xdy+4zomgKOA/qkx9a6iAO57Int522n8Mz8f2GzU/0RAsDCh4AvrvOPS8FvlhHltujggwu5Sp0RI8yI40qkeXzVpI9Q5Z8j57D+SCFOXaisfeZLAYoZqRWKGQGkrr3yA0YQDVGpQGgjJkTK84Gw2sVCjQgK2VyhmHxz9yYY3DEZFSYber/2d42LOJRXBrudx9ML/gUAXNMmAcO7pFzcuIhLn90/sv8n1gMtrwvsWLyFkwiorPZBM0tZNo2Q2msAZdMEimqLDXd/tQUAcH2HpACPxkuQZaRWyDICSJYReQCrYBkJTwAiU9nrEvWAUo+wqosRAAgz6BAbblD92ru3dcW25zNh1GlQabbhZJH0xHCu3D89dYhLBJVjy9ucKTHhvD+OOyfLSGVVlZsZPWTjx8CSJ5jrVVZnREjt1dtNigDWA2fO41TRZfKU3sCptkgxFVWXS6FHhRi5TLbJy5AYAdQDWEUxEg/Et2avCw9e/LpEy4h6CXitRjJ/J8ua52V2SEJCpBFtkyMBAPvPShdKctM0AAJtxpdf4CxeumHXQKXZjtWH/BBU6ORyqqjykmXkr+eBbV8Cp7eL+84GDSwOy4ieNyuKnr21dC+ufmu1d9ZN1IhVdj3jcZlc28hNUyskRgDXAFabVYoPCU8AEtqx1wXeECOOddTw9Co00ruvX7o4LdLIPGrtk1mg664cKX6FxEgAsduBs7uAN9KAdW8Hbhzym7YfxAgPIL/UBy4TZ5wu3FXVXrDGyAPRTaVSNg00sGqYZURrNyMc0vbpHbEq8qd2wjdYZFWmLVaZMLmUe3PJLCOni8pc6kcRJEYYzpYRIV4EHIsXSXSIkcKDbJ49812yYeqE3SZe+GoSI79N7odfH7kSD1zdAsM7p2DCVc2hcVhMmieEAwD2ySwjJitdIOvFhRPAT2OAU/WM1K8uBWZ1Aj7vz8o9r3rVm6PzDLkZ2C9ihEOBP8SIk2WkyhsxI3KBw/OybBoOJo5dCzSwIxSS8NGBnWPHCt03tSS8g1yAmGTCxGS9hFthyCwjVosFH6z0XuPVywUKYAVkYsRxoRNcNGFxgEYrWUZyd7Nshfy9QFg88FgWYIys+3rkNww3bhoAiI8wIj6CjemTMT0Un6XHsWJn+85Kkf6l1WT2qxe//wfIXgccXAy8XFL7/M7s/w0oderRYqkG9HWvGeM1FJYR38eMAEB+qf9iRmw8By3Ho9rkhXXKs+Z4O+BwBdihgZWTYrb0nCTyBcvI0YJydEyNvvgxEG6R9+MyywRIpdmGEL02EEO6eGTXfh1nw65TxYEbC8AeWA4tZYHuod6uIlg/yDICSAGsvI25aISeNOEJ7H9sS0CjYxfG/L1sWuU5FgTnCXK/YT2DDJvFMcuIXICUVFINhHpxkQHJF6pURGDB/otaZr2RC12T75/eeXDIL/OHZYTt4yowcW4yVYsdreuN/DzkbYpy8FZOr/oVQZgcKyDLiK+Ru2nKqqVrW6X5EnrocnYpyY45Lew4ml+GgLLuHWD+/cAPowM7DhkkRgCldaO6BKgQxEg8+68zAM2ulOYJc0xf9zZw4p+6r0f+9GquAE7vcO2Bo4bdBmz9Aig4gGZxrmXgS6tJjNQLT6xaKpy4oHIzzvv3opZZb+RZJ9X1sPLUaR3STYIHUOBHy0iVI8tFy9su/niXixFrtSy1l4NGo5FiyGSIlhFy0/gcuTVE/qBVealk1lhNwMe9gXn3sPc8r8jU1MGGCrMtsNlZ+39j/09tCdwYnCAxAgBaPWB0VECtKlK6aQR63ie9vuVzoPNt7Klq86d1X488dbjgAPDldcCH3Wr/3s5vgKVPArOvQJhBh4RI5cVS3tmX8ABj3aveqhFqdL1pIT9QlhHZMWAqdT9fXbGaXS0s8uMXHArKvGClqA3HdlXzbF/rOBuKLjb4T+6msVSJqb02aKDhOFU3mxAzsnRPHnq9ugJ/7sm9uDEQblFYRkySNeSSESMn/gHOHwEO/M7eO8U9aR3H0hfrA9j0NKGt9DrQmYAOSIwICH6zqgtAmeNCE5ksfd5uBJDYAYhvA6RfBXS9g013bqhXE3JTev6eun/v9HbF23Qn6wiJkXoSIvP92z0PjrPyKlVITQEyv8qPrYu1jPA88OmVwIfdWQyMyjp4cLDYeFyo9HFWgJNlRA/bxd+U5JYRS5Usm4ZjqfU6VzGih7TOc+VmPPz9zosbA+EWecyInEvGTaORxbVYzU4inrlpAOCnbacCd+2OlBXJLDoWmDE4QWJEQKisWlkElApiRPaD6QzAg+uBR7awgNdGzdn0Cyfq3rPGXZnu2pQprzw505w69pKbpp4YIqTX1cUef92iFt1v9UMchRo2L7ppCg+xJ7uKAmVcjVUSHpwj6NPnQaxOMSM6WC8+vdbZMuIQXCZeDw0HKaBdRu+mEWjhyGQjfIvFpn49rTRdIpYRjSzuyFyuOG8AQM/xaJEQDrPVjn+OnPPz4BzI3bp5HjwY+xASIwKiZaRIsoxEpSrn0eoAjWOXRacBnIaldJYXSIKivBAodtNZ1Ormwm0uB/L3uRcrTmIn2anDb4MKYK2HhSFwyPZrhecXBatZ5fd09xv7GoUYuUg3zYn10muLzK9tkwfhsRtDga+DWB0XzWqZZaTqYsWIcxq0I5W/CJHMTaNiGRnaPg5GnTKTg1LqfUBZHsLytqt+VHmp1HixO7lMnVoJ6DgbrmubCABYfajAnyOTkAukcw0jzZjEiECo3DLi6Ngrt4w4ozMAUU3Y698eAd5oCpzNAt5pxWpPqN0Q3ImNTZ8ws/hvk9Q/d7KMpDiKogmUmay+993XhbVvAW81bzAHd62oVdz1AJua8LAFSox40U0jD8qWp6DLtlcI6PR5EKsYM8LEiA62iy8RrrCMVIpi5ALvXozAZsHTQ9oqJh3Jp2BWr/N+R/RedQe6cUddPqq6VNw08uuCqczVTcPbMKAdEyNrD/uhirEazudAA4DEiIDgpqnJMuJMbDr7f/RvVtthzRvSZ2qxJO5uVOveYv//naf+uZMYSXWyjPA88PjPWfhu88max+trVr/G3B1r3wzsOOqK/AZe6bllxGZViZcImGVEntpbyixUh/9iVjtPObNDtizZDVe2DiHV1adVWO02CNYryU3jBcuIczaN47cv4iPdxozAZsGAtolY9cS16NyYxRrd8NE/2CmrhEyA/WZLngB2/+R+nlWvAuvfc/N9Jjiu0LgGgldcKm4aedFBk6ubRgsbOqWy4PnCMlNgqvrKrxcNpAkkiREBwTJy4aSkFGuyjABAo3Tle7loUHs6re+P7iRGnN00APBb1lm8sGhvw7CQqF3MvY03AkUv0jJit6i5aQIUMyI3DZvLgeOrgB9uYzcGj5Zjl8Q4AJjVO9caHJYRn9YakVkSxQBWztsxI5XMGgqgCFHg3MSMCPu3RUIEuqXFiJM3HTt/aZcp9zZ75rOePwsfVP+85AwribByhqtwl+1HM1zrvVRZbNhy/Dyenr+7YQftu7GMyGvYRBk10GtZAPxFZ4fVB2dB3gAgMSIgWEby97H/ITGAwbWmh4KUbsr35Xmy1ypPpO7cNLWivNilRoe6mQ/I80eJbgdniqswZNY6/LAlRxnXcpEps7Xy78/AzCbA9q8vbjlycViPmBFezQrSEGJGAOC8wzLnXCG2NqouKMuly4vzyZ7wDJxDjPjSTSMTWFVeddO4iRkRLCN6lfNetn//c10rRIaw4tVvLz+Ea99eoyjOFdTI6+yodaeV3/icU8dlD3AWuFZarTRbMfrzzfh5+2m8ujhAKfR1wSq3jJSKx5uJkx7SON4mdmgPjBiR/Q5qFt4AQGJEQAhgLXT0nKnNRQMA3e5Svi88LL0uz3ed35Mb1fljwMlN7LXTk1dMmKSwhbLxAv7snfHxqiM4mFeG5xbuUd60QnwsRn6dyP4vfvziluP8BOMJu39C97MqpuiG4KYBpOwgs4f+YOfjVuGmkbbN4HDT+LQ/jUwAVCvcNBcZJO2cTeMQoheEAFa9itiXjSUxKgQzbuwovs8pqsS6wwHKimhoyMWvmsBXVKF2OufEnmCS5U2O3E3zz9EGvL/dWEaqOdm12m5FbDh7f648ANcMhZsmQNcsJ0iMCAhuGoHotNq/ow8F7l4gvZcr4gvZrAnbzm+laZ64aT7qAcwZwtIs5W4auw0cJ9W36NVM2VfgeKF/+pIASo20Yvte6Y3Gx/0juBqWX1Vc9yI+8pPQk+Zydhuw8EFEWlQuiP46sQsPAd+MALIdmS/OlpGqYvZf3qfmn/dZU7+aAlzl1j2ghgBWIWbEl5YR6ebj1WwaZzEis4xoOAAGlRRep3M3LVZpPdEG+5XUamJWNcGyDLgeS4AyWNLZMlIlxd/IOyYL/L77rPjaLx2j64siZqRMPG+qIXNf262IC6hlxKT+OoAE+ykkEebULKj51XX7XqtM4PZvXadv+5I1YPv9P6JPul4xI2d2KsWIYxkPXdsSjcL0eHFEB7x7W1exTLw/LSMxYVJTsdmLN0sf1MXsd3QlyyKqj7/dneWl9CzwZjPgq8y6LcfZXF9Xaog+V3Xd+IJ5d7Mmf/8byd47ixHRMiITI3+/DJzdBWz7yv1ynd2LcouRbB063gKAR15pNaxuilRdNA43jYXXimZ779QZkT+dl4s3wSI+yr1lRHBdOY5X51o/32w8iSX/BnFV1m9GAG82B87JrMNlKtZhubh1tkbKLCMRnOv5KL9p23kEJvCzLrhYRti4hSBsAIA9AG4aSzXw3ShWqj43S5pOYqSBEduC1Q0RaHld3b9bW9fDxY8D39zIUn89xVKhvGE7Dpxnh7bDjheuR+OYUIzq2QSTB7QCwAJZ+72xCkv9UK463CBZKOI4WSpzXQKivrsFWP6cZ719BNz1lDmwmP0/u6tuy5FbMayeiBH385qq/JQmJ1z0hZuks9AVLCNqbho1X76As5vGrO6m4cBDzzERcvVbq30jSBzixwotbA4xovdGzIj8dy89CyEmqxjhNceM5O0F3mkDbPkciU4tGTYdP49JP+xEXkkDfmL3FeZKR48TpwcLNVe1/Hh07lwuFyOo/Xw8lBfgZnPukF9LzOXiNdsEPexC1Wa7VRQj5/0lRnI2ssxPuWAEKIC1wRHaCOj7sPQ+sYNn362J/YuA7LXAljr2sZG7GcyVTpYR6elUo5HcNa2T2A26pMqCM8VVeMQP5apNsgqkCjHiiQWoPp1z5QGyblwyoz7diGV7VczEchRlwT04IWvouGzzZDnexEWMOEze1iqWISMXIEK2SHUp8PO9wN5fpc/Ep1nHseUmgBUArm/Djvvckmpk+aIlukNoWaAF76hq6fXU3uIctg59FKzQMReoOzfNmpmsKu2fTynOPTnL99VyzF2OuCsnXqubRiYmzBWo2rtYfCu3jGg44PsJfV0WldtQhZ/CMiIVPTPxeliFW67NIrppzvsqZiRnC3sQFtyy7iqtUmpvA2TgS8BVjwNj5gOc+sVGlZAY743hs6uAszIhYalUPsm5iUlol+zIBPAjcjNpHGRxCLUpbbmlh6vHISg3o1fI3QrScnecvICHvpPVy1BDIUY8sGjUYBkxwg9ZFTLfOnSOfWF3EmXy8vaWSslSAkhiJOt7JpTnj5M+E55m45ilzV0AKwB8ckcnZLZPAlBD8aaarDC1IbOM2DUse0XHXYQYqS5h1jOFmGTHjDk0AYAj9sOdm0YrSze1WfHSDa4PLP6wSDY43BU5VHXTyPa9XIz8PQOhR5eIb+UxIyF6rWop/mJf90WqL25iRky8TrTw4ePeSAxhx/HP20/jYJ4Xmls68/UglnG4bBp7n7dXfT6yjDRA9CFA5stA6+s9+15tlhFPyNsD/DxWei8z8wFwq2JD9Fq0dDphy02+rVhYLSuHHa9w09Si9OUHf32CXeXWhzL1J1EN6uA2kO9LT07IGsSIjrP7vgumvDOwzcwsHy7ZNDJxKKsyCkD99xEqBgv7M64l+y/PeHD6HmezYkgn1kxSVYz8+QzwVgtWW6I+2AUxogM4Jkb0sNbfTfPFQGDeGGDXdy4fHev6JAA4YkbU3DRmpXvw/FHcf1VzvHJTR8Vs+8764KbS0HEWI3Gt2X81N43FjZtm638Vs8ktIyF6LZKjXGsXFTfUWiPy86TkNFDAztdqXgebcMu1VCDdzPZbJ+44/u+Xjb4bT/Y69t+dZYRSey8jnM26Qpn4+iIvOlVV7JoT7iboM8KoU7w/ku9bn2qVWbrhN+Lc37RcqMHNUSfkGSJqFzwARtRwggn7z7l7a33Wr7Z4Xz9pXMiWrczGLCXuYkYAtr+riqT3wrbKb7r5e5kV5IyjL0jjXuy/wjLidPG3mZHRMg4AuwlbnONGtnzGLDQbZtVlq1xxrM8CLXitzE1THzFSXcqa/wGuTRHb34j8lIEAHGJEfj5rjdJYKmX70FFPIzpUWZyr3GS9fHvWlJwB3mkL/P6ocrpzDEKLa9l/tRo3CsuI49hSiWuSx4yE6DSKDEKB4obUk0uOPGbkzA5g08cAgGq7DhGcdG1ItJ7FA9o/sNj4AsYVvOF6/niL0rPsnHf+nQQotfcyQn6iGKM9t6zURGWR0hKw+0fgzXTg+FqXWYUy1QJHCnybWSO3jITBg1QxN+midUZ+8ZJbRmQiLdSdGDFXsrTp+ffXP7W3lnnLKi5CbBUeqr3RnfPFu6LQ1RojF0zmCuWNdM3rwA93KJ9Sc3cDh5cx4RvbAmh6heO77t00sJmRGh2CEL0GNjuPMxfc7Be5MPIEx5itvBdiRg4vc/9Zs36wO44dDQelm0YQJnarkxhhT5mNZBllAgFJ1fQH/7zP4kB2fiMdb1YTcGqrcr6ON7P/eXtdz2+FGHEc5/J0YAdyN028U7CwQIN107i5plXxSitws7KdeE7/IwAgU7sLxwp89PDI29jxz9uAsDjXzymb5jKl2ZVKMdJvysUtr6pIaRnZMIs92f10l8usTw1ph9t7NUGnxizAc72P21Nrqkvwrv5TzL6iBK0ayU402U1r75kS16I+8pupJ1ksAnIxIO8pI1tviDsxkrub9Q3au0A53RNrhhvLjsVxsTlfXM9GdYeXA5/0qb2Ym/M+qyisOQjNUqm0jADA4T+BggPS+9zdwL6F7HWHkYAxgr2WWUZcyt/bLOA4Dulx7IZ94rxsv8jFUX1K9x9aBnw/CgCLGYFGcNPY6pfSebIGM3hab7GNAsumkVlGDI79YDMrjzWHRU5egFDgfHkDvUleLKdlouPcISbOfr4XKMkBIpKAER8CN30CNOvHbno2E5D7r3IZam4aR5rpKS4Vc62DACjdNBOubgEA+H1yP4zomooHr2XvG6xlRH59uvI/4stGUJ4HmiPLFe+PZav0M3PHnvnA3BtqqRwte0jOYqIHPca6zkZi5DLjmqeAmKbAsLeBtsOAYe8A9ywCrnsBuHUOkDmjfsutLFI/WFRKy0cYdXjr1q5445YuAIA/dp9F82lLMOyD9ej16t8YP3cbK93uJdqVb8Io7Xr0Pj0HTSIkq8S54lJsO1GEQ3lluOGjf3DTxxuUX5TfzD2xSIjfcRMEJ1tWCMduCK69etRdXFZTBb5cf7xuplI3YxZ6TxSV1NMi9ccU9n/v/FrW7yScKgrUm/YJmCuUMSMCcndF3l6W9gewJ1tDpPRdB1az03odAkiocXPyvOxGI1+2qR5xFAvGiy8t0InBozpY62cZETpxO5PSFUjpBuEw4ZzrjAiizGZR7kNHEHFMqKtl5HyFGXvPlDTcOhj1ofQsE6wCZ3YCix5mohYAhrwB9LwX6H43sxSnObJfTm1RLkfNTVN4EACwWnMFvrAOByC5aYZ3TsENnVmPsC5NYvDRnd3RKZVZgIurGoDoqyxydZsL1+vbvgEGvSpO7qlxcpPIA9EBXDjpJqZDjQXjgRPrgZWvuJ8nMll6LQifLqNd5yMxcplx3QvAlD1ATBo7GftMBFoOYBfRTrcAsc3rt1xny4hADdk+nRpH46pW8QDYebI/txTnyk1YebAAL/22FxVeCmzVW9jFJNRUCCMvHdAnCy7gts824ZtNJwCwHjZmWRqwwvTvaSqs1azMHHEjRgQ3TYVz23E3Vg1TVQVeXXIAczecqH0MbjJvrBp2Y8orqodlxFQOlMlumDUVg3O2jJw/Br6mJ39ZMzgFpbLYpPw9DhdNSyC5s3QTNpeJY7G4ESOqlhH5+uqTvi07RqzQimJEz9UzZqRMRYzc9AnwwFpAo4XNsY1ajlP2pBLcNEKFUQGH6ykmXLKMCMlsT8/fjRs++gez/naTZSIsb+f/6tdVORCc3qZ8//tkyfV1xw/sGienSW/2X4hBElBL7XWkV5/kE1EGJgRDOAuWdFiJT9rtYWnUZfksUBuSNSrglpGDS4C3mgPr31FOF85PQdR2GAkA+NnWv8bFHdm7HW1e+NN9ZpoaNTb4dLpHcBogvg2gcbLmUcxIkCH31RkiXT/X6FynAQ7LiNoNu+Y03ll3dBMFiRyrncfWEyo3pnqgt7Kbj7H6vEIICOmtW7Ol9ShuVArLiIdFwpzndyNGhADWsuq6iRHm1uGx/WQd9o2bMdscYmT23/s8by2f52TOdhYP275k3U4BVwG3+jXocmuoK2NWcdMAykBpgQ43MqEr3IR5u3j8udRQcVjnmjnEiMIyIrcilJy+qAwjK7TghNTe+saMqFlG9GGiqBc672o0cHLTOF5XFCrr/TiESaQsaLxdMnOPCiXyf9oms0LmbGauQUFk7vmF3dDfa38RDTT9yBnH8eXcybzTrUC74a7zC9lYzplUakXPHGIkxx6PCkhWqY7HvwL+eJRVa363DbDiRQBSnE7Axcjy59n/Va8qpwuWBiGFfuSnODf8S7xrva3GxbXmTsNsteOJn3fXOJ+CmlLnne8bYfEse1HrLEbMotALJCRG/IVcjDiXngdce+MI8DbXSoUAU7lr3mSVXVWi0eMjjPjvPT3F988Na4fRvVi/nXFztuEDx1PbkfwyfLTyiNJyUUd0NrZevaVUEaRocIiRo7IA2iP5cmuIPGbEQ8uIsxCQB3vKLAahHLsg1FWMaDkeBljh4tVRQ62q6fD3oDOwC6kRFmzwNF7H+QmnVGZNsFmAJU+wi965o9J2pvao27LN5W4sIyo36MaOY0aeaePYXpvJabtFywibVyE45eKHt3nePViGBVpwsmwaj90flmp1N5VOCoy02YUAVic3jRAzUun0ezrECMdxmDW6G14Y3h6905XndV/LVti/vJ79Zj+PZUHTix5mv6dQJdhuBbbP8Wx7fMm5o8D3twPH10jTqi4A+xyF8fo/Cwx5k2Vb9RwHDH9XfTkRDhdB6VnW9HPbV2wfyOsCmRxWN4cYOWGNhw1asUOzyD/vs/+bPgZ4XsxgulBpxv8t3o9PVh+9yI2uJxGJ0uvqUsmNJTwUCTWADGEoSR+KKnlvGhl8LIuBacKxY6zEE/cTX8O54BxHJozX2TKiNm8AcPM4TngduRgJkWW9xLcBBjwH/PGYZ8vjOJYVAbBmfOlXAWGxim7D4UYdXhjeHpuPn8ddfZth1cECzNt+CgDw/t+H8f7fkg8zMkSH+/p55koy2GTuAtlFxrnwlw5WHCkoA+B4qjK7ESZgDbASI42qqXzsu3WzjISIlhGnpydnMaMLEQVRCMzoVrYO2LwduOIh9fWrLKN6zG8Iad0fEVs/ByoBI2fBaXeZJe5wvlmWnmXxDIAyfbnsrGQZaTdcWSAPQBVvQCjndGGxVKo3x1O7kAmVhzVadjG1VrHfKzwOvLOQc9QBaeLo03K2uAo8z7Pfznl7Ks4BjZq5rk8Np/VYeSlmRA8bLDYeFpsd+rp2p1OzAAEKMSKIUJfUXucCaMI+qbrAbqQch5HdGwOAKPAF/qt9GzgN4M+npN9w94/s95DXLNm3EOj7QN22xRtYqljWVkpXV3fvihdZfMGR5cB/drLMqq8Gi4IBjXsyN15N5wcARLJieCg9zbLX1KguBta/6zifOJyyMTFXhEg0huz4kV8vL2SjUXhTAKwC9Ff/sDT38Vc1R4jexw06nZFf09/vBJhKgDELJMuIXhIfpho6TXPxbYCi44ji2HFvsfE4U1yFxjEqxfec8cQyIogRrcpt31qtGG8gIMuIv3BXGO2m2Sxg0NPCM/LKpYeWAJ/1YyZfJyZc3QJf3tsbEUYdMtsn4rp2iS7zAMDxc9INIL+0Gle9uQrvr3CTl+7AaFd3V4RorBilWYfbtavRnMvFbuNEdNrncDHY7U5uGumE+S3rDPq+vhIfrFTxtVuqgG9HAqtfU053K0bYjdLVMuJkZbJWi/syDNWYVDgDWPYMcNrJ1w0AxafYU6zTMgwOi4hwczPAilMXPHQ/VTiLEYclobpUCiwVxiBYRkKigf7PKb5WBpWCXebKugcKN0qXXguxEw7xxTsLQYd7ITk6BBwHVFvsUp8NZ0uMJ0GsxacUby3QQiMLYAVUfteacCtGpIuvMptGtg+d3adCZVre5pIlFCuLH4k0ym6MhYcU8/H7f0dpnqxWzKktLsGMPuW3ycDn17pmlJWcBg4tld6vf4+N/Zxj/I2aAwmu1xhVIpJrn6c4B1j1fwAAPjIZ1Tzbf3m8k5VYLsaPrUa4QQudU7Xps8X1CIS/WOQC3+R4vetb6fyUHV9CGYRP9Pcyi2a7G6TvxrMicdGQrovbsuvoSndnGbHbXKsyhzdsywiJEX8h99PJgxOFpzNPg4jkyxAq7AEsKDHrR9WUrzCDDl/f1xsvDHe9oCzceQYf/H0EdjuPORtO4PSFKnVRIMPgRozEayvwruEzvKX/Aq/o5iCcMyHzws/Ahg+BN5qytE0BmWvl2QUsmlw18O/EBuD4aslcLGCSXRAsrm6aUmfLiIqLhXeYU9tpZDfBgv1AeSHw+QBgIytahFmdgMVTWIllGRq9w6zsKJBlhPniLSOCr/37W5VWs+IcmRk4BLj2aZbJ5aCMV3maslTUXYzIK+IKFgJBPDpblRwXMINOIzaO25pdxFx+ztvjkRhRZnxpYAenY/s4RMueMLPPqbvbVBHcUXKhBYg9jmx2XllnRCd/QnSyHEQ3lj53EhAVssDacZ1kIka40IfEwKoLAwcemrOyVgW8DTi4VN395wuEbC1nYb//N+X7A79LQrhJH2DyNvWnajX0IUqLRi2FIDmZYMx1FiNFspTXs7vAcZzYZE7gTCDEiJrrz26TxYy4WkYWho4CHlgNpHaTvuOoWNsy0oYJVzHr9La6xvW5i/VQy5CJYC0PFPcioahfA8ioITESCBRixHHA8h7GbLgL/PzxDmDRQ+y/G8b1a45P7uqB54a1E6eVmax4/+/DWH2oQJHe6q6aJM/zCLWrXwC0MmHVMUx2wV7xIsvOOCnr1Cu7SfJu0m4BuGaQCOpe/nQqM0sKbhqXkvgqMSO8wxTfhZNd9PL2AEf+Ym6Qv54Hds9zPzat48KoE8SIBWeLq8Q4hDohxCSEOy4YJQ5h5JwaWXxSVkE1lJnZoxqLH5dBRYyYK+sUKHxOIwU82+w8cisdlwdRjDj9BrJ4HcGk/Mj3O/HRqiOuAbPC71SXQDmni3woZxYtIyEa9v1jhR6kTxc5rBAJ7ZTTG6XjfLkJfV77G8/+yoSwhuMcUawOnHsnhcVLVk4nMTKsXTQ+0X+I/0vZgB4GWcyPEA8UFosSI3OjipU42w5j/397hBUzlNflsFQzAS+UW3e2nh5aBvz+H+DwXzVtvXucH1iEjKyB05l4MJWyYx8AWg9yDXysDbkbo+e9ys9aD1a8taT3F1+7WEYUwdDsvBjcUWl58Vj8ewM1McLblQ8LDoTraIjecTw1krnEBWtbdTF6NWPH1vYTNVjK5PcPwfpxcImy+Jzaw22Ew3Umt/YJYyQxEqQ4zHIA6u+nc2eecxQQcknFk6HVcBjeJQUPXNMSP0xUdsNclHVW0Q7+VJH6Tcxi4xHGVat+JqeRtZY0NUsVSqosuOuLzaiuwa+qKEsOSCeWEAQHKG64ztk087bl4PmFe2BzXg4Au+PpYKpeVt/jbJayVsbCGnz6wtOF48QO1VhhtfPIK60G1r4NbPjA/XcFhAtbi/7sf/Y6ddN9cY4kuoQLieyiX14Py8hTlgfwtXUIbquaJk7LOlWM/GpmJTFVMSGhcQjC81pBMEk33MaNJNfGR6uOurqdqkuBA38Ab7cA/nrB7VgAuAjPUJjFAFajIEY8qS4suB5aDlROD4nCvO2nFC3cNc4xFM7vw2IlMeJUVr5p8TYM127GPRc+Qb/9rnWFTPpo5GmdslEGvya5PmwmlC19ScrEWvIEE/ALH2SZVDObsH0IMIHy42gWL7biJddttlldzxlAeSMzlQKLHgG+u5UJn5OOekDpVwMDpkFhFWp5neuyakN+g0tTXmcwYBoTOAOeB659BqXXvSl+5CJG5DhceOP6pSsmn/bULXqx8Lx6ULilCmItI9m1Xbi2GXUOy6O81INwP7CZ0asJ+86h/DKUuMsWkrtfeBsLEP7pLuArWbFNNXER5njYkItKh8WxIaT3khjxJ/f+AfR5ELjmSWmacENxelLwCjardFM4uUm1CqVzkNSK/Xk4m1+Ar/Rv4xbNOmSfk07y15cewE2fbEBxpRnVVpuiZLM7uNp8kZYqzN1wAhuPKW9ePM+7r00CwCY81dmt0o1WFn+SEsYuCGXVFpisNjyzYA++35KD7FzXug42jYogzNsDU3EdO7AKJ7fDMpIYxi7ieaezgdWvsptFxXn2/7SbTsLC9nQYydwH5fnsSdmZCyel7RQudmHSxVs9ZqRmMXLCnoxXrGORzUs3yrWHC1HBs+VXlDELiMbKjoUCoyMQtURyp6RGO+1DocR3rCPF8/xRYN7dTGDt+NbtWAC4pC6Hohpax0XTwDERXmfLyIUTzLrFaYCOI2ULZftM6yQ2XDpfy10NABN+ghjZ9hXwy33STV8W66I3ud6oNp61Y3NRhPieBwdEpwH3LWYNOgFEnlqNuZ+9jeLiIiDL0dDvzA6WSWUzsX1YckYZRyTEVBxcAnx6FVB4GPj2JuCDrq4ZU8430KzvgaMrmOipusDSmlO7seJlj+8Fbv4vcOc8oElPeIz8nJUHLxujWKD0mF+Ym3HAc6iMaCp+XKMYKTkF2O1okRCBWaO7oXk8cyX63TJSXaL+QChmjUmxR+fLTfh4NbNuGXWOW25CexbDkdyZWUMdFrh4bTVaOLZp5yk31hH5uWy3Kev4CK4+4YFFHh8iWPnk00TLSOA795IY8SfNrwGGvSWZ4gEpZuTmz4ChbwExdcw4qAsLHwTebsnMd3OGAHOGShfOsnxg9etIhtJUW22xo/3J7zFQuwvvGT5D9rlylFRasHRPLj5fdxy7TxXj63+yUW2xIcxZjKgFRtXCueISRVaPwFPz/8WHr0xC5ZfD2QnmJEa0tmrYeceNw+ECsJokF0ybWGaKXLY3D4Pel2JqCs653iRsaqeBtQplx7e6TldD+A0d/1Mj2LhO5pyU5vnzaWYh+dLNE6Zwk4hKqbm3UekZKU5GCLSUWUbKeHcBrO6fHKvgWkV07eFCVIJtT1UFEyNaxwXrXGg6m0l28zXJhGNTQ5lDqHBS0zTBYgdIgbF75gOzr5TcKALOlhHODE4npfYCQF5+HvC/W2p2nwGSAG/SR5mKGc6eEnVOGTkuSVwRSUBIjPJ7ghg58DvLhBEaAZacQk1cQCRyeGkMFmMjJmTD44GrHkdpnykAgDf0X6I6q4YqvLt/AI6tkt5XFzP31093seJ1y55hrtDKc65xIe6Cebd+zv63GiiJ6+gmQNc7gLZDatwutxjdxIw066fIZAIAs8wa6xIzIsdmFsXXyO6N8eSgtgDgvi+Sr1Bz0QBSM7qIJHE/PrNgD/aeYeeQmPFjCAMeywImrGIHnSB6q4vRLoVlWrm1/smtHnarmNXGxuW4ngsuPXlmmJCdJ4/7EVzMDaBzL4mRQBASzRSpPkx2Q4kF+j5Y/0qtauydD4BXugmEzqU/3wOsfRPG3x92+Zq8A++O7Vuw97/j8Nr3Uh+FP/7NRYXJhnBnN01ojMdDLCtTD2ycv+M0ntT8gLDT/7ALasFBxefrbJ1RLuTtO8RIZaV08qZHs0P7WGGFohiXXc1NIyvM9Jn9JiCpEwAgttC9q0uBGDPCxtMihp3s67NkvV+cgwPl8Lx0EQmLVwSkqswsBXiquGkq1GoZVJcoL1hOVEG6MRwvLEdRhRn/ni4Wp5sq2f7VOVK5i8NZXQT5zXdMX+nJto3VcUFOaCelmsuLXwlPbwvGAwX7XG+YTibmUJjEmBEtz0zUN5d+DxxbWbP7DJDiWqKc3CMOk7VzVoZBECcjPmTWyt7j2U1Z/F6c63EuBFiqVJrdYZdcsiV8uEKMVBri8Nayg7hy5kpsOnYeB9pMwhF7Y4RxJsRsr8G1t+pVFs8kwNuBQtmxdk5Wd+PAYuWNRt5Y0hmtEbi+nm0r1LjlcyC6KTD6O3YDdARqImOSy6yCFTQqRId8uBEjQjaI7LhLi2WW3aOF5Z7FaF0sooXJTQmCaCmO6+8DUjaQaBkBmFAQ3CSC4K0uQYt4Zj077i5IWy7WLZXKRpRCfJLoyjWyFO37lgCJjpgpeWanIArJMhKk6EOBu35mZkrnoDBZMKLXkHfFFILhhMDIkxvQo2kMAOCDO7oBcPQCcfBayTPoV/IHNoQ8hhWGp9AYhcg+V4EXvl+DJK5YuR5jpGvAXy2EcmY8qZuHGzSbnD6RXVg2fQz8+xN7nTEZuPEjPGF5WHRJ2A7/heKtPyrMl6nhUCXMkWWDIW/iENLxlOUBREASK2+YR8MWz05aDeoYVOwUwCoIIU6ekliDGIC5QroYhMUBie2BF1WKpjkXxhNqYMguLqpjdi7Y5US1rMjUde+uxYw/9oHnIbpphJgRnZ2NsTRCECOnxYDU1kmR2PPyICRqSvG8zuFeaNJTejoul90ETaXKUvTOwdtOLqUQmHG+ypHtYreia1oMmnB1LJltdghrQ4Ryegrr3+TslolyFNRCz3uBMT+zG4b8nAyLc01vFVwhghhx/B4nDG2w3y5ZOi/wkdhhb43jdhZ8uaysBWavOYazJdWY8cc+nC4xY7GNdUsOKVcpoX/NU8rgw7QrpMJau3+UpsvcZ6guZv1j3mkL/PWi9DvIrT1XPQ5c9yJwx/esroi3aJYBPL4HaD+Cvb97ATB+BdD8apdZBTESYdQhpUm667L0YVJshSzbqn1KFKJD9SiutGC7lypL1wnBMuIcFO3AEi6J31BZ/ROFGJEjCNyqYtH1lF3oTozIxLq5QhlbJrjlBfe4zsiq4aZfJc0z/D0gvi0rKyFmc16ilpFPPvkE6enpCAkJQd++fbF1a93M2T/99BM4jsPIkSPrs9rLixbXKg8QAflTWHii6+f14YLMDL7sWeBHWcdffTi+ub8P/nzsatzUrTE6NY6CBdLJE89JlovWmjP4OeFrJGpKMbtogut69OFSMGcdSeYuYLLuN3xs+EgxPQxuAqrCYoEeY1GIGDFYU/vXNMQsfQhRsk6fOrsJLRPYSa3hgGcGt8XoTpHSchPa4pHID/CLSr+IquiWHm2DKEYcpbITbUyEJHBuetQ4l0YXnlZ1oZJZVasHrn6Cibu7FwCTt7sGEQqWEZmg1UPFjy1mTag/xcktIwDwW9ZZxXRrVTlgs0LHM0GliW/NxiUzmQNAZIgek8NWoLnGMa3TrcriXiI8q40j4BzP4vSUZuSsyC13iDm7BROSjuIKzX7VbXFBsIQJ47jjR6DjLeAHPIcHvt2OFxbtVcweGaKSuhrtJEacf4fcf5koE8TI6O+AzBk4N2IO9mtaibNdQARKEYHrzO+ie/VneLZa6qB6MK8MT/yyG0vtToGefR8CwAHdxgDXPsOEQ3JnlvEy9jfJKravBsvbb/9hImTjh1IZgHY3SEHg3e9hcWw1uQe9QaNmQFof1Y+EDD6DToPvH7wG5ZP2sONfIDSWNSIFFNczvVaDge3ZdXL5Ppn49zWCBUJWZFJOkRDkDeUx5bYwmyAOf5+MQdvGoQlXiOPn3Lhp5OeLudJJjDhZRtSux3Etgclbge5jLu3U3nnz5mHq1KmYPn06du7cia5du2Lw4MEoKKi54dOJEyfw5JNP4uqrXVUxIUMuRuQ+bs5L1QUrzytvBBEJiAzRo30Kq7nw2sjO0NVgEWhcthv/7VOIaE4lBkEf6uIL9oRoSCdfrMxVpMAQKTZKK1dLYxWwVOHTu3vi1p5N8M8z1+Hh/Ol48+gIqZaIIUJM+z1pZ/t5oa0fAKAs0lMx4hAD8W0AAJrC/Xh7eBoSnC1HAv8XB+z/XXqf7+jWmdheGbQw4AXguVygVSZ7KoxrpVyOc3VQAEd4FcuakPVhCMfBe3ZioOlthSm8WiVmhC2eWRNs1eWKmJNGsfGStaBQ6T5rr2E35OzW41ijyJAo1WXj2GrptbN7QyXYttrmOP55O0bsfVR5/J3cBPzvZha46YwQayRYRtoNA26bg0KzEX/td715RYWoxD0JWQiAZLlSjLeC7QchHiO+DXDVFPTq3AkvTLxbnK19unBuc7iAKAji8JH+0vF2hG+CvfZ0adnXPAU8dwYYOZsdZ9e9ADz0D3D1VBbALLSWkFtDnJHX4vnXEWMTmcyaBD60QeojE0AEy4hBp4FBp0FEQlOpGjDAtlOw2jjFGA1xpPku2XPW964acyWweCqw2lH9Wn69lnFOJkYiZGLE7q4BphAzUlGIyLytGKjZifxSk3pTU4VlxKnVgyhGXGudqKK7hMXIe++9h4kTJ2LcuHHo0KEDPvvsM4SFheHrr792+x2bzYYxY8ZgxowZaNHCi2bAyxGFZUQW6OqulHaM5KuvTwApNDqmrB0xGV3TYjCmm5tqsQ66l69VThCE0kWKkcEplejahJ2UjeBOjITjXDk7cQr4GPcLM1egTfk2vDOyDVJjQpWF4QBUwig2NBtreRbvWm7Fi5ZxAIBzobUfo+XaaJzlY5Ef2VESEAksmA65u3Hb6utwe+S/7hfw8z2ApQr8sdXA0qfZNIfrQESjUaZ+y1PCAYUYmdX8M0yzjMdPtgHu16kPRQkXg2N8YxRppDgTd2IkIZYJFrssG8fOc0iIiWRCAwAWTGA1WRw05VlsyJHoDDZB1TIC5Q3FuW+N48JodVjoPrSOxBNDO7rfrjlDWEDnL/e5fiZaRpRuGnnApBxVy4j8NwiJZr/31U9AYWk6vQ0Az540ZeIlvLF0Q+3T1HVf9Ggag6eHtBOLXQHAR9abHV9OYMGtBjc+R0Dp/9caWHaOgPza4EzTK1gcTXIn9/P4EZPj91CU+JdnMiV3lcTI+WOK717bNgExYXrkl5qw4aiHPaE8ZcEEYPtXUhdoN/v4tF32u8j0R1GFG3eI03WziYG5aPadVYmpk8eM8DalC9RFjKif2yKCVffXCTXHE/kBj8SI2WzGjh07kJmZKS1Ao0FmZiY2bXL2+Uu88sorSExMxPjx4+u0HpPJhNLSUsVf0CC/mMjFiLssG3lVSTcqvUaqitkJ9mmGGFsij6FQRZ5WCIiZCTCEK0s3C3S/hxX5GfmpNE1Fsb91XSQ+H9sLQA2WEWMECsrYibbK3r2GMa5gT8trZrIT06kCaLbs7Uk+GR/ZbkG5IwbliC25VndTkS4R15hmYdkV30kTY2VPmHYLoqpqbg63cu4r4P43UurrI0S7u8P5CVYniZHDujb40TYQHZqlYKfdyYIioA8VrUHFOumGaVe5DIQbtIiKctwMzBViX5oqGJAYFQr0up99VnkO+DKTVau1mpBgYRfpo7zDfG10YxkplmUaVZ5nT5y/Pgj8MUW82H4X8xAGm95As1GvokPjOPXlyDlXB8uIsEqzijsLspgROfJOtYLwHPAC8GwO0NyRLXTGkbId3VhZNE2jBfo9BsS3RfOrR2NE11Q8el0rLJrUD8M7p+CpwSzm4O4rpPN7ub033ol9GdV3LEBBWbVYql4VeRxRag+g2ZXSe3nJ8R5ORcfqUzfEh8gtIyJxrZg70BgFXP+KdPwXKcWIUafFjV3Z8bZoV/0bMtZK8SmlVRlg12WHq+w0L51TJ03ScV8qa11wrtyNGHGyDnaIZvPd/t9N+HL9ceW8zlYM+XeFWBZbHS0j8mDsnf+reV4f45EYOXfuHGw2G5KSkhTTk5KSkJenrqr++ecffPXVV/jiiy/qvJ6ZM2ciOjpa/EtLS6v9S5cL8mA5uVp2LmUtkNKV9bbpca+i3oQLoY2AiatYZPVYmX+5qoh16OTtUgU/kxsh4A4htkXFbQCAFfJ6LAvodpf0JNf9btf5io4jKSoErRIjarCMRIiWkRU2qfbBAttVYm8LBRs+UC2Nv+IoW37bJNen1QN5ZbA16esyXTFUrhGs0CEiRPbkUdtTiBMDz8xWTkiuRYzIxQ6nUcSKCK6r0b3TMC/9FeTYE3AsdYTi6xZNiChGSg2SC3DmLZ1dVtU0Llx000SY8mH/8U4AzKKUGGUEUrsDXUazma3VLIBywwfQwI5SPhRZRY6LoDs3jXOPoDPbWZDyjjmi2bnCbsQhvikMOp1roLcadgvwXkdlbQ3hWHay0Kiav+HGMtLpVnZjH/y6NE2jYdsmxF2ccTQsVHsguP4VYPJWaMMa4aM7u2PqoLbolhaDT8b0QEZLdiNLjw/HnX2kp+yPz7ZBu09Oo89rK/F/S/bjbHEVvtl4QlEdGYDynG+WAVzztPS+RX8gcwYw7B3ghlnADY7utwNfUpb9bwCIYkRuGYltDjy8EZiyBwiPk47/ikKXxo/DOjPBuO5IIXh3rpCL5bRKbGRMGjDhb7xpG4NhptdRrYtCGR+K3SapQqy8JYVw7XLh6ieYZTuZWUfbRkmi5dUlB5BbIs+gcXJjyjPVnC0j2lquSdc8LYntA7/XPK+P8Wk2TVlZGe655x588cUXiI+Pr/0LDqZNm4aSkhLx79SpmvP3LyuEGgyA8oLd9U4WuNZ7ovIJ2hgF3DYXuPHDmiPho5qwjptxLdlF6mmHmZy3SxX9zjtSAqs9tEQJPQ/0ocDt3wK9xgOjvpI+j5SVbr5vCQvEu/7/lE+cgGi6n/9QBib0ZDexQj4KVZxM5BgjUeiwjBQjEvOs/VHMhyNjwgcomLCTXbicKXcVyt/vYk8QvZs3QrtkdqPq4Iib+WJ9Nl7OZVkNpXwoHjRPgZnXYor5EfH7B8vZ7xThfPMSKqh6SmQKkFSDKwJQ3vx0oYr4EuFJP1SvRXhCc1xjnoUFqU8ovn602C5WpL0QIgn8O/s0FTOpBFKjQxASxvZHG/N+aM8xN56ZM0pVJG/5nN3kAFY0y5Gim8MnIet0CbafKMK6k3X0Rcv7jzguqJUOcanXalwb1rmj9DSw7h3pvWgZUbo63FpG1GJGdAaWaaKSkirGdQlxP9H1f3CaeUtnfHO/a4Dn4n9zceunGzH99334eNVR5YdyN03TK2Fp1AIlQz9Bcbs7sAldgKumAH0mMvHUcxzwaBZw1dR6j9FXqFpGABafIzy9h0RJ1mK5q6bqAro3DkeoXotz5WYcypc9yNgsNXe29QS15pkxTWGPaY5PLcNRigisvn4pBpnewuFSdhxVW2yK4o3/d5Mbt1irgcAL+ez3AhDHleGKJB7Jju7Fqw/KssecLSNlMvHtacxIQhsWg8RpgLx/WVHFAOGRGImPj4dWq0V+vtIUn5+fj+Rk1y6Nx44dw4kTJzBixAjodDrodDp8++23+P3336HT6XDs2DGX7wCA0WhEVFSU4i+oEPyQnW+XpjXpzQ6a4e8AkbIIbrn5edCrrLCaGtFOgY1hsa4HqnCC18UyojUCSZ1ZOWfRMhIGdLgJuOE9ZZS5XHQkdQQGPMdEl3P0vkMMxYQZ0CmG3TSX2K7AV5ZB0jyGCBSUStkWz1gnYpjxW6Q2a4WmaU3ZvrviEcVikeNIY3a4us7zkSisZod+p9RoLHj4Snw5thcev76N+JX/lXbDRPNU3GKegeX2PuhomoNF9qvwnXUgyvhQ0a8faXS6SY78lP1NO8MCT+vA153+x54A69IaQAhidZq32iKJkSaNQgFwOFFsgzyuodSqEy0ju+JvZGW/HV1/b+rWGMdeHybOG2LQIjTC9byzaJysXx1vhnOWzma+AwrKTLj1s00YP6/mzs+iJVBuai5nbivB0qXXaVSqkbmHV1SkdKRHOrlpPLKM1IQ8yByon6tURpysAZwgjgvLTDhbwo75D1YewX/XHpOe/h1uGh4cLI17Y8Yf+9B1YSN0y7oRd369EyfktSo4jlkbPNiX/kKI4XGb+iogHP/Cg9O5o8CsrjD+fCf6NGf7Ysis9dh/tpTdkL8aBHzQjbkBzx9T1uSQc/gvVsG2JgQxIk8miEhWxB+lNWmKXMTh9IUqWG120SrCccDulwZhgJuu6QCYtcrh8uEqzuFH7YtYF/Y00rh8Ra0Slz5dcoSU+brGjADMzd70ShajI5R+CAAeiRGDwYCePXti5cqV4jS73Y6VK1ciIyPDZf527dphz549yMrKEv9uvPFGDBgwAFlZWcHlfvGEhzYAk7axyP9bvmCpnXI/tDw9Th6YF5nMSs73fYi97ypL4ZXHnwjIn6oA6QSvS4fV+DbAw/+wcs6CRcZd9k+E0q0n4uy3Pr2VldiuLBJ9nxf4SFzgZSZ2YwR2CL072IoQG+kU35E5g5mkBdF2zHG8xrXE6uHrcL3pbQg30E6NoxFu1CGzQxJaJypvWCvsvXCUb4IWCeFi7ZUXrPejh+m/OAO2P8OdxUhUKnNHGSPY7zZVKka13NbLZReYeD32WBrX7GKT4/Cb87oQbDp2HmXVFhSWmVDlECNhBkGMAEv35qNSVgStijdix0m270JCw1kZ8v7PiJ/La24YtRrERDuVQgegdc60Co1RWnT6PYb3MUZ8a4EOJr1sOc4CWLh5y2ONHNk/lXbBMuLZzbP0vKyGiZsA1gqzuhhRjRmpCef0+4uwjAAQux8DwLfj+6BbWozLPDP/PChlAjnitQ7a0/DayrP4brMyq6bO3V8DjFvLiDNChk3ubvZ/6RMsW+jYKkXzvPk7TgObZ7NWACU5rNv3Rz1YOX2XlVcA88awz9wFcVYVS1WEO42Spms04oMAALRMiEBUiA5VFhveXHYQpVXsOIs06hAdVodjSwh+PncIXNFxGOxVeFL3C1YdLJDiYWrKfKkoYG6cusaMCNzyOfDkUaB13R6gfIHHbpqpU6fiiy++wDfffIMDBw7g4YcfRkVFBcaNY5kIY8eOxbRprOFWSEgIOnXqpPiLiYlBZGQkOnXqBIPBMx970BASxcxnANDldtcnbPnJoNZ2/PpXgHsXSz5iQD0iX174CGDVWd9oWmtZawBK4XHFQ8DNn7O+OwLySrJONwKR1oNZ8Z20KyTryZKpwPe3ihcFPiwWpbJ+K1n5FmzLZjfUGMfJPTYjXblcnYEFWLZzPOkfdYiR8AR0bt8ORZCe+FsnSWMTbuLO9GjKRNvwLikAOEVROBc3jTNRqSi++y+MMszGy5Z78ZvtSmT1fVf8OIdPxJlSDwoOOZ4My2x63PnFZnR++S/0fu1vHM5nN92ESCP6NI8Vn7ArZAXNqmDAJkcPIHfjbuUQZCO7N0ZcI1eBlAiVm1uT3tLrrnfi7iuVgbbnDTIrmdxixmklF165a2mACodlxOBUsh1dRjOrjhvCig9Lpnmx6JlzzIi66T7QlpHEqBC8NaoLZo/pgfgII/q2UBepYjO91oOwQns13rSOxtyNJ1zm23vGTa2bBoZqzIgajXuw/2d2sriR42vEj+7oFodbejBLW05RBbBNluH598vs/4n1rjEX54+xmjm8XbV/FwBg909snsQOwLC3gWZXsVgcSE3wtBoOoQYt3hzF4j7mbjwhWnHrLHKFZAAZI7RbEIZqvPPXITbBMX4bLxPpYXFS0H3xKVmdkTreY6Mbexzz5m08FiOjR4/GO++8g5deegndunVDVlYWli1bJga15uTkIDe3jg3GiPrRqJnkyknv5/q5zsiqHMpN+fow1/kEy4g846G6jhcv+UU4JBroOloZrBiRyNxKj+12vwxDGPDIZmD8cmXthjM7WNVIAOOu74mEGGm5t83ZC7PNjrTYUKx+oj++vb8Pbuvp5gaQLKTKOkza4QmIjzDi+WFsXb2aNZLiH8D6lLx2cyc8eE0LRMluShOubo6B7RLx0DUt8cXYXhjeWXI7ubhpVIhp1RcLnhuDjB5d8JhlMjbqpYyHMoQir8SDUszxLH24VKUHzdWt49E6KRIxYQbMf/hKhOq1qOSlJ+1KGEU3TYRR/eL484MZmP9QBq5pkwAupiksUM5ntKuIX3n9k/g2eDyzDRY+ciWmj2BPsWchs4zJA7RDG0nuE5UsrAqbLGYEYIGkca2BIW8A9y3GueY3qW6D3l4NLH6cldl3YxmpdGMZkR8PdcJFjFy8tff23mliQOa9zkLbwQ5Hi3mrMQZTrf/BGjeZZbtPXyJixFZHy0hjR9B67m5lrRoAmqrzuLk7O742HzihrL1SIYu5yHW6JgkWYcBVjBSfYk0HlzksiN3uYtbAcUtYLA4Ak5UJ2xDH2Id0SkZqdAgsNh6rDjKRrRqLpIastYO4XbCho+YkTl+oYoGsDsvIAts1OGZPgRl6rOj4ppRtVJwjtQCowTKSdaoY9369FYfz6+CW9wMePgYwJk+ejMmTJ6t+tmbNmhq/O3fu3PqsknDmwfXMgpHsmgmhinASyxHESFof1kgsZxNwfLXrfGqouX2cqcvYBPdT59uVzb8cNOp4PTqfKAMcBTcFq8TIbo3RKNyAa9rUMA7nVFnHU8eEq5ujQ2qUWHZZzpi+LK7kQF4Z1h1mF7B2yVH46j729N8Z0bi+QxJarjiM0ioLEqPqaAaFlLnz1t/ZeMTxNTP0yCth6Zsa546xarTKBK6aip+PNwaKWF2YxjHMZfP8cEnQNY8Pxx//uQqVH0tipFomTNxZRmLDDYgNdzyNRyTg0/bfYHXWESw0Tnc/ph5jgUN/sngljRYhGqB700a4UMkuiMds8RAdVHLLSGgjSSSXu5Z3r7CzMYpi5NavmMBwxDwYI2Lcj2nnN0CbIVJnVZeYES8FNTq7IN1U5KwvqTGh+HvqtZi3LQdfrJdqs+w+XYwKkxU7cy6gzCn+5apW8fi/kZ0w4J01yDpVjN93nxVTXyvNVoQZ6nXZ9yl1dtPEt2GVni0VUnM/gYpCNIttjziU4AHdYreL2LFxBXo2vUKaIA+GdRYjvz0i1SiKTGWJBE4IlhGjo7oqx3HIaBmPBTtPY9k+ZuGNCq3jPtfqWQsFk1JEZsbkYVtRW2zNLsJNDqtHBUIw0vx/CIEJhevDcKRjGvQF+1navLw3jRtunr0BPA8cP1eO9U8HPtW74R2VRN0IjalbY7rxfzNfZ7vhrp8JQa3pV4tR3PhulFRH5LZvgPnjXHuHAK5PhBdLl9EsgKv5tcAfjzHLyOCZQHg8QltfA+xnlVLn3NcbrRIj0DimhuqrAontWRaGkC3kEFAcx6Ffq5qzu27qmiqKETWmyoJd68qtPZvgf5tPKtqdF/LRMNvsOFdhQmJkzcLmxLkK6LQcmmROx9pPNgAoxqT+LTFI5iuX06RRKPbJY0Zkxc3qYtEBgE5de+G9XcCf9r4YqtmirIopEBLFnhSdaBrLhMaBqlgpxlXR5yVWch+qWkbYGA06mUiTBV+GRroW59tvb4Z92na4jV8umeYBlWwadcuIx8ifZLvfo8yG8xKtEiPw/PAO6J0ei2V787D+6DkUlpnw+brj+G4zy34Y3jkFS/Ywi3S4UYv0uDBktIjDpuPnMeWnXcgurIDZZsMnq4/h0etaYaqj221DwaxW9EwNjRZI7Qac3MD+5JScQeO8/Zht+Bh9NQdVvw4Aufv+YS/K8ljBvvOyoM2CfSxmreQ08OsDjgaEHDD+L/ZAp5IS7WwZAYArW8Zhwc7T4rleZ8sIAITKxEhKNyA3Cw9W/hfRukP4998HcFMSExomGFCGMLFHV3loKhoBTIzIe9O4QYiBPlXk547HbiAxcrmT1pv9qXHNU+yGLdSLANjNRhAjHUcCbQYDrzludoobu5fFiEbD4mMA4LY5rN+HI1C3S5sW6IevERMdiT/aJNTNggCwEzG2hVQMy7mceg3c0qMxSqst6NTYNYizvsRFGPHjxCvw1T/Z+CPvMQyt+A3zbPcD54DvN+cosnmcKTdZ0f+dNQCAQ68OQZ6j7kBKtHtRFuLkppH3oImooxi5rl0SPru7B9rFzAGO/wh0vq1O3wOAJo3YRTLHEgVRB8ktdNZqSSSoNBIstzu5aZzQhbr+Nlvs7fC5+QbcGr4a3DmHj10f5nITKfeWZUSjBW78mFWRrbHb8sUzqGMyBnVMxtwN2Xj5j/34YCW7ibZJisCbt3ZBn+axeHPZQTxwTUtwHIdvx/dB/7fX4ExxFd7/W8pq+nDVUfRKj63Zquhn6mwZAViNG0GIaHRA417Aqc3Az2OhBY++skVss7dBbw3b9kqEIgxVaM+dBM5mAf8bqezrIrBmJrBvkVSMsM1gtz11AMkyIu87M6BdIgw6jbhdbZPdVCJWI74Nc7VEJLFO7otYZ/U7dGuw6kgxVuSn4XoA1U4u1CJ9MhMjZ7Mky3VdA1gbANS1N5iJSGSBnvKnxt6OKrlCgKA+lKXwAlKWDsCKEPkKfSjQtK/4FBwbbsCiJ27AD48MqLsQEeh2F6vJMeB5VgK7jnAch3H9mqN3eh2zXOpIWmwYXr6xI0Y88Ap0j+/GnYNYKvZ/1x3D2WL3TyhnZNaUHScuiFVok6NrvtjIBUiVLJi11sBbGUM6pSC9SWPWTM1dWwIVQvRadEiJwmledtNrO0w2Q4x6LJODMpuTm8YZWcr4Hns6vteNxBzj3chFHPJ71FxLQ80yMmlAPXu09LgH6P+s3wqJ3dYrTQzeBoAHr2mJCKMO916Zjn0zBqNnM2Yx0ms1YkCnM8v3Bbb0t0B+aTWyThWLv4exNssIIAWxAsy9LAbLuxY722VvjTJHQ80H8DwAoDmXB6x61VWIpDkKHW79XBIiiR1YP6AaELJp5EIqNtyAYZ0ki+V9V6bXslEyRn4GjFkAPL6PBfnLgq87aU7gQgnLdjTxyoDTbDhi57LXAnvns9cJDcsCVhNkGSGUNEpnKV7yipVjf2NPHi2vAzZ9zKZ52J33YklwTt+tK1c9Dlz5aIOrOCkwrHMy+qTHYuuJIry17CBm3aEeiCiv3PjrrjPgeZbyKq9LoUalTIzk8FJ8Q10tIxfL7DE98Mj3wAeFNyM6Ngn3aTSsA/HKV1izt7NZbr9bXpsY6TQKC1f9g07Fq/GW9Q4kdByKmMJy5FSUYE+ze5G87Q02n8U16NY5ZuSXhzK8Ljx9RbhRh5k3d8bD3+9ETJheDHYFmIiWc3P3xvjvuuOKwlsAsMdPWTY8z7uMSSDrVDFGfqJ0tdTNMiITIy0HuC1FYNGE4HvbQMy3XQM7OBzlm6DQGM26aR9dwWZqdwNL7U3vB0Q3BU45ahJ1GsXcxJFuyhLIMFldLSMA8MzQdsgrrcbIbo0RF+HB9SsiQUqxDY8DHt/LXOVvNUciV4xbtSyGpUPLpoCsjM+ETTEYo70fz4QvRpS5gAV8t7+x1tV5+nznK8gyQrgSkaDMxAmPYzEn+lD2pGCIVD6dNHQaqBAB2M1DCDxd/G+uy01DQC5G5u9gRb2SokJqtRT1iJXShtfZpUZ8ifUVdx6SHh+OT+/ugfett+HlgmtY2fr41sDo/zFzew1N4KrdpfYK6Az4NXosrje/jfX2LhjYPgnJjoDivHILqzcDsPU44WwZCW+AQZ01MbRzCuaO640fJlyBUIP747tFQgTWPTUAv03qh/YpUXhqMHtSPpBbKsY6eJujBWW4UGHGusOF6PLyX277xQg1b+TUSYw0SgciHFaHVpnqwfRhcTgx8RBO8sk4zKfhKM+sBgfsysZ2hQPexPZrvmYutjaDgIT2LEj1li/rJEQAyTLiXLAtJToUPz2QgTv6NFX7Wt0JjWHxVY2YBUjD8cgNbYPuQ517vXH43paJ2w2zgQkrWe2QOhS40zSQIniX1hlIBJ4H1rLgqJo6iRIe0aVJNCKMrDrqifMVaKPSL0cogS8npRYXDQCkGaQKnKVgv1lMmL7+lqZ60DQ2DOEGLSrMNuSWVKFFgiyzxY2bhteFQIh61evcXywzWsZh/ZFzSIw0YljnZGzNZnVU8kqqgEGPMRN+kmsJbucKrCH6S++5rH/busVtJUeHIDk6BH8+djV4nseX64/jQqUFS/fk4ubuF1cXxZkzxVUYMms9wo06WG12VJhtmDIvCyO7u7qLclXckqF1EYUcx8rzF+ewh6IcWZPWgS8B698DRn2JlknRaBobhpwiyTJ2kG+Ka+Ao3R+RjAGz96HcZMWChzPQs1ksMGmzx9vszjLidcLjgQssoyr2njmwx6nH/BwvtqI6qTv2nSpBj6aN3FqmBBqIFiHLCOEhOoP7ImZEveA4Tiw25i7nX02MtFYRLS7LHvQqEBYP86hvxWmp0aG1XqC8CcdxSHFkP50tdqqp4i77ROYGrCnD4p4rmuG1mzthxePXguM4JDkEWm5JNbvKdrjJtdsxgAqn3jRGX99IGggcx4lVXR+ftxtra8gYqw+H88pgtfMoqbK47GNnckurXaYl1NWd0aQX0OkW9lqewn/1E8BzZ4CW10Gj4TD/4Qz89x4paFpuHYQxQqy7s+6wazNNgf1nS/HNxhNuuye7s4x4HVmlY2NqJ7cWMbPVjoe/24FRn27CD1tzVOeR489rQU2QGCGIBkAbRyXYI/nl+Pd0MUZ9ulFRyltNjGS0qEMQcetM4OljMHSWCoQlRvk33geQrDhnS5yehg3qwtYuywLQ1eCKigzRY0zfZmKpbWE9tRWSq3SyjPj8RtKAeH54BzHI9en5u1FSZcH/Np/EhqOuN+Sc85VYf6TugsVdV1p5yXQAKKu2KPvmOEiIrEcV0PSrgDvnAVP2unyUGBmCwR2T8datXfD6zZ1xqlFfbLKx9HR7R6mSdU0tB4Z9uB7Tf9+HhW7cTcK2+dwyct2LrDv7w5Il6PsJfTGmr+QGEmLBVh9iv9nbyw+pLsomE1YUM0IQhEjrRGblOFJQhhs/3oAdJy/g/xbvFz8vdFzk5YGnV9RFjKjgr3gROamOFORcZ8tIjW4aFi/iyZNbchRbT57jqft4YbnYrOzk+Qp8tvYYKkxWV8tIEImRVokR+G58XzSPD0d+qQnXvbMGLy7aizFfboHFpoxZuvad1bjnq63IOlVcp2UXuhEjxwqlDuTny024+q3V2HfWNfA0IaKeqahthwAx7qvf3t4rDXf1bYqU6BDcZ3ka23u/h/NdpPYVvLrRQ8F2lRgXQO6m8fExFB7PurMnSbV++rWKx4s3dEB6XBiubZOAHs2UtXfc1TeR92bi0DDUSPCcgQTRgBF65CzdI6VcCk9cPM8j33FzHdmdVdLUajiP4z76Orqa3nNF+sUO12NSYthN5v2/D+NQXhmqLTas2J8Ps3MXYAc2o5Ce6tmFUm4Z2X6iCNe9uxZP/MzKf4/8ZAPe+PMgPlh5RBQoAj5/qm1ghBq0YluE8xVSkPP2E9INt8psE2/S2+vYcO9cmXqfpSP5khj55+g5FFe61pUBgPj6WEY8ICbUABMMOBA7EPkmSdiXVKmPRx7ozLtRLCbRTROYYyhEr8WqJ/pj7rjeaBqrPJ9yiipx/9xtOO8kEuUxUzY37id/QwGsBNEA6JUeKwZ5Clht7CLx8Hc7xUZ4d/Zpir7N49AhNUp1OTUxZ1xvFJSakK5SBt/XpMqKsz303Q60iA/HyoMFeLF/HBQ5AX0eBDQ6FDQZDnxXDL2HFguh7kql2YYXf9sHAFixPx88z+OC4wb4595clyfhmlxBlysD2ydibEYz/G/zSXF/rDlcgBYJ4Zi78QT6yFKdy011q1jrzjIij4WSCxNnfF2qvlE4sxRcqLQoxlpUqS6i5LV/nAXUhQozPlx1RKxgGsggaCGrrkW8q9tz1cECvLJ4PyZe3QIhei1aJUYoxIjZZofFZq+9+q2PITFCEA2ACKMOjwxopfDxni2pgsVmF/tbAEDjmFB0TK1fVdgwgw7p8YE55eXF2bLPVSDbES/wzfZCpRjJmAQ0aoby3FIA6z2+QIbotWgWF4aT5ytxIFdyAzw9/1/xtV0le7qhBPH5E47j8MpNnTB5QCss35+PFxftxaJdZ/DrzjMoLDPhU0g9W3LOu9ZqsdjsyD5XgdaJEeL+KyxjFrwHr2mBH7fmICbMgJyiShwpYAJk/ZFCfLz6qMuy/EV0KLO8fLvpJJJksVMXKtTFyBmZW/F0MdsH1RYb9FoN7v5qi8LVFCjLiJxbezXB6QtViAnT470VUhGS37LO4resswCA468Pc6lAXGm2ITo0sGKE3DQE0UCYcHVzPDW4LeY9wCrFVlvs+Pd0sfj5nHG9ERMW2Dbf9aVrkxjVAm2nnB+So1mqqRC7UGtLeRX6qBQv+8VRmwWQnt5TokOg4YB2npTqvgxJjArBbT2boGlsGPJLTarB0vvOlrpYR2b8sQ+D3l+HRVlSYOe5cnZTv7ZNAv59eTBm3sKqNx/JL4PZaseD/9uhWMaD17bw9ubUSHyEwTFOk0JIFLlxG8ktI6eKqnChwowrZq7EuLnbXGJeGkJ6eFSIHi+N6IBHB7bGyyM6oHFMqMs5VFRpdkltv/bt1ShQyW7yJ4HfewRBAGBPVpMGtELfFnHijXutI+WwS5NoDKhjXYmGSHSYHlufz8SChzNg0GlwV9+m6JMeC15+CYpvIxaos4iN0zy3WPRpXnMlVaGwXHpcOPbOGIzF/7nK43VcboTotXjlpo5uPz+UX4bB769TFEr7bjNLG31/hdRoThAyQjxTa0fK+smiSuzPLUWlWcg80eDg/w3BtKFSp2l/cFO3xrhZpeaJW8uIrA2DkHVUXGlRbaLZECwjcu7r1xwbnr0Oo3oqa8nklVSjrFopRoorLfhs7XF/Ds8FEiME0QARAj6FtMpWCZd+bRethkPPZrHYP2MwXr+5M3o3d+q6K6sXYbayIIb6+LHlWUZta6jFEhdhQJhBB12AfeUNhf5tE/HJXT3cdsQ+U1yFjcdYUTmrLOtGEB4mq00MBI131AtJiDQiOlQPngcW7z4rfufb+/sGJGg4IdKI90d3w44XMhXT3YkR535RG4+5r0cS7qcWC57SIUV5DuSXVrtYRgCWbh1I6CwkiAaI0I13V04xAKBl4qUvRgSEm38n59gXWUdgS11byquQFhuG927vitljeuCBayQ3QE+ntMd4T/qFBAnDu6Rgw7PXKaYJNXAAYN7WU+B5HsdlNUJCHaLipCOuxKDTIDqUBYpyHCd+f8FO5iq7N6OZwnr1/uiu0Go4fHyXel8mXxAXYVSkuJeZrKqtGE47iZHNx91nFfmzqrEntE9RBrtP+SkLf+13bZRoq0t+sw9pmFKOIIKcVokRWLE/X3zf+jISIwKdGjMxMsLyBhaOToSuzWDxM1GM1LP+xy09mGnabudh1GvQt3kcYsL0aP38n+I8QvwA4coj/Vti9ppj+O89PXFFizgs35uHpxf8i2X78jD15924unW8OK9Q6Ew4XjNaxCl6Jo3u3RTbTlwQs5mcb443d2+CoZ1S/G4p+fOxq5F1qhgTv90OOw8UV5phtfOIDtWD49j2HHQEQT/cvyU+XXOsxuU1VDHSzml/l5msWL4v32U+uUsqEJAYIYgGyI1dUxUXv6tkF//LhSaNQhEdqseeqqbYH9sPskLdsgDWi8ty0Wg43NAlVXwfG25AkcMk71En1SBj6vVtcPcVzZDqcNnc3KMxlu/Lw8qDBViUdUbch4AUJ/KXQ4wM7pisWNatPZtAwwHPLPgXFhuPro5y9HIC4bKJizBiYPskNGnE+tcMfHctKsxWdEuLQZcmMZi78YQ47119mmL+jtOqwb0CDVWMRBh1ePe2rnh7+SGxGKAapwMsRshNQxANkPYpUaIZ+Z4rmvm8/kIg4DgOvdOZ6+S9FYcVRaXMtvrHjNSE3Bqilt1DMHRajShEAPY7fHVfb/RpHgueh6KnzfkKM3bmXMDuU8XQcEBme9dA61t6NMGiSf3w2d09XCwjgeZJRyfjMpMVdh7YmVOsECJaDYfUmFDMua83okLcn4cN2dI2qmcTcTvdkesoJRAoSIwQRANl7rg+mDygFZ4d2i7QQ/EZzw5tB4NOgzWHCjFfln5rsdY/ZqQmOjeOEV83D0Dxt0ud23tJJddjww1ix9fnfmWdcEd2a4zEKPWS7h1TozGkU4rPx+gpI7qk4Nmh7XBTt1TVz5OjQqDVcOjUOBr/PHsdBnVIUp2voWXTOBNhdB3fFS2k2B07X3tPJ19CYoQgGigdUqPw5OC2DTZK3xu0SozE1OvbAABeWbxfLNN+MQGsNfHGqM746t5e+Ob+PnXqekwouaV7Yzx4TQuE6rWYktkaSZFMeBzMK4NWw+HRga0DPELP4TgOD13bEh/c0R1vjeri8rlO5iqMCtErLEaXEle0iEOjMGWvmjn39cGChzNEYX7qgmtxO39BYoQgiIAy4armaBEfjrJqK1YdKAAgixnRebcyql6rwcD2Sbi2TYJXlxssaDQcpg1rj/2vDMbYjHRFnMStPZoEpNWANxnUMcmlNYBzTQ5jAyhuVh9iwgzY9nwmvhvfFwBwf7/mCDVo0bNZLKaP6IDvJ/QVg8oDweX7yEUQxCWBTqvBDV1S8OGqo1iUdQbbThSJ1S11mkvzwn+5I5R/79msEfacKUGzuDA8lnnpWUWciQkz4MpW8YqiZq2cMtnsDaSxXH3QaTW4qnU8Nk27DnHhkpDs3wAKKpIYIQgi4AzpxMTImkPKypbN4sICNCKiLrwwvD3GX9UcjWNCFem8lzJ39E7DusOF6NE0BvERRkwbpqwSa5Clm+u1HCw2HrGXWDB0SnTDczWRGCEIIuC0T4kUG9zJua5d4J/YCPfotBqkxV5egnFY5xT89fg1aBobpppyfN+VzbFo11nc3L0x+rdNwBt/HsRLIzoEYKSXFxzPB7jsWh0oLS1FdHQ0SkpKEBXVsNLCCILwDjOXHsB/1yn7Yxx/fdhl88RNEMFIXe/f5JAlCKJBMKSTVCwrOSoE04a2IyFCEEECuWkIgmgQdG/aCP+9pycSI43o3rRR7V8gCOKygcQIQRANBudS4gRBBAfkpiEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqCQGCEIgiAIIqBcEl17eZ4HAJSWlgZ4JARBEARB1BXhvi3cx91xSYiRsrIyAEBaWlqAR0IQBEEQhKeUlZUhOjra7eccX5tcaQDY7XacPXsWkZGR4DjOa8stLS1FWloaTp06haioKK8tl3CF9rV/oP3sH2g/+w/a1/7BV/uZ53mUlZUhNTUVGo37yJBLwjKi0WjQpEkTny0/KiqKDnI/QfvaP9B+9g+0n/0H7Wv/4Iv9XJNFRIACWAmCIAiCCCgkRgiCIAiCCChBLUaMRiOmT58Oo9EY6KFc9tC+9g+0n/0D7Wf/QfvaPwR6P18SAawEQRAEQVy+BLVlhCAIgiCIwENihCAIgiCIgEJihCAIgiCIgEJihCAIgiCIgEJihCAIgiCIgBLUYuSTTz5Beno6QkJC0LdvX2zdujXQQ7qkWLduHUaMGIHU1FRwHIdFixYpPud5Hi+99BJSUlIQGhqKzMxMHDlyRDFPUVERxowZg6ioKMTExGD8+PEoLy/341Y0fGbOnInevXsjMjISiYmJGDlyJA4dOqSYp7q6GpMmTUJcXBwiIiIwatQo5OfnK+bJycnB8OHDERYWhsTERDz11FOwWq3+3JQGzaeffoouXbqIFSgzMjLw559/ip/TPvYNb7zxBjiOw5QpU8RptK+9w8svvwyO4xR/7dq1Ez9vUPuZD1J++ukn3mAw8F9//TW/b98+fuLEiXxMTAyfn58f6KFdMixdupR//vnn+V9//ZUHwC9cuFDx+RtvvMFHR0fzixYt4nfv3s3feOONfPPmzfmqqipxniFDhvBdu3blN2/ezK9fv55v1aoVf+edd/p5Sxo2gwcP5ufMmcPv3buXz8rK4ocNG8Y3bdqULy8vF+d56KGH+LS0NH7lypX89u3b+SuuuIK/8sorxc+tVivfqVMnPjMzk9+1axe/dOlSPj4+np82bVogNqlB8vvvv/NLlizhDx8+zB86dIh/7rnneL1ez+/du5fnedrHvmDr1q18eno636VLF/6xxx4Tp9O+9g7Tp0/nO3bsyOfm5op/hYWF4ucNaT8HrRjp06cPP2nSJPG9zWbjU1NT+ZkzZwZwVJcuzmLEbrfzycnJ/Ntvvy1OKy4u5o1GI//jjz/yPM/z+/fv5wHw27ZtE+f5888/eY7j+DNnzvht7JcaBQUFPAB+7dq1PM+z/arX6/lffvlFnOfAgQM8AH7Tpk08zzPhqNFo+Ly8PHGeTz/9lI+KiuJNJpN/N+ASolGjRvyXX35J+9gHlJWV8a1bt+ZXrFjBX3vttaIYoX3tPaZPn8537dpV9bOGtp+D0k1jNpuxY8cOZGZmitM0Gg0yMzOxadOmAI7s8iE7Oxt5eXmKfRwdHY2+ffuK+3jTpk2IiYlBr169xHkyMzOh0WiwZcsWv4/5UqGkpAQAEBsbCwDYsWMHLBaLYl+3a9cOTZs2Vezrzp07IykpSZxn8ODBKC0txb59+/w4+ksDm82Gn376CRUVFcjIyKB97AMmTZqE4cOHK/YpQMeztzly5AhSU1PRokULjBkzBjk5OQAa3n6+JLr2eptz587BZrMpdjAAJCUl4eDBgwEa1eVFXl4eAKjuY+GzvLw8JCYmKj7X6XSIjY0V5yGU2O12TJkyBf369UOnTp0AsP1oMBgQExOjmNd5X6v9FsJnBGPPnj3IyMhAdXU1IiIisHDhQnTo0AFZWVm0j73ITz/9hJ07d2Lbtm0un9Hx7D369u2LuXPnom3btsjNzcWMGTNw9dVXY+/evQ1uPwelGCGIS5VJkyZh7969+OeffwI9lMuStm3bIisrCyUlJZg/fz7uvfderF27NtDDuqw4deoUHnvsMaxYsQIhISGBHs5lzdChQ8XXXbp0Qd++fdGsWTP8/PPPCA0NDeDIXAlKN018fDy0Wq1L1HB+fj6Sk5MDNKrLC2E/1rSPk5OTUVBQoPjcarWiqKiIfgcVJk+ejMWLF2P16tVo0qSJOD05ORlmsxnFxcWK+Z33tdpvIXxGMAwGA1q1aoWePXti5syZ6Nq1Kz744APax15kx44dKCgoQI8ePaDT6aDT6bB27Vp8+OGH0Ol0SEpKon3tI2JiYtCmTRscPXq0wR3TQSlGDAYDevbsiZUrV4rT7HY7Vq5ciYyMjACO7PKhefPmSE5OVuzj0tJSbNmyRdzHGRkZKC4uxo4dO8R5Vq1aBbvdjr59+/p9zA0VnucxefJkLFy4EKtWrULz5s0Vn/fs2RN6vV6xrw8dOoScnBzFvt6zZ49C/K1YsQJRUVHo0KGDfzbkEsRut8NkMtE+9iIDBw7Enj17kJWVJf716tULY8aMEV/TvvYN5eXlOHbsGFJSUhreMe3VcNhLiJ9++ok3Go383Llz+f379/MPPPAAHxMTo4gaJmqmrKyM37VrF79r1y4eAP/ee+/xu3bt4k+ePMnzPEvtjYmJ4X/77Tf+33//5W+66SbV1N7u3bvzW7Zs4f/55x++devWlNrrxMMPP8xHR0fza9asUaToVVZWivM89NBDfNOmTflVq1bx27dv5zMyMviMjAzxcyFFb9CgQXxWVha/bNkyPiEhgVIhZTz77LP82rVr+ezsbP7ff//ln332WZ7jOP6vv/7ieZ72sS+RZ9PwPO1rb/HEE0/wa9as4bOzs/kNGzbwmZmZfHx8PF9QUMDzfMPaz0ErRnie5z/66CO+adOmvMFg4Pv06cNv3rw50EO6pFi9ejUPwOXv3nvv5Xmepfe++OKLfFJSEm80GvmBAwfyhw4dUizj/Pnz/J133slHRETwUVFR/Lhx4/iysrIAbE3DRW0fA+DnzJkjzlNVVcU/8sgjfKNGjfiwsDD+5ptv5nNzcxXLOXHiBD906FA+NDSUj4+P55944gneYrH4eWsaLvfffz/frFkz3mAw8AkJCfzAgQNFIcLztI99ibMYoX3tHUaPHs2npKTwBoOBb9y4MT969Gj+6NGj4ucNaT9zPM/z3rW1EARBEARB1J2gjBkhCIIgCKLhQGKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiAQmKEIAiCIIiA8v+PjG/VLKZ8pwAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}