notebooks/evaluate_classifier_keras.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "evaluate_classifier-keras.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yGtnt7BrPkwa"
      },
      "source": [
        "# Environmental Sound Classification using Deep Learning - <ins>Keras</ins>\n",
        "## >> Urban sound classification with CNNs"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N4jlbQhjQevI"
      },
      "source": [
        "* [0. Load the preprocessed data](#zero-bullet)\n",
        "* [1. Data augmentation](#first-bullet)\n",
        "* [2. CNN model](#second-bullet)\n",
        "* [3. Helper functions](#third-bullet)\n",
        "* [4. 10-Fold Cross Validation](#fourth-bullet)\n",
        "* [5. Results](#fifth-bullet)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KZvnMHAfRckQ"
      },
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GXvNOAkHfyzC"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "from keras import regularizers, activations\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout, Flatten, Activation, Conv2D, MaxPooling2D, GlobalAveragePooling2D\n",
        "from keras.utils import np_utils, to_categorical\n",
        "\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "from datetime import datetime \n",
        "\n",
        "from matplotlib import pyplot as plt\n",
        "%matplotlib inline"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k4vFD1-3gtGl",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "08ab49c1-f8ac-4f38-b885-d9817f623b68"
      },
      "source": [
        "USE_GOOGLE_COLAB = True\n",
        "\n",
        "if USE_GOOGLE_COLAB:\n",
        "    from google.colab import drive \n",
        "    drive.mount('/content/gdrive')\n",
        "\n",
        "    # change the current working directory\n",
        "    %cd gdrive/'My Drive'/US8K\n",
        "else:\n",
        "    %cd US8K"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/gdrive\n",
            "/content/gdrive/My Drive/US8K\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1GkqUlX8RXft"
      },
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dUZ4Lgh0RVKR"
      },
      "source": [
        "## 0. Load the preprocessed data <a name=\"zero-bullet\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CEcVX1xmgHMy"
      },
      "source": [
        "us8k_df = pd.read_pickle(\"us8k_df.pkl\")"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "fwIqLRs5Rg70",
        "outputId": "d944d0b6-45ca-4193-a275-30dd28b5a38d"
      },
      "source": [
        "us8k_df.head()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>melspectrogram</th>\n",
              "      <th>label</th>\n",
              "      <th>fold</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>[[-61.70703, -74.49714, -65.133354, -65.751175...</td>\n",
              "      <td>3</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>[[-11.593532, -3.6112566, -12.501208, -13.6347...</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>[[-24.203024, -14.915937, -6.091387, -12.99589...</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>[[-15.058913, -12.812816, -12.299819, -11.4387...</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>[[-9.8777, 0.0, -11.115205, -10.564414, -3.854...</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                      melspectrogram  label  fold\n",
              "0  [[-61.70703, -74.49714, -65.133354, -65.751175...      3     5\n",
              "1  [[-11.593532, -3.6112566, -12.501208, -13.6347...      2     5\n",
              "2  [[-24.203024, -14.915937, -6.091387, -12.99589...      2     5\n",
              "3  [[-15.058913, -12.812816, -12.299819, -11.4387...      2     5\n",
              "4  [[-9.8777, 0.0, -11.115205, -10.564414, -3.854...      2     5"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jbfmf036RY3x"
      },
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uTJCle0qSG1E"
      },
      "source": [
        "## 1. Data augmentation <a name=\"first-bullet\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y1hcGDoSg8pw"
      },
      "source": [
        "def init_data_aug():\n",
        "    train_datagen = ImageDataGenerator(\n",
        "        featurewise_center=True,\n",
        "        featurewise_std_normalization=True,\n",
        "        fill_mode = 'constant',\n",
        "        cval=-80.0,\n",
        "        width_shift_range=0.1,\n",
        "        height_shift_range=0.0)\n",
        "\n",
        "    val_datagen = ImageDataGenerator(\n",
        "        featurewise_center=True,\n",
        "        featurewise_std_normalization=True,\n",
        "        fill_mode = 'constant',\n",
        "        cval=-80.0)\n",
        "    \n",
        "    return train_datagen, val_datagen"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "K4IRqawLUnNY"
      },
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qQUZdDRuSt1P"
      },
      "source": [
        "## 2. CNN model  <a name=\"second-bullet\"></a>\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ElWVJ09JhDBQ"
      },
      "source": [
        "def init_model():\n",
        "    model1 = Sequential()\n",
        "    \n",
        "    #layer-1\n",
        "    model1.add(Conv2D(filters=24, kernel_size=5, input_shape=(128, 128, 1),\n",
        "                      kernel_regularizer=regularizers.l2(1e-3)))\n",
        "    model1.add(MaxPooling2D(pool_size=(3,3), strides=3))\n",
        "    model1.add(Activation(activations.relu))\n",
        "    \n",
        "    #layer-2\n",
        "    model1.add(Conv2D(filters=36, kernel_size=4, padding='valid', kernel_regularizer=regularizers.l2(1e-3)))\n",
        "    model1.add(MaxPooling2D(pool_size=(2,2), strides=2))\n",
        "    model1.add(Activation(activations.relu))\n",
        "    \n",
        "    #layer-3\n",
        "    model1.add(Conv2D(filters=48, kernel_size=3, padding='valid'))\n",
        "    model1.add(Activation(activations.relu))\n",
        "    \n",
        "    model1.add(GlobalAveragePooling2D())\n",
        "    \n",
        "    #layer-4 (1st dense layer)\n",
        "    model1.add(Dense(60, activation='relu'))\n",
        "    model1.add(Dropout(0.5))\n",
        "    \n",
        "    #layer-5 (2nd dense layer)\n",
        "    model1.add(Dense(10, activation='softmax'))\n",
        "\n",
        "    \n",
        "    # compile\n",
        "    model1.compile(loss='categorical_crossentropy', metrics=['accuracy'], optimizer='adam')\n",
        "    \n",
        "    return model1"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "awJfOlpChFNL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6233c99a-ec16-47d8-ccfb-cb2c0ecbaed0"
      },
      "source": [
        "model = init_model()\n",
        "model.summary()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 124, 124, 24)      624       \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 41, 41, 24)        0         \n",
            "_________________________________________________________________\n",
            "activation (Activation)      (None, 41, 41, 24)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 38, 38, 36)        13860     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 19, 19, 36)        0         \n",
            "_________________________________________________________________\n",
            "activation_1 (Activation)    (None, 19, 19, 36)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 17, 17, 48)        15600     \n",
            "_________________________________________________________________\n",
            "activation_2 (Activation)    (None, 17, 17, 48)        0         \n",
            "_________________________________________________________________\n",
            "global_average_pooling2d (Gl (None, 48)                0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 60)                2940      \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 60)                0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 10)                610       \n",
            "=================================================================\n",
            "Total params: 33,634\n",
            "Trainable params: 33,634\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jdDK1TBkUlaz"
      },
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bJSCcMe_S2dX"
      },
      "source": [
        "## 3. Helper functions  <a name=\"third-bullet\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ywrCPzgFg5Ea"
      },
      "source": [
        "def train_test_split(fold_k, data, X_dim=(128, 128, 1)):\n",
        "    X_train = np.stack(data[data.fold != fold_k].melspectrogram.to_numpy())\n",
        "    X_test = np.stack(data[data.fold == fold_k].melspectrogram.to_numpy())\n",
        "\n",
        "    y_train = data[data.fold != fold_k].label.to_numpy()\n",
        "    y_test = data[data.fold == fold_k].label.to_numpy()\n",
        "\n",
        "    XX_train = X_train.reshape(X_train.shape[0], *X_dim)\n",
        "    XX_test = X_test.reshape(X_test.shape[0], *X_dim)\n",
        "    \n",
        "    yy_train = to_categorical(y_train)\n",
        "    yy_test = to_categorical(y_test)\n",
        "    \n",
        "    return XX_train, XX_test, yy_train, yy_test"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FJQ43K0Cg-0m"
      },
      "source": [
        "def process_fold(fold_k, data, epochs=100, num_batch_size=32):\n",
        "    # split the data\n",
        "    X_train, X_test, y_train, y_test = train_test_split(fold_k, data)\n",
        "\n",
        "    # init data augmention\n",
        "    train_datagen, val_datagen = init_data_aug()\n",
        "    \n",
        "    # fit augmentation\n",
        "    train_datagen.fit(X_train)\n",
        "    val_datagen.fit(X_train)\n",
        "\n",
        "    # init model\n",
        "    model = init_model()\n",
        "\n",
        "    # pre-training accuracy\n",
        "    score = model.evaluate(val_datagen.flow(X_test, y_test, batch_size=num_batch_size), verbose=0)\n",
        "    print(\"Pre-training accuracy: %.4f%%\\n\" % (100 * score[1]))\n",
        "    \n",
        "    # train the model\n",
        "    start = datetime.now()\n",
        "    history = model.fit(train_datagen.flow(X_train, y_train, batch_size=num_batch_size), \n",
        "                        steps_per_epoch=len(X_train) / num_batch_size, \n",
        "                        epochs=epochs, \n",
        "                        validation_data=val_datagen.flow(X_test, y_test, batch_size=num_batch_size))\n",
        "    end = datetime.now()\n",
        "    print(\"Training completed in time: \", end - start, '\\n')\n",
        "    \n",
        "    return history"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-Epn1AfTg6xC"
      },
      "source": [
        "def show_results(tot_history):\n",
        "    \"\"\"Show accuracy and loss graphs for train and test sets.\"\"\"\n",
        "\n",
        "    for i, history in enumerate(tot_history):\n",
        "        print('\\n({})'.format(i+1))\n",
        "\n",
        "        plt.figure(figsize=(15,5))\n",
        "\n",
        "        plt.subplot(121)\n",
        "        plt.plot(history.history['accuracy'])\n",
        "        plt.plot(history.history['val_accuracy'])\n",
        "        plt.grid(linestyle='--')\n",
        "        plt.ylabel('Accuracy')\n",
        "        plt.xlabel('Epoch')\n",
        "        plt.legend(['train', 'validation'], loc='upper left')\n",
        "\n",
        "        plt.subplot(122)\n",
        "        plt.plot(history.history['loss'])\n",
        "        plt.plot(history.history['val_loss'])\n",
        "        plt.grid(linestyle='--')\n",
        "        plt.ylabel('Loss')\n",
        "        plt.xlabel('Epoch')\n",
        "        plt.legend(['train', 'validation'], loc='upper left')\n",
        "            \n",
        "        plt.show()\n",
        "\n",
        "        print('\\tMax validation accuracy: %.4f %%' % (np.max(history.history['val_accuracy']) * 100))\n",
        "        print('\\tMin validation loss: %.5f' % np.min(history.history['val_loss']))"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3zhCJ0IaUjKi"
      },
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "55A8iD_uS7XC"
      },
      "source": [
        "## 4. 10-Fold Cross Validation  <a name=\"fourth-bullet\"></a>\n",
        "\n",
        "* [fold-1](#fold-1)\n",
        "* [fold-2](#fold-2)\n",
        "* [fold-3](#fold-3)\n",
        "* [fold-4](#fold-4)\n",
        "* [fold-5](#fold-5)\n",
        "* [fold-6](#fold-6)\n",
        "* [fold-7](#fold-7)\n",
        "* [fold-8](#fold-8)\n",
        "* [fold-9](#fold-9)\n",
        "* [fold-10](#fold-10)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pWdts6xvTsef"
      },
      "source": [
        "### fold-1 <a name=\"fold-1\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5uOcnV1cTr1q",
        "outputId": "7bbf1106-2358-405c-f788-b793d7285ebd"
      },
      "source": [
        "FOLD_K = 1\n",
        "REPEAT = 1\n",
        "\n",
        "history1 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history1.append(history)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 2.8637%\n",
            "\n",
            "Epoch 1/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 2.0781 - accuracy: 0.2257 - val_loss: 1.7888 - val_accuracy: 0.3058\n",
            "Epoch 2/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.8673 - accuracy: 0.2980 - val_loss: 1.6559 - val_accuracy: 0.3677\n",
            "Epoch 3/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.7244 - accuracy: 0.3640 - val_loss: 1.4030 - val_accuracy: 0.4502\n",
            "Epoch 4/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.5954 - accuracy: 0.4185 - val_loss: 1.3822 - val_accuracy: 0.4548\n",
            "Epoch 5/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.5095 - accuracy: 0.4495 - val_loss: 1.2706 - val_accuracy: 0.5475\n",
            "Epoch 6/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.4303 - accuracy: 0.4816 - val_loss: 1.2291 - val_accuracy: 0.4914\n",
            "Epoch 7/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.3772 - accuracy: 0.4960 - val_loss: 1.1378 - val_accuracy: 0.5487\n",
            "Epoch 8/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.3315 - accuracy: 0.5212 - val_loss: 1.0493 - val_accuracy: 0.5979\n",
            "Epoch 9/100\n",
            "246/245 [==============================] - 11s 45ms/step - loss: 1.2821 - accuracy: 0.5393 - val_loss: 1.0833 - val_accuracy: 0.6323\n",
            "Epoch 10/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.2517 - accuracy: 0.5591 - val_loss: 1.0994 - val_accuracy: 0.6208\n",
            "Epoch 11/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.2185 - accuracy: 0.5665 - val_loss: 1.0508 - val_accuracy: 0.6243\n",
            "Epoch 12/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.1964 - accuracy: 0.5741 - val_loss: 1.0354 - val_accuracy: 0.6266\n",
            "Epoch 13/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.1629 - accuracy: 0.5865 - val_loss: 1.0339 - val_accuracy: 0.6838\n",
            "Epoch 14/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.1422 - accuracy: 0.5949 - val_loss: 1.0118 - val_accuracy: 0.7159\n",
            "Epoch 15/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.1165 - accuracy: 0.6118 - val_loss: 0.9841 - val_accuracy: 0.6987\n",
            "Epoch 16/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.0815 - accuracy: 0.6225 - val_loss: 0.9845 - val_accuracy: 0.7033\n",
            "Epoch 17/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.0774 - accuracy: 0.6243 - val_loss: 0.9358 - val_accuracy: 0.7388\n",
            "Epoch 18/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.0542 - accuracy: 0.6277 - val_loss: 0.9675 - val_accuracy: 0.6930\n",
            "Epoch 19/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.0394 - accuracy: 0.6423 - val_loss: 1.0483 - val_accuracy: 0.6827\n",
            "Epoch 20/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.0111 - accuracy: 0.6528 - val_loss: 0.9789 - val_accuracy: 0.6816\n",
            "Epoch 21/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 1.0138 - accuracy: 0.6573 - val_loss: 0.9206 - val_accuracy: 0.7411\n",
            "Epoch 22/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 0.9802 - accuracy: 0.6648 - val_loss: 0.9425 - val_accuracy: 0.7125\n",
            "Epoch 23/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 0.9588 - accuracy: 0.6707 - val_loss: 0.9008 - val_accuracy: 0.7446\n",
            "Epoch 24/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 0.9398 - accuracy: 0.6825 - val_loss: 0.9146 - val_accuracy: 0.7331\n",
            "Epoch 25/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 0.9290 - accuracy: 0.6872 - val_loss: 0.9032 - val_accuracy: 0.6942\n",
            "Epoch 26/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 0.9075 - accuracy: 0.6972 - val_loss: 0.8850 - val_accuracy: 0.7342\n",
            "Epoch 27/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 0.8886 - accuracy: 0.7062 - val_loss: 0.8038 - val_accuracy: 0.7652\n",
            "Epoch 28/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 0.8739 - accuracy: 0.7067 - val_loss: 0.8952 - val_accuracy: 0.7423\n",
            "Epoch 29/100\n",
            "246/245 [==============================] - 11s 44ms/step - loss: 0.8705 - accuracy: 0.7133 - val_loss: 0.8538 - val_accuracy: 0.7503\n",
            "Epoch 30/100\n",
            "246/245 [==============================] - 11s 43ms/step - loss: 0.8596 - accuracy: 0.7281 - val_loss: 0.8298 - val_accuracy: 0.7640\n",
            "Epoch 31/100\n",
            "246/245 [==============================] - 11s 43ms/step - loss: 0.8311 - accuracy: 0.7343 - val_loss: 0.8208 - val_accuracy: 0.7869\n",
            "Epoch 32/100\n",
            "246/245 [==============================] - 11s 43ms/step - loss: 0.8347 - accuracy: 0.7304 - val_loss: 0.8680 - val_accuracy: 0.7721\n",
            "Epoch 33/100\n",
            "246/245 [==============================] - 10s 43ms/step - loss: 0.8328 - accuracy: 0.7339 - val_loss: 0.7541 - val_accuracy: 0.7858\n",
            "Epoch 34/100\n",
            "246/245 [==============================] - 10s 43ms/step - loss: 0.8132 - accuracy: 0.7292 - val_loss: 0.7828 - val_accuracy: 0.7847\n",
            "Epoch 35/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.7747 - accuracy: 0.7551 - val_loss: 0.7866 - val_accuracy: 0.7640\n",
            "Epoch 36/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.7972 - accuracy: 0.7517 - val_loss: 0.7944 - val_accuracy: 0.7721\n",
            "Epoch 37/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.7696 - accuracy: 0.7619 - val_loss: 0.7808 - val_accuracy: 0.7904\n",
            "Epoch 38/100\n",
            "246/245 [==============================] - 11s 43ms/step - loss: 0.7702 - accuracy: 0.7627 - val_loss: 0.7882 - val_accuracy: 0.7778\n",
            "Epoch 39/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7429 - accuracy: 0.7747 - val_loss: 0.7509 - val_accuracy: 0.7847\n",
            "Epoch 40/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7405 - accuracy: 0.7721 - val_loss: 0.7939 - val_accuracy: 0.7721\n",
            "Epoch 41/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7373 - accuracy: 0.7726 - val_loss: 0.7601 - val_accuracy: 0.7927\n",
            "Epoch 42/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7241 - accuracy: 0.7815 - val_loss: 0.7920 - val_accuracy: 0.7927\n",
            "Epoch 43/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.7305 - accuracy: 0.7775 - val_loss: 0.7527 - val_accuracy: 0.8121\n",
            "Epoch 44/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7053 - accuracy: 0.7846 - val_loss: 0.8318 - val_accuracy: 0.7468\n",
            "Epoch 45/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.7026 - accuracy: 0.7832 - val_loss: 0.8127 - val_accuracy: 0.7812\n",
            "Epoch 46/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6848 - accuracy: 0.7886 - val_loss: 0.8632 - val_accuracy: 0.7721\n",
            "Epoch 47/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6839 - accuracy: 0.7881 - val_loss: 0.8308 - val_accuracy: 0.7824\n",
            "Epoch 48/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6748 - accuracy: 0.7960 - val_loss: 0.7551 - val_accuracy: 0.7869\n",
            "Epoch 49/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6719 - accuracy: 0.7982 - val_loss: 0.7793 - val_accuracy: 0.7950\n",
            "Epoch 50/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6784 - accuracy: 0.7941 - val_loss: 0.7391 - val_accuracy: 0.7938\n",
            "Epoch 51/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6528 - accuracy: 0.8004 - val_loss: 0.8667 - val_accuracy: 0.7961\n",
            "Epoch 52/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6437 - accuracy: 0.8076 - val_loss: 0.7702 - val_accuracy: 0.8076\n",
            "Epoch 53/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6456 - accuracy: 0.8024 - val_loss: 0.8094 - val_accuracy: 0.7858\n",
            "Epoch 54/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6397 - accuracy: 0.8100 - val_loss: 0.7847 - val_accuracy: 0.7881\n",
            "Epoch 55/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6277 - accuracy: 0.8104 - val_loss: 0.7777 - val_accuracy: 0.7915\n",
            "Epoch 56/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6220 - accuracy: 0.8116 - val_loss: 0.9045 - val_accuracy: 0.7526\n",
            "Epoch 57/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6370 - accuracy: 0.8070 - val_loss: 0.8163 - val_accuracy: 0.8053\n",
            "Epoch 58/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6179 - accuracy: 0.8121 - val_loss: 0.7605 - val_accuracy: 0.8007\n",
            "Epoch 59/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6148 - accuracy: 0.8158 - val_loss: 0.7690 - val_accuracy: 0.8030\n",
            "Epoch 60/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6000 - accuracy: 0.8216 - val_loss: 0.8790 - val_accuracy: 0.7640\n",
            "Epoch 61/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6015 - accuracy: 0.8210 - val_loss: 0.8372 - val_accuracy: 0.7904\n",
            "Epoch 62/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5954 - accuracy: 0.8180 - val_loss: 0.7666 - val_accuracy: 0.7961\n",
            "Epoch 63/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5982 - accuracy: 0.8224 - val_loss: 0.7975 - val_accuracy: 0.7721\n",
            "Epoch 64/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5853 - accuracy: 0.8242 - val_loss: 0.8486 - val_accuracy: 0.7995\n",
            "Epoch 65/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5912 - accuracy: 0.8263 - val_loss: 0.8932 - val_accuracy: 0.7858\n",
            "Epoch 66/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5796 - accuracy: 0.8318 - val_loss: 0.8502 - val_accuracy: 0.7766\n",
            "Epoch 67/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5715 - accuracy: 0.8317 - val_loss: 0.8539 - val_accuracy: 0.7847\n",
            "Epoch 68/100\n",
            "246/245 [==============================] - 10s 43ms/step - loss: 0.5583 - accuracy: 0.8385 - val_loss: 0.9224 - val_accuracy: 0.7927\n",
            "Epoch 69/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5690 - accuracy: 0.8327 - val_loss: 0.8551 - val_accuracy: 0.7595\n",
            "Epoch 70/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5616 - accuracy: 0.8347 - val_loss: 0.8776 - val_accuracy: 0.7297\n",
            "Epoch 71/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5491 - accuracy: 0.8378 - val_loss: 0.8645 - val_accuracy: 0.7468\n",
            "Epoch 72/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5404 - accuracy: 0.8427 - val_loss: 0.8894 - val_accuracy: 0.7732\n",
            "Epoch 73/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5458 - accuracy: 0.8393 - val_loss: 0.8377 - val_accuracy: 0.7755\n",
            "Epoch 74/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5408 - accuracy: 0.8339 - val_loss: 0.8562 - val_accuracy: 0.7824\n",
            "Epoch 75/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5306 - accuracy: 0.8504 - val_loss: 0.9283 - val_accuracy: 0.7400\n",
            "Epoch 76/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5361 - accuracy: 0.8434 - val_loss: 0.9776 - val_accuracy: 0.7388\n",
            "Epoch 77/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5140 - accuracy: 0.8529 - val_loss: 0.9832 - val_accuracy: 0.7457\n",
            "Epoch 78/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5449 - accuracy: 0.8451 - val_loss: 1.0569 - val_accuracy: 0.7583\n",
            "Epoch 79/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5292 - accuracy: 0.8465 - val_loss: 0.8369 - val_accuracy: 0.7881\n",
            "Epoch 80/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5052 - accuracy: 0.8518 - val_loss: 0.9646 - val_accuracy: 0.7572\n",
            "Epoch 81/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5191 - accuracy: 0.8485 - val_loss: 0.9243 - val_accuracy: 0.7743\n",
            "Epoch 82/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4970 - accuracy: 0.8583 - val_loss: 1.1219 - val_accuracy: 0.7205\n",
            "Epoch 83/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5026 - accuracy: 0.8541 - val_loss: 0.9241 - val_accuracy: 0.7847\n",
            "Epoch 84/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5151 - accuracy: 0.8547 - val_loss: 0.9491 - val_accuracy: 0.7537\n",
            "Epoch 85/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.4966 - accuracy: 0.8574 - val_loss: 1.0566 - val_accuracy: 0.7365\n",
            "Epoch 86/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5093 - accuracy: 0.8542 - val_loss: 0.9546 - val_accuracy: 0.7721\n",
            "Epoch 87/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5012 - accuracy: 0.8585 - val_loss: 1.0061 - val_accuracy: 0.7491\n",
            "Epoch 88/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4813 - accuracy: 0.8644 - val_loss: 1.1609 - val_accuracy: 0.7560\n",
            "Epoch 89/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4750 - accuracy: 0.8666 - val_loss: 1.1679 - val_accuracy: 0.7194\n",
            "Epoch 90/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4833 - accuracy: 0.8600 - val_loss: 0.9671 - val_accuracy: 0.7617\n",
            "Epoch 91/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4791 - accuracy: 0.8675 - val_loss: 1.1817 - val_accuracy: 0.7342\n",
            "Epoch 92/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4696 - accuracy: 0.8679 - val_loss: 1.1386 - val_accuracy: 0.7721\n",
            "Epoch 93/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4723 - accuracy: 0.8654 - val_loss: 1.1684 - val_accuracy: 0.7171\n",
            "Epoch 94/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4854 - accuracy: 0.8618 - val_loss: 1.0863 - val_accuracy: 0.7468\n",
            "Epoch 95/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4616 - accuracy: 0.8675 - val_loss: 1.0418 - val_accuracy: 0.7503\n",
            "Epoch 96/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4652 - accuracy: 0.8679 - val_loss: 1.0979 - val_accuracy: 0.7583\n",
            "Epoch 97/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.4746 - accuracy: 0.8647 - val_loss: 1.1480 - val_accuracy: 0.7480\n",
            "Epoch 98/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.4555 - accuracy: 0.8725 - val_loss: 1.2418 - val_accuracy: 0.7549\n",
            "Epoch 99/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.4508 - accuracy: 0.8738 - val_loss: 1.1835 - val_accuracy: 0.7491\n",
            "Epoch 100/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4636 - accuracy: 0.8672 - val_loss: 1.0084 - val_accuracy: 0.7377\n",
            "Training completed in time:  0:17:27.446960 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h4YP5DsnhmXb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "outputId": "6dbe32b7-4ab9-41d3-86a6-fe506be7d895"
      },
      "source": [
        "show_results(history1)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 81.2142 %\n",
            "\tMin validation loss: 0.73915\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LtSLzK3CWaCK"
      },
      "source": [
        "### fold-2 <a name=\"fold-2\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iZcEwPHwWcxR",
        "outputId": "60f7b6f5-4e2c-4c3f-eba9-bce5c70f7d32"
      },
      "source": [
        "FOLD_K = 2\n",
        "REPEAT = 1\n",
        "\n",
        "history2 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history2.append(history)"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 14.1892%\n",
            "\n",
            "Epoch 1/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 2.0893 - accuracy: 0.2235 - val_loss: 1.8992 - val_accuracy: 0.2354\n",
            "Epoch 2/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.9092 - accuracy: 0.2708 - val_loss: 1.7946 - val_accuracy: 0.2354\n",
            "Epoch 3/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 1.8013 - accuracy: 0.3146 - val_loss: 1.6632 - val_accuracy: 0.3604\n",
            "Epoch 4/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.6710 - accuracy: 0.3877 - val_loss: 1.5538 - val_accuracy: 0.3333\n",
            "Epoch 5/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.5538 - accuracy: 0.4326 - val_loss: 1.4228 - val_accuracy: 0.4212\n",
            "Epoch 6/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.4486 - accuracy: 0.4786 - val_loss: 1.4721 - val_accuracy: 0.4313\n",
            "Epoch 7/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.3725 - accuracy: 0.5119 - val_loss: 1.3924 - val_accuracy: 0.4944\n",
            "Epoch 8/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.3340 - accuracy: 0.5260 - val_loss: 1.3989 - val_accuracy: 0.4617\n",
            "Epoch 9/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.2997 - accuracy: 0.5407 - val_loss: 1.6702 - val_accuracy: 0.3863\n",
            "Epoch 10/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.2801 - accuracy: 0.5602 - val_loss: 1.3545 - val_accuracy: 0.4876\n",
            "Epoch 11/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.2211 - accuracy: 0.5741 - val_loss: 1.2824 - val_accuracy: 0.4741\n",
            "Epoch 12/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.1875 - accuracy: 0.5880 - val_loss: 1.3172 - val_accuracy: 0.5045\n",
            "Epoch 13/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.1627 - accuracy: 0.6062 - val_loss: 1.2061 - val_accuracy: 0.5338\n",
            "Epoch 14/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.1379 - accuracy: 0.6086 - val_loss: 1.2205 - val_accuracy: 0.5484\n",
            "Epoch 15/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.0961 - accuracy: 0.6295 - val_loss: 1.2875 - val_accuracy: 0.5721\n",
            "Epoch 16/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.0962 - accuracy: 0.6246 - val_loss: 1.2752 - val_accuracy: 0.5901\n",
            "Epoch 17/100\n",
            "246/245 [==============================] - 10s 40ms/step - loss: 1.0699 - accuracy: 0.6368 - val_loss: 1.3018 - val_accuracy: 0.4730\n",
            "Epoch 18/100\n",
            "246/245 [==============================] - 10s 40ms/step - loss: 1.0520 - accuracy: 0.6456 - val_loss: 1.3146 - val_accuracy: 0.5349\n",
            "Epoch 19/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.0330 - accuracy: 0.6562 - val_loss: 1.1601 - val_accuracy: 0.5552\n",
            "Epoch 20/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 1.0122 - accuracy: 0.6608 - val_loss: 1.1970 - val_accuracy: 0.6014\n",
            "Epoch 21/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.9894 - accuracy: 0.6699 - val_loss: 1.1763 - val_accuracy: 0.5563\n",
            "Epoch 22/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.9688 - accuracy: 0.6809 - val_loss: 1.0881 - val_accuracy: 0.6498\n",
            "Epoch 23/100\n",
            "246/245 [==============================] - 10s 40ms/step - loss: 0.9504 - accuracy: 0.6889 - val_loss: 1.1057 - val_accuracy: 0.5845\n",
            "Epoch 24/100\n",
            "246/245 [==============================] - 10s 40ms/step - loss: 0.9348 - accuracy: 0.6925 - val_loss: 1.1878 - val_accuracy: 0.5495\n",
            "Epoch 25/100\n",
            "246/245 [==============================] - 10s 40ms/step - loss: 0.9280 - accuracy: 0.6998 - val_loss: 1.0829 - val_accuracy: 0.6374\n",
            "Epoch 26/100\n",
            "246/245 [==============================] - 10s 40ms/step - loss: 0.9073 - accuracy: 0.7049 - val_loss: 1.1684 - val_accuracy: 0.6475\n",
            "Epoch 27/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.8959 - accuracy: 0.7068 - val_loss: 1.1446 - val_accuracy: 0.6194\n",
            "Epoch 28/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.8896 - accuracy: 0.7143 - val_loss: 1.0501 - val_accuracy: 0.6655\n",
            "Epoch 29/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.8660 - accuracy: 0.7226 - val_loss: 1.1553 - val_accuracy: 0.5912\n",
            "Epoch 30/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.8427 - accuracy: 0.7342 - val_loss: 1.0789 - val_accuracy: 0.6464\n",
            "Epoch 31/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.8392 - accuracy: 0.7302 - val_loss: 1.0157 - val_accuracy: 0.6520\n",
            "Epoch 32/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.8359 - accuracy: 0.7278 - val_loss: 1.1145 - val_accuracy: 0.6577\n",
            "Epoch 33/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.8128 - accuracy: 0.7407 - val_loss: 1.0424 - val_accuracy: 0.6610\n",
            "Epoch 34/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7962 - accuracy: 0.7489 - val_loss: 1.0454 - val_accuracy: 0.6239\n",
            "Epoch 35/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7902 - accuracy: 0.7485 - val_loss: 1.1069 - val_accuracy: 0.6565\n",
            "Epoch 36/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7830 - accuracy: 0.7543 - val_loss: 1.0124 - val_accuracy: 0.6599\n",
            "Epoch 37/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7699 - accuracy: 0.7568 - val_loss: 0.9978 - val_accuracy: 0.6464\n",
            "Epoch 38/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7673 - accuracy: 0.7550 - val_loss: 1.1235 - val_accuracy: 0.6295\n",
            "Epoch 39/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7559 - accuracy: 0.7662 - val_loss: 1.2439 - val_accuracy: 0.6745\n",
            "Epoch 40/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7423 - accuracy: 0.7718 - val_loss: 1.0903 - val_accuracy: 0.6779\n",
            "Epoch 41/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7337 - accuracy: 0.7662 - val_loss: 1.0083 - val_accuracy: 0.6892\n",
            "Epoch 42/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7435 - accuracy: 0.7717 - val_loss: 1.2040 - val_accuracy: 0.6430\n",
            "Epoch 43/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7177 - accuracy: 0.7810 - val_loss: 1.0525 - val_accuracy: 0.6791\n",
            "Epoch 44/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6993 - accuracy: 0.7852 - val_loss: 1.1562 - val_accuracy: 0.6982\n",
            "Epoch 45/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7064 - accuracy: 0.7838 - val_loss: 1.0252 - val_accuracy: 0.6441\n",
            "Epoch 46/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.7075 - accuracy: 0.7770 - val_loss: 1.0536 - val_accuracy: 0.6903\n",
            "Epoch 47/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6885 - accuracy: 0.7903 - val_loss: 1.0169 - val_accuracy: 0.6836\n",
            "Epoch 48/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6975 - accuracy: 0.7856 - val_loss: 1.0949 - val_accuracy: 0.6610\n",
            "Epoch 49/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6674 - accuracy: 0.7984 - val_loss: 1.0563 - val_accuracy: 0.6689\n",
            "Epoch 50/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6766 - accuracy: 0.7963 - val_loss: 1.0274 - val_accuracy: 0.6464\n",
            "Epoch 51/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6663 - accuracy: 0.8001 - val_loss: 1.0392 - val_accuracy: 0.7050\n",
            "Epoch 52/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6531 - accuracy: 0.7991 - val_loss: 1.1181 - val_accuracy: 0.6723\n",
            "Epoch 53/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6624 - accuracy: 0.7998 - val_loss: 1.1775 - val_accuracy: 0.6791\n",
            "Epoch 54/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6483 - accuracy: 0.8052 - val_loss: 1.1132 - val_accuracy: 0.6520\n",
            "Epoch 55/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6406 - accuracy: 0.8006 - val_loss: 1.0937 - val_accuracy: 0.6655\n",
            "Epoch 56/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6236 - accuracy: 0.8072 - val_loss: 1.2044 - val_accuracy: 0.6712\n",
            "Epoch 57/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6332 - accuracy: 0.8097 - val_loss: 1.1881 - val_accuracy: 0.6802\n",
            "Epoch 58/100\n",
            "246/245 [==============================] - 10s 40ms/step - loss: 0.6251 - accuracy: 0.8107 - val_loss: 1.2915 - val_accuracy: 0.6633\n",
            "Epoch 59/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6252 - accuracy: 0.8114 - val_loss: 1.1339 - val_accuracy: 0.6453\n",
            "Epoch 60/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.6164 - accuracy: 0.8143 - val_loss: 1.3016 - val_accuracy: 0.6768\n",
            "Epoch 61/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6085 - accuracy: 0.8173 - val_loss: 1.2214 - val_accuracy: 0.6712\n",
            "Epoch 62/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6070 - accuracy: 0.8183 - val_loss: 1.1804 - val_accuracy: 0.6926\n",
            "Epoch 63/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6024 - accuracy: 0.8197 - val_loss: 1.1431 - val_accuracy: 0.6700\n",
            "Epoch 64/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.6022 - accuracy: 0.8210 - val_loss: 1.1437 - val_accuracy: 0.7083\n",
            "Epoch 65/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5956 - accuracy: 0.8200 - val_loss: 1.2902 - val_accuracy: 0.6667\n",
            "Epoch 66/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5788 - accuracy: 0.8269 - val_loss: 1.2203 - val_accuracy: 0.6982\n",
            "Epoch 67/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5932 - accuracy: 0.8275 - val_loss: 1.1387 - val_accuracy: 0.6847\n",
            "Epoch 68/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5800 - accuracy: 0.8290 - val_loss: 1.3084 - val_accuracy: 0.7050\n",
            "Epoch 69/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5819 - accuracy: 0.8275 - val_loss: 1.1540 - val_accuracy: 0.6599\n",
            "Epoch 70/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5611 - accuracy: 0.8327 - val_loss: 1.1230 - val_accuracy: 0.6926\n",
            "Epoch 71/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5704 - accuracy: 0.8274 - val_loss: 1.3102 - val_accuracy: 0.6836\n",
            "Epoch 72/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5785 - accuracy: 0.8334 - val_loss: 1.2008 - val_accuracy: 0.6847\n",
            "Epoch 73/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5500 - accuracy: 0.8401 - val_loss: 1.1813 - val_accuracy: 0.7061\n",
            "Epoch 74/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5495 - accuracy: 0.8395 - val_loss: 1.0374 - val_accuracy: 0.7072\n",
            "Epoch 75/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5544 - accuracy: 0.8377 - val_loss: 1.0227 - val_accuracy: 0.7005\n",
            "Epoch 76/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5455 - accuracy: 0.8410 - val_loss: 1.1911 - val_accuracy: 0.6723\n",
            "Epoch 77/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5394 - accuracy: 0.8385 - val_loss: 1.1588 - val_accuracy: 0.6847\n",
            "Epoch 78/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5239 - accuracy: 0.8427 - val_loss: 1.1237 - val_accuracy: 0.6858\n",
            "Epoch 79/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5464 - accuracy: 0.8433 - val_loss: 1.0750 - val_accuracy: 0.6543\n",
            "Epoch 80/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5353 - accuracy: 0.8392 - val_loss: 1.0955 - val_accuracy: 0.6813\n",
            "Epoch 81/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5306 - accuracy: 0.8447 - val_loss: 1.1324 - val_accuracy: 0.6791\n",
            "Epoch 82/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5251 - accuracy: 0.8443 - val_loss: 1.1287 - val_accuracy: 0.6655\n",
            "Epoch 83/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5269 - accuracy: 0.8460 - val_loss: 1.1759 - val_accuracy: 0.7027\n",
            "Epoch 84/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5239 - accuracy: 0.8460 - val_loss: 1.2211 - val_accuracy: 0.6723\n",
            "Epoch 85/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5070 - accuracy: 0.8506 - val_loss: 1.2986 - val_accuracy: 0.6712\n",
            "Epoch 86/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5278 - accuracy: 0.8459 - val_loss: 1.2124 - val_accuracy: 0.6836\n",
            "Epoch 87/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.5196 - accuracy: 0.8470 - val_loss: 1.3773 - val_accuracy: 0.6689\n",
            "Epoch 88/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4977 - accuracy: 0.8514 - val_loss: 1.1748 - val_accuracy: 0.6937\n",
            "Epoch 89/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4920 - accuracy: 0.8557 - val_loss: 1.3444 - val_accuracy: 0.7173\n",
            "Epoch 90/100\n",
            "246/245 [==============================] - 10s 43ms/step - loss: 0.4881 - accuracy: 0.8600 - val_loss: 1.0937 - val_accuracy: 0.7185\n",
            "Epoch 91/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.5027 - accuracy: 0.8557 - val_loss: 1.1227 - val_accuracy: 0.7027\n",
            "Epoch 92/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4977 - accuracy: 0.8543 - val_loss: 1.1451 - val_accuracy: 0.6791\n",
            "Epoch 93/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.4791 - accuracy: 0.8656 - val_loss: 1.2961 - val_accuracy: 0.7005\n",
            "Epoch 94/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4985 - accuracy: 0.8586 - val_loss: 1.2999 - val_accuracy: 0.6588\n",
            "Epoch 95/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4986 - accuracy: 0.8591 - val_loss: 1.1046 - val_accuracy: 0.6723\n",
            "Epoch 96/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4996 - accuracy: 0.8565 - val_loss: 1.1392 - val_accuracy: 0.6914\n",
            "Epoch 97/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4819 - accuracy: 0.8591 - val_loss: 1.1635 - val_accuracy: 0.6914\n",
            "Epoch 98/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4884 - accuracy: 0.8609 - val_loss: 1.2265 - val_accuracy: 0.6745\n",
            "Epoch 99/100\n",
            "246/245 [==============================] - 10s 41ms/step - loss: 0.4624 - accuracy: 0.8707 - val_loss: 1.1951 - val_accuracy: 0.6982\n",
            "Epoch 100/100\n",
            "246/245 [==============================] - 10s 42ms/step - loss: 0.4709 - accuracy: 0.8627 - val_loss: 1.2082 - val_accuracy: 0.6723\n",
            "Training completed in time:  0:16:55.076367 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 406
        },
        "id": "92UmreoEWlWG",
        "outputId": "915b2159-0637-4f25-ea9b-9b70bb6b25be"
      },
      "source": [
        "show_results(history2)"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 71.8468 %\n",
            "\tMin validation loss: 0.99784\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LYj732yqWo4T"
      },
      "source": [
        "### fold-3 <a name=\"fold-3\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xpcna5HBWoX3",
        "outputId": "b89254f0-4a7a-4b76-fb05-da40e6846c4a"
      },
      "source": [
        "FOLD_K = 3\n",
        "REPEAT = 1\n",
        "\n",
        "history3 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history3.append(history)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 10.1622%\n",
            "\n",
            "Epoch 1/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 2.1087 - accuracy: 0.2158 - val_loss: 1.9886 - val_accuracy: 0.2335\n",
            "Epoch 2/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.8496 - accuracy: 0.3069 - val_loss: 1.8190 - val_accuracy: 0.3859\n",
            "Epoch 3/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.7104 - accuracy: 0.3729 - val_loss: 1.7063 - val_accuracy: 0.4216\n",
            "Epoch 4/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.6095 - accuracy: 0.4189 - val_loss: 1.5826 - val_accuracy: 0.4703\n",
            "Epoch 5/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 1.5111 - accuracy: 0.4542 - val_loss: 1.5241 - val_accuracy: 0.4649\n",
            "Epoch 6/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.4461 - accuracy: 0.4862 - val_loss: 1.3918 - val_accuracy: 0.5362\n",
            "Epoch 7/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.3661 - accuracy: 0.5135 - val_loss: 1.4522 - val_accuracy: 0.4876\n",
            "Epoch 8/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.3210 - accuracy: 0.5204 - val_loss: 1.3596 - val_accuracy: 0.5081\n",
            "Epoch 9/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.3027 - accuracy: 0.5329 - val_loss: 1.2061 - val_accuracy: 0.5416\n",
            "Epoch 10/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.2419 - accuracy: 0.5587 - val_loss: 1.3494 - val_accuracy: 0.5049\n",
            "Epoch 11/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.2127 - accuracy: 0.5680 - val_loss: 1.2823 - val_accuracy: 0.5351\n",
            "Epoch 12/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.1781 - accuracy: 0.5846 - val_loss: 1.1295 - val_accuracy: 0.6119\n",
            "Epoch 13/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.1512 - accuracy: 0.5986 - val_loss: 1.1534 - val_accuracy: 0.5859\n",
            "Epoch 14/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.1408 - accuracy: 0.5977 - val_loss: 1.2686 - val_accuracy: 0.5784\n",
            "Epoch 15/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.1208 - accuracy: 0.6111 - val_loss: 1.1931 - val_accuracy: 0.5762\n",
            "Epoch 16/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.0931 - accuracy: 0.6228 - val_loss: 1.1099 - val_accuracy: 0.5957\n",
            "Epoch 17/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.0670 - accuracy: 0.6344 - val_loss: 1.1895 - val_accuracy: 0.5276\n",
            "Epoch 18/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.0580 - accuracy: 0.6458 - val_loss: 1.0548 - val_accuracy: 0.6519\n",
            "Epoch 19/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 1.0264 - accuracy: 0.6494 - val_loss: 1.0914 - val_accuracy: 0.6270\n",
            "Epoch 20/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9919 - accuracy: 0.6635 - val_loss: 1.1031 - val_accuracy: 0.6249\n",
            "Epoch 21/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.9954 - accuracy: 0.6666 - val_loss: 1.1779 - val_accuracy: 0.6032\n",
            "Epoch 22/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.9762 - accuracy: 0.6693 - val_loss: 1.1882 - val_accuracy: 0.5914\n",
            "Epoch 23/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9480 - accuracy: 0.6884 - val_loss: 1.1643 - val_accuracy: 0.6216\n",
            "Epoch 24/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.9349 - accuracy: 0.6901 - val_loss: 1.0743 - val_accuracy: 0.6292\n",
            "Epoch 25/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.9422 - accuracy: 0.6891 - val_loss: 1.2812 - val_accuracy: 0.5611\n",
            "Epoch 26/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.9100 - accuracy: 0.7004 - val_loss: 1.1268 - val_accuracy: 0.6595\n",
            "Epoch 27/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9109 - accuracy: 0.7028 - val_loss: 1.0566 - val_accuracy: 0.6378\n",
            "Epoch 28/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.8738 - accuracy: 0.7168 - val_loss: 1.0965 - val_accuracy: 0.6324\n",
            "Epoch 29/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.8606 - accuracy: 0.7255 - val_loss: 1.0888 - val_accuracy: 0.6703\n",
            "Epoch 30/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8455 - accuracy: 0.7309 - val_loss: 1.1482 - val_accuracy: 0.6432\n",
            "Epoch 31/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.8305 - accuracy: 0.7365 - val_loss: 1.1039 - val_accuracy: 0.6724\n",
            "Epoch 32/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8010 - accuracy: 0.7398 - val_loss: 1.2395 - val_accuracy: 0.6086\n",
            "Epoch 33/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.8089 - accuracy: 0.7401 - val_loss: 1.2442 - val_accuracy: 0.6432\n",
            "Epoch 34/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7971 - accuracy: 0.7466 - val_loss: 1.1420 - val_accuracy: 0.6800\n",
            "Epoch 35/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.7885 - accuracy: 0.7552 - val_loss: 1.1616 - val_accuracy: 0.6919\n",
            "Epoch 36/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.7715 - accuracy: 0.7575 - val_loss: 1.1772 - val_accuracy: 0.6703\n",
            "Epoch 37/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7586 - accuracy: 0.7635 - val_loss: 1.1522 - val_accuracy: 0.6778\n",
            "Epoch 38/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7447 - accuracy: 0.7711 - val_loss: 1.3429 - val_accuracy: 0.6119\n",
            "Epoch 39/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7516 - accuracy: 0.7644 - val_loss: 1.2088 - val_accuracy: 0.6811\n",
            "Epoch 40/100\n",
            "244/243 [==============================] - 11s 43ms/step - loss: 0.7285 - accuracy: 0.7767 - val_loss: 1.3336 - val_accuracy: 0.6605\n",
            "Epoch 41/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.7324 - accuracy: 0.7725 - val_loss: 1.1645 - val_accuracy: 0.6897\n",
            "Epoch 42/100\n",
            "244/243 [==============================] - 11s 43ms/step - loss: 0.7124 - accuracy: 0.7835 - val_loss: 1.2928 - val_accuracy: 0.6951\n",
            "Epoch 43/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6971 - accuracy: 0.7846 - val_loss: 1.2375 - val_accuracy: 0.6584\n",
            "Epoch 44/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7056 - accuracy: 0.7856 - val_loss: 1.2111 - val_accuracy: 0.6703\n",
            "Epoch 45/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.6894 - accuracy: 0.7951 - val_loss: 1.2988 - val_accuracy: 0.6692\n",
            "Epoch 46/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6874 - accuracy: 0.7916 - val_loss: 1.0915 - val_accuracy: 0.6951\n",
            "Epoch 47/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6803 - accuracy: 0.8003 - val_loss: 1.3543 - val_accuracy: 0.6757\n",
            "Epoch 48/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6711 - accuracy: 0.7992 - val_loss: 1.3109 - val_accuracy: 0.6368\n",
            "Epoch 49/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.6639 - accuracy: 0.7965 - val_loss: 1.3847 - val_accuracy: 0.6249\n",
            "Epoch 50/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6469 - accuracy: 0.8050 - val_loss: 1.4240 - val_accuracy: 0.6076\n",
            "Epoch 51/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6497 - accuracy: 0.8057 - val_loss: 1.3403 - val_accuracy: 0.6497\n",
            "Epoch 52/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6474 - accuracy: 0.8075 - val_loss: 1.2913 - val_accuracy: 0.6562\n",
            "Epoch 53/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.6253 - accuracy: 0.8147 - val_loss: 1.4540 - val_accuracy: 0.6378\n",
            "Epoch 54/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.6384 - accuracy: 0.8111 - val_loss: 1.3687 - val_accuracy: 0.6249\n",
            "Epoch 55/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.6116 - accuracy: 0.8230 - val_loss: 1.2785 - val_accuracy: 0.6605\n",
            "Epoch 56/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6193 - accuracy: 0.8176 - val_loss: 1.3752 - val_accuracy: 0.6659\n",
            "Epoch 57/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6059 - accuracy: 0.8254 - val_loss: 1.3854 - val_accuracy: 0.6595\n",
            "Epoch 58/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5971 - accuracy: 0.8285 - val_loss: 1.5343 - val_accuracy: 0.6757\n",
            "Epoch 59/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.6009 - accuracy: 0.8220 - val_loss: 1.3232 - val_accuracy: 0.6584\n",
            "Epoch 60/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.5796 - accuracy: 0.8313 - val_loss: 1.3633 - val_accuracy: 0.6692\n",
            "Epoch 61/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.5676 - accuracy: 0.8341 - val_loss: 1.3862 - val_accuracy: 0.6519\n",
            "Epoch 62/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.5765 - accuracy: 0.8340 - val_loss: 1.3761 - val_accuracy: 0.6530\n",
            "Epoch 63/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5632 - accuracy: 0.8354 - val_loss: 1.3199 - val_accuracy: 0.6692\n",
            "Epoch 64/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5679 - accuracy: 0.8410 - val_loss: 1.4426 - val_accuracy: 0.6486\n",
            "Epoch 65/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5448 - accuracy: 0.8444 - val_loss: 1.3963 - val_accuracy: 0.6486\n",
            "Epoch 66/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.5427 - accuracy: 0.8422 - val_loss: 1.3113 - val_accuracy: 0.6714\n",
            "Epoch 67/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5570 - accuracy: 0.8439 - val_loss: 1.4404 - val_accuracy: 0.6378\n",
            "Epoch 68/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5377 - accuracy: 0.8450 - val_loss: 1.4682 - val_accuracy: 0.6411\n",
            "Epoch 69/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5603 - accuracy: 0.8384 - val_loss: 1.4894 - val_accuracy: 0.6443\n",
            "Epoch 70/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.5374 - accuracy: 0.8460 - val_loss: 1.3675 - val_accuracy: 0.6530\n",
            "Epoch 71/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5304 - accuracy: 0.8517 - val_loss: 1.3728 - val_accuracy: 0.6411\n",
            "Epoch 72/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5137 - accuracy: 0.8522 - val_loss: 1.3697 - val_accuracy: 0.6811\n",
            "Epoch 73/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5186 - accuracy: 0.8492 - val_loss: 1.4723 - val_accuracy: 0.6605\n",
            "Epoch 74/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5228 - accuracy: 0.8487 - val_loss: 1.5396 - val_accuracy: 0.6422\n",
            "Epoch 75/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5259 - accuracy: 0.8494 - val_loss: 1.3969 - val_accuracy: 0.6735\n",
            "Epoch 76/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5193 - accuracy: 0.8522 - val_loss: 1.4223 - val_accuracy: 0.6638\n",
            "Epoch 77/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5127 - accuracy: 0.8564 - val_loss: 1.6386 - val_accuracy: 0.6378\n",
            "Epoch 78/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4908 - accuracy: 0.8626 - val_loss: 1.5469 - val_accuracy: 0.6259\n",
            "Epoch 79/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5036 - accuracy: 0.8564 - val_loss: 1.5939 - val_accuracy: 0.6378\n",
            "Epoch 80/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5070 - accuracy: 0.8496 - val_loss: 1.4693 - val_accuracy: 0.6573\n",
            "Epoch 81/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.4822 - accuracy: 0.8610 - val_loss: 1.4697 - val_accuracy: 0.6595\n",
            "Epoch 82/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4764 - accuracy: 0.8654 - val_loss: 1.3582 - val_accuracy: 0.6692\n",
            "Epoch 83/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4882 - accuracy: 0.8622 - val_loss: 1.2220 - val_accuracy: 0.6832\n",
            "Epoch 84/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4907 - accuracy: 0.8608 - val_loss: 1.3527 - val_accuracy: 0.6724\n",
            "Epoch 85/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4858 - accuracy: 0.8649 - val_loss: 1.5343 - val_accuracy: 0.6432\n",
            "Epoch 86/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4773 - accuracy: 0.8652 - val_loss: 1.4053 - val_accuracy: 0.6800\n",
            "Epoch 87/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4758 - accuracy: 0.8632 - val_loss: 1.4220 - val_accuracy: 0.6638\n",
            "Epoch 88/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4764 - accuracy: 0.8626 - val_loss: 1.6382 - val_accuracy: 0.6292\n",
            "Epoch 89/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4706 - accuracy: 0.8681 - val_loss: 1.6100 - val_accuracy: 0.6422\n",
            "Epoch 90/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4633 - accuracy: 0.8705 - val_loss: 1.4670 - val_accuracy: 0.6486\n",
            "Epoch 91/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4475 - accuracy: 0.8724 - val_loss: 1.4290 - val_accuracy: 0.6454\n",
            "Epoch 92/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4648 - accuracy: 0.8691 - val_loss: 1.4759 - val_accuracy: 0.6508\n",
            "Epoch 93/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4701 - accuracy: 0.8664 - val_loss: 1.4020 - val_accuracy: 0.6659\n",
            "Epoch 94/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4523 - accuracy: 0.8718 - val_loss: 1.5453 - val_accuracy: 0.6476\n",
            "Epoch 95/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4511 - accuracy: 0.8713 - val_loss: 1.5486 - val_accuracy: 0.6573\n",
            "Epoch 96/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4342 - accuracy: 0.8793 - val_loss: 1.7468 - val_accuracy: 0.6324\n",
            "Epoch 97/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4471 - accuracy: 0.8782 - val_loss: 1.6119 - val_accuracy: 0.6530\n",
            "Epoch 98/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4426 - accuracy: 0.8715 - val_loss: 1.5363 - val_accuracy: 0.6205\n",
            "Epoch 99/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4375 - accuracy: 0.8769 - val_loss: 1.6744 - val_accuracy: 0.6368\n",
            "Epoch 100/100\n",
            "244/243 [==============================] - 10s 41ms/step - loss: 0.4404 - accuracy: 0.8754 - val_loss: 1.5067 - val_accuracy: 0.6519\n",
            "Training completed in time:  0:17:00.595198 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "id": "VUViHcdaWzQg",
        "outputId": "c3c60e79-c0b9-40a5-f41c-5a97d4fca9ef"
      },
      "source": [
        "show_results(history3)"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 69.5135 %\n",
            "\tMin validation loss: 1.05481\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tO1Fmd8QW1rv"
      },
      "source": [
        "### fold-4 <a name=\"fold-4\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JBNj_UkKW1Jh",
        "outputId": "b6ac3304-8a34-40c1-82aa-d6c2efa96494"
      },
      "source": [
        "FOLD_K = 4\n",
        "REPEAT = 1\n",
        "\n",
        "history4 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history4.append(history)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 10.8081%\n",
            "\n",
            "Epoch 1/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 2.0681 - accuracy: 0.2235 - val_loss: 1.9486 - val_accuracy: 0.2596\n",
            "Epoch 2/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.8538 - accuracy: 0.2838 - val_loss: 1.7600 - val_accuracy: 0.3707\n",
            "Epoch 3/100\n",
            "242/241 [==============================] - 10s 43ms/step - loss: 1.7129 - accuracy: 0.3599 - val_loss: 1.7088 - val_accuracy: 0.4313\n",
            "Epoch 4/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 1.5875 - accuracy: 0.4186 - val_loss: 1.5224 - val_accuracy: 0.5020\n",
            "Epoch 5/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.4594 - accuracy: 0.4664 - val_loss: 1.5201 - val_accuracy: 0.4828\n",
            "Epoch 6/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.3928 - accuracy: 0.4978 - val_loss: 1.5095 - val_accuracy: 0.5293\n",
            "Epoch 7/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.3190 - accuracy: 0.5220 - val_loss: 1.3643 - val_accuracy: 0.5444\n",
            "Epoch 8/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.2855 - accuracy: 0.5409 - val_loss: 1.3121 - val_accuracy: 0.5778\n",
            "Epoch 9/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.2387 - accuracy: 0.5665 - val_loss: 1.3712 - val_accuracy: 0.5566\n",
            "Epoch 10/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.2133 - accuracy: 0.5726 - val_loss: 1.2890 - val_accuracy: 0.5879\n",
            "Epoch 11/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 1.1829 - accuracy: 0.5803 - val_loss: 1.3383 - val_accuracy: 0.5909\n",
            "Epoch 12/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.1462 - accuracy: 0.5988 - val_loss: 1.2381 - val_accuracy: 0.6121\n",
            "Epoch 13/100\n",
            "242/241 [==============================] - 10s 40ms/step - loss: 1.1222 - accuracy: 0.6057 - val_loss: 1.2951 - val_accuracy: 0.6283\n",
            "Epoch 14/100\n",
            "242/241 [==============================] - 10s 40ms/step - loss: 1.0916 - accuracy: 0.6218 - val_loss: 1.2341 - val_accuracy: 0.6051\n",
            "Epoch 15/100\n",
            "242/241 [==============================] - 10s 40ms/step - loss: 1.0787 - accuracy: 0.6258 - val_loss: 1.3139 - val_accuracy: 0.6364\n",
            "Epoch 16/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.0746 - accuracy: 0.6285 - val_loss: 1.2702 - val_accuracy: 0.6131\n",
            "Epoch 17/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.0444 - accuracy: 0.6400 - val_loss: 1.2356 - val_accuracy: 0.6182\n",
            "Epoch 18/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 1.0216 - accuracy: 0.6474 - val_loss: 1.2765 - val_accuracy: 0.6000\n",
            "Epoch 19/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.9956 - accuracy: 0.6571 - val_loss: 1.3245 - val_accuracy: 0.5737\n",
            "Epoch 20/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.9947 - accuracy: 0.6633 - val_loss: 1.2595 - val_accuracy: 0.5697\n",
            "Epoch 21/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.9681 - accuracy: 0.6656 - val_loss: 1.3418 - val_accuracy: 0.6313\n",
            "Epoch 22/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.9535 - accuracy: 0.6821 - val_loss: 1.3586 - val_accuracy: 0.5768\n",
            "Epoch 23/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.9192 - accuracy: 0.6930 - val_loss: 1.2519 - val_accuracy: 0.6212\n",
            "Epoch 24/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.9267 - accuracy: 0.6912 - val_loss: 1.1547 - val_accuracy: 0.6535\n",
            "Epoch 25/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.8854 - accuracy: 0.7076 - val_loss: 1.3365 - val_accuracy: 0.6485\n",
            "Epoch 26/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.8766 - accuracy: 0.7123 - val_loss: 1.3757 - val_accuracy: 0.6152\n",
            "Epoch 27/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.8616 - accuracy: 0.7118 - val_loss: 1.3441 - val_accuracy: 0.6020\n",
            "Epoch 28/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.8373 - accuracy: 0.7249 - val_loss: 1.4633 - val_accuracy: 0.5758\n",
            "Epoch 29/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.8642 - accuracy: 0.7142 - val_loss: 1.1966 - val_accuracy: 0.6455\n",
            "Epoch 30/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.8365 - accuracy: 0.7245 - val_loss: 1.1594 - val_accuracy: 0.6606\n",
            "Epoch 31/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.8339 - accuracy: 0.7294 - val_loss: 1.3718 - val_accuracy: 0.5899\n",
            "Epoch 32/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.8098 - accuracy: 0.7366 - val_loss: 1.2441 - val_accuracy: 0.6343\n",
            "Epoch 33/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7910 - accuracy: 0.7474 - val_loss: 1.2093 - val_accuracy: 0.6495\n",
            "Epoch 34/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7956 - accuracy: 0.7510 - val_loss: 1.2802 - val_accuracy: 0.6586\n",
            "Epoch 35/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7758 - accuracy: 0.7506 - val_loss: 1.2730 - val_accuracy: 0.6475\n",
            "Epoch 36/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7848 - accuracy: 0.7548 - val_loss: 1.1914 - val_accuracy: 0.6404\n",
            "Epoch 37/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7716 - accuracy: 0.7590 - val_loss: 1.1875 - val_accuracy: 0.6687\n",
            "Epoch 38/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7467 - accuracy: 0.7702 - val_loss: 1.2875 - val_accuracy: 0.6475\n",
            "Epoch 39/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7497 - accuracy: 0.7595 - val_loss: 1.3528 - val_accuracy: 0.6374\n",
            "Epoch 40/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7475 - accuracy: 0.7647 - val_loss: 1.4073 - val_accuracy: 0.6081\n",
            "Epoch 41/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7127 - accuracy: 0.7782 - val_loss: 1.3751 - val_accuracy: 0.6374\n",
            "Epoch 42/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.7023 - accuracy: 0.7815 - val_loss: 1.2545 - val_accuracy: 0.6596\n",
            "Epoch 43/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.7048 - accuracy: 0.7776 - val_loss: 1.3754 - val_accuracy: 0.6566\n",
            "Epoch 44/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6927 - accuracy: 0.7852 - val_loss: 1.4066 - val_accuracy: 0.6384\n",
            "Epoch 45/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6965 - accuracy: 0.7867 - val_loss: 1.2392 - val_accuracy: 0.6646\n",
            "Epoch 46/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6905 - accuracy: 0.7902 - val_loss: 1.3876 - val_accuracy: 0.6263\n",
            "Epoch 47/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6595 - accuracy: 0.7985 - val_loss: 1.1335 - val_accuracy: 0.6707\n",
            "Epoch 48/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6691 - accuracy: 0.7927 - val_loss: 1.2727 - val_accuracy: 0.6525\n",
            "Epoch 49/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6479 - accuracy: 0.8069 - val_loss: 1.4083 - val_accuracy: 0.6667\n",
            "Epoch 50/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6661 - accuracy: 0.7924 - val_loss: 1.2940 - val_accuracy: 0.6828\n",
            "Epoch 51/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.6396 - accuracy: 0.8088 - val_loss: 1.4559 - val_accuracy: 0.6293\n",
            "Epoch 52/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6449 - accuracy: 0.8043 - val_loss: 1.4028 - val_accuracy: 0.6333\n",
            "Epoch 53/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6223 - accuracy: 0.8099 - val_loss: 1.4144 - val_accuracy: 0.6586\n",
            "Epoch 54/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6098 - accuracy: 0.8131 - val_loss: 1.3602 - val_accuracy: 0.6414\n",
            "Epoch 55/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.6202 - accuracy: 0.8137 - val_loss: 1.5041 - val_accuracy: 0.6515\n",
            "Epoch 56/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6121 - accuracy: 0.8130 - val_loss: 1.5057 - val_accuracy: 0.6384\n",
            "Epoch 57/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6106 - accuracy: 0.8183 - val_loss: 1.3357 - val_accuracy: 0.6687\n",
            "Epoch 58/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5841 - accuracy: 0.8252 - val_loss: 1.4668 - val_accuracy: 0.6303\n",
            "Epoch 59/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.6067 - accuracy: 0.8163 - val_loss: 1.5641 - val_accuracy: 0.6263\n",
            "Epoch 60/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.6033 - accuracy: 0.8220 - val_loss: 1.5690 - val_accuracy: 0.6162\n",
            "Epoch 61/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5868 - accuracy: 0.8268 - val_loss: 1.4387 - val_accuracy: 0.6556\n",
            "Epoch 62/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5889 - accuracy: 0.8216 - val_loss: 1.4132 - val_accuracy: 0.6646\n",
            "Epoch 63/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5737 - accuracy: 0.8312 - val_loss: 1.4114 - val_accuracy: 0.6737\n",
            "Epoch 64/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5629 - accuracy: 0.8357 - val_loss: 1.4127 - val_accuracy: 0.6566\n",
            "Epoch 65/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5706 - accuracy: 0.8349 - val_loss: 1.3928 - val_accuracy: 0.6596\n",
            "Epoch 66/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5570 - accuracy: 0.8373 - val_loss: 1.5172 - val_accuracy: 0.6586\n",
            "Epoch 67/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5620 - accuracy: 0.8318 - val_loss: 1.4354 - val_accuracy: 0.6535\n",
            "Epoch 68/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5401 - accuracy: 0.8411 - val_loss: 1.5214 - val_accuracy: 0.6374\n",
            "Epoch 69/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5582 - accuracy: 0.8385 - val_loss: 1.6828 - val_accuracy: 0.6374\n",
            "Epoch 70/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5386 - accuracy: 0.8425 - val_loss: 1.5548 - val_accuracy: 0.6313\n",
            "Epoch 71/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5478 - accuracy: 0.8375 - val_loss: 1.4148 - val_accuracy: 0.6717\n",
            "Epoch 72/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5348 - accuracy: 0.8460 - val_loss: 1.5248 - val_accuracy: 0.6091\n",
            "Epoch 73/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5147 - accuracy: 0.8478 - val_loss: 1.5821 - val_accuracy: 0.6545\n",
            "Epoch 74/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5318 - accuracy: 0.8437 - val_loss: 1.6005 - val_accuracy: 0.6465\n",
            "Epoch 75/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5155 - accuracy: 0.8490 - val_loss: 1.4683 - val_accuracy: 0.6192\n",
            "Epoch 76/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5095 - accuracy: 0.8473 - val_loss: 1.5303 - val_accuracy: 0.6505\n",
            "Epoch 77/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5071 - accuracy: 0.8516 - val_loss: 1.6126 - val_accuracy: 0.6384\n",
            "Epoch 78/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.5089 - accuracy: 0.8487 - val_loss: 1.5445 - val_accuracy: 0.6677\n",
            "Epoch 79/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.5118 - accuracy: 0.8500 - val_loss: 1.6584 - val_accuracy: 0.6192\n",
            "Epoch 80/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.4966 - accuracy: 0.8548 - val_loss: 1.5007 - val_accuracy: 0.6616\n",
            "Epoch 81/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.4881 - accuracy: 0.8565 - val_loss: 1.7398 - val_accuracy: 0.6525\n",
            "Epoch 82/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4914 - accuracy: 0.8580 - val_loss: 1.6850 - val_accuracy: 0.6677\n",
            "Epoch 83/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4862 - accuracy: 0.8606 - val_loss: 1.5287 - val_accuracy: 0.6949\n",
            "Epoch 84/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.4880 - accuracy: 0.8586 - val_loss: 1.5300 - val_accuracy: 0.6778\n",
            "Epoch 85/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4924 - accuracy: 0.8575 - val_loss: 1.6992 - val_accuracy: 0.6525\n",
            "Epoch 86/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.4675 - accuracy: 0.8628 - val_loss: 1.6509 - val_accuracy: 0.6424\n",
            "Epoch 87/100\n",
            "242/241 [==============================] - 10s 41ms/step - loss: 0.4862 - accuracy: 0.8577 - val_loss: 1.6931 - val_accuracy: 0.6657\n",
            "Epoch 88/100\n",
            "242/241 [==============================] - 10s 43ms/step - loss: 0.4577 - accuracy: 0.8676 - val_loss: 1.6055 - val_accuracy: 0.6737\n",
            "Epoch 89/100\n",
            "242/241 [==============================] - 10s 43ms/step - loss: 0.4711 - accuracy: 0.8626 - val_loss: 1.7761 - val_accuracy: 0.6424\n",
            "Epoch 90/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4763 - accuracy: 0.8614 - val_loss: 1.7864 - val_accuracy: 0.6929\n",
            "Epoch 91/100\n",
            "242/241 [==============================] - 10s 43ms/step - loss: 0.4734 - accuracy: 0.8644 - val_loss: 1.5354 - val_accuracy: 0.6879\n",
            "Epoch 92/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4635 - accuracy: 0.8646 - val_loss: 1.5652 - val_accuracy: 0.6556\n",
            "Epoch 93/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4599 - accuracy: 0.8652 - val_loss: 1.8321 - val_accuracy: 0.6404\n",
            "Epoch 94/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4639 - accuracy: 0.8645 - val_loss: 1.5959 - val_accuracy: 0.6606\n",
            "Epoch 95/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4504 - accuracy: 0.8699 - val_loss: 1.8920 - val_accuracy: 0.6687\n",
            "Epoch 96/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4523 - accuracy: 0.8708 - val_loss: 1.8129 - val_accuracy: 0.6424\n",
            "Epoch 97/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4472 - accuracy: 0.8708 - val_loss: 1.6394 - val_accuracy: 0.6737\n",
            "Epoch 98/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4620 - accuracy: 0.8684 - val_loss: 1.7850 - val_accuracy: 0.6586\n",
            "Epoch 99/100\n",
            "242/241 [==============================] - 10s 43ms/step - loss: 0.4431 - accuracy: 0.8714 - val_loss: 1.6254 - val_accuracy: 0.6576\n",
            "Epoch 100/100\n",
            "242/241 [==============================] - 10s 42ms/step - loss: 0.4496 - accuracy: 0.8734 - val_loss: 1.6005 - val_accuracy: 0.7010\n",
            "Training completed in time:  0:16:48.162298 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "id": "ZtheeNUNW7fb",
        "outputId": "4bbfc863-00f3-4f25-90cc-c4520cc371fb"
      },
      "source": [
        "show_results(history4)"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 70.1010 %\n",
            "\tMin validation loss: 1.13355\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vW_27wyZW-IC"
      },
      "source": [
        "### fold-5 <a name=\"fold-5\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cC2yy-cCW9sC",
        "outputId": "3174d17e-c051-4c73-9854-c61d0ed6c8b8"
      },
      "source": [
        "FOLD_K = 5\n",
        "REPEAT = 1\n",
        "\n",
        "history5 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history5.append(history)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 14.9573%\n",
            "\n",
            "Epoch 1/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 2.0727 - accuracy: 0.2224 - val_loss: 1.9841 - val_accuracy: 0.2650\n",
            "Epoch 2/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 1.8736 - accuracy: 0.2735 - val_loss: 1.8616 - val_accuracy: 0.3632\n",
            "Epoch 3/100\n",
            "244/243 [==============================] - 11s 45ms/step - loss: 1.7510 - accuracy: 0.3456 - val_loss: 1.6841 - val_accuracy: 0.4049\n",
            "Epoch 4/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 1.6294 - accuracy: 0.4025 - val_loss: 1.4871 - val_accuracy: 0.4156\n",
            "Epoch 5/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.5093 - accuracy: 0.4541 - val_loss: 1.5251 - val_accuracy: 0.4594\n",
            "Epoch 6/100\n",
            "244/243 [==============================] - 11s 43ms/step - loss: 1.4106 - accuracy: 0.4964 - val_loss: 1.3159 - val_accuracy: 0.4850\n",
            "Epoch 7/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.3756 - accuracy: 0.5083 - val_loss: 1.2424 - val_accuracy: 0.4882\n",
            "Epoch 8/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.3102 - accuracy: 0.5419 - val_loss: 1.2526 - val_accuracy: 0.5021\n",
            "Epoch 9/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.2685 - accuracy: 0.5564 - val_loss: 1.2779 - val_accuracy: 0.5182\n",
            "Epoch 10/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.2191 - accuracy: 0.5682 - val_loss: 1.1571 - val_accuracy: 0.5876\n",
            "Epoch 11/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 1.1996 - accuracy: 0.5798 - val_loss: 1.0961 - val_accuracy: 0.6100\n",
            "Epoch 12/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 1.1845 - accuracy: 0.5865 - val_loss: 1.1791 - val_accuracy: 0.5620\n",
            "Epoch 13/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.1378 - accuracy: 0.6062 - val_loss: 1.0493 - val_accuracy: 0.5855\n",
            "Epoch 14/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 1.1179 - accuracy: 0.6162 - val_loss: 1.0456 - val_accuracy: 0.6442\n",
            "Epoch 15/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.0945 - accuracy: 0.6199 - val_loss: 1.0302 - val_accuracy: 0.6175\n",
            "Epoch 16/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.0776 - accuracy: 0.6310 - val_loss: 1.0192 - val_accuracy: 0.6389\n",
            "Epoch 17/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 1.0528 - accuracy: 0.6434 - val_loss: 0.9695 - val_accuracy: 0.6442\n",
            "Epoch 18/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.0272 - accuracy: 0.6601 - val_loss: 0.9506 - val_accuracy: 0.6613\n",
            "Epoch 19/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 1.0086 - accuracy: 0.6570 - val_loss: 1.0140 - val_accuracy: 0.6432\n",
            "Epoch 20/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9926 - accuracy: 0.6655 - val_loss: 1.0251 - val_accuracy: 0.6239\n",
            "Epoch 21/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9862 - accuracy: 0.6687 - val_loss: 1.0236 - val_accuracy: 0.6453\n",
            "Epoch 22/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9677 - accuracy: 0.6775 - val_loss: 0.9917 - val_accuracy: 0.6303\n",
            "Epoch 23/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9611 - accuracy: 0.6807 - val_loss: 0.9362 - val_accuracy: 0.6741\n",
            "Epoch 24/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9322 - accuracy: 0.6896 - val_loss: 0.9129 - val_accuracy: 0.6923\n",
            "Epoch 25/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9109 - accuracy: 0.7050 - val_loss: 0.9425 - val_accuracy: 0.6848\n",
            "Epoch 26/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.9047 - accuracy: 0.6987 - val_loss: 0.8698 - val_accuracy: 0.7051\n",
            "Epoch 27/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8986 - accuracy: 0.7074 - val_loss: 0.9101 - val_accuracy: 0.6667\n",
            "Epoch 28/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8848 - accuracy: 0.7105 - val_loss: 0.8973 - val_accuracy: 0.6870\n",
            "Epoch 29/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8708 - accuracy: 0.7201 - val_loss: 0.8652 - val_accuracy: 0.7051\n",
            "Epoch 30/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8553 - accuracy: 0.7177 - val_loss: 0.8593 - val_accuracy: 0.6923\n",
            "Epoch 31/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8439 - accuracy: 0.7229 - val_loss: 0.8521 - val_accuracy: 0.7639\n",
            "Epoch 32/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8416 - accuracy: 0.7276 - val_loss: 0.8562 - val_accuracy: 0.6944\n",
            "Epoch 33/100\n",
            "244/243 [==============================] - 11s 43ms/step - loss: 0.8411 - accuracy: 0.7331 - val_loss: 0.9357 - val_accuracy: 0.6827\n",
            "Epoch 34/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.8020 - accuracy: 0.7454 - val_loss: 0.8704 - val_accuracy: 0.7094\n",
            "Epoch 35/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8103 - accuracy: 0.7429 - val_loss: 0.8522 - val_accuracy: 0.7019\n",
            "Epoch 36/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.8081 - accuracy: 0.7420 - val_loss: 0.9715 - val_accuracy: 0.6774\n",
            "Epoch 37/100\n",
            "244/243 [==============================] - 11s 43ms/step - loss: 0.7843 - accuracy: 0.7538 - val_loss: 0.8328 - val_accuracy: 0.7564\n",
            "Epoch 38/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.7637 - accuracy: 0.7518 - val_loss: 0.8373 - val_accuracy: 0.7404\n",
            "Epoch 39/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7855 - accuracy: 0.7476 - val_loss: 0.8371 - val_accuracy: 0.7361\n",
            "Epoch 40/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7598 - accuracy: 0.7601 - val_loss: 0.8392 - val_accuracy: 0.7521\n",
            "Epoch 41/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7528 - accuracy: 0.7623 - val_loss: 0.7780 - val_accuracy: 0.7479\n",
            "Epoch 42/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7242 - accuracy: 0.7757 - val_loss: 0.8428 - val_accuracy: 0.7532\n",
            "Epoch 43/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.7249 - accuracy: 0.7730 - val_loss: 0.8016 - val_accuracy: 0.7212\n",
            "Epoch 44/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.7420 - accuracy: 0.7663 - val_loss: 0.8365 - val_accuracy: 0.7425\n",
            "Epoch 45/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.7235 - accuracy: 0.7754 - val_loss: 0.8822 - val_accuracy: 0.7073\n",
            "Epoch 46/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.7059 - accuracy: 0.7818 - val_loss: 0.8472 - val_accuracy: 0.7212\n",
            "Epoch 47/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.7044 - accuracy: 0.7774 - val_loss: 0.8387 - val_accuracy: 0.7062\n",
            "Epoch 48/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6868 - accuracy: 0.7816 - val_loss: 0.9420 - val_accuracy: 0.7073\n",
            "Epoch 49/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6809 - accuracy: 0.7881 - val_loss: 0.8160 - val_accuracy: 0.7500\n",
            "Epoch 50/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6896 - accuracy: 0.7887 - val_loss: 0.8940 - val_accuracy: 0.7073\n",
            "Epoch 51/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6645 - accuracy: 0.7967 - val_loss: 0.8772 - val_accuracy: 0.7671\n",
            "Epoch 52/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.6821 - accuracy: 0.7976 - val_loss: 0.7909 - val_accuracy: 0.7682\n",
            "Epoch 53/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.6627 - accuracy: 0.7986 - val_loss: 0.7758 - val_accuracy: 0.7618\n",
            "Epoch 54/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6619 - accuracy: 0.7980 - val_loss: 0.8132 - val_accuracy: 0.7767\n",
            "Epoch 55/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.6564 - accuracy: 0.7999 - val_loss: 0.7800 - val_accuracy: 0.7703\n",
            "Epoch 56/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.6499 - accuracy: 0.7984 - val_loss: 0.7868 - val_accuracy: 0.7639\n",
            "Epoch 57/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6351 - accuracy: 0.8066 - val_loss: 0.7734 - val_accuracy: 0.7853\n",
            "Epoch 58/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.6248 - accuracy: 0.8127 - val_loss: 0.8586 - val_accuracy: 0.7660\n",
            "Epoch 59/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6231 - accuracy: 0.8079 - val_loss: 0.7408 - val_accuracy: 0.7831\n",
            "Epoch 60/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.6233 - accuracy: 0.8113 - val_loss: 0.8649 - val_accuracy: 0.7660\n",
            "Epoch 61/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6256 - accuracy: 0.8139 - val_loss: 0.9784 - val_accuracy: 0.7190\n",
            "Epoch 62/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.6149 - accuracy: 0.8182 - val_loss: 0.7905 - val_accuracy: 0.7639\n",
            "Epoch 63/100\n",
            "244/243 [==============================] - 11s 43ms/step - loss: 0.6077 - accuracy: 0.8222 - val_loss: 0.7804 - val_accuracy: 0.7831\n",
            "Epoch 64/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6075 - accuracy: 0.8135 - val_loss: 0.8611 - val_accuracy: 0.7479\n",
            "Epoch 65/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.6056 - accuracy: 0.8186 - val_loss: 0.8824 - val_accuracy: 0.7233\n",
            "Epoch 66/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5985 - accuracy: 0.8238 - val_loss: 0.8385 - val_accuracy: 0.7853\n",
            "Epoch 67/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5814 - accuracy: 0.8275 - val_loss: 0.8255 - val_accuracy: 0.7917\n",
            "Epoch 68/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5753 - accuracy: 0.8306 - val_loss: 0.9569 - val_accuracy: 0.7308\n",
            "Epoch 69/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5925 - accuracy: 0.8207 - val_loss: 0.8656 - val_accuracy: 0.7585\n",
            "Epoch 70/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5686 - accuracy: 0.8291 - val_loss: 0.7936 - val_accuracy: 0.7511\n",
            "Epoch 71/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5793 - accuracy: 0.8297 - val_loss: 0.7871 - val_accuracy: 0.7714\n",
            "Epoch 72/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5657 - accuracy: 0.8325 - val_loss: 0.9821 - val_accuracy: 0.7404\n",
            "Epoch 73/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5590 - accuracy: 0.8317 - val_loss: 0.7949 - val_accuracy: 0.7735\n",
            "Epoch 74/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5456 - accuracy: 0.8389 - val_loss: 0.7899 - val_accuracy: 0.7799\n",
            "Epoch 75/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5546 - accuracy: 0.8334 - val_loss: 0.8538 - val_accuracy: 0.7543\n",
            "Epoch 76/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5508 - accuracy: 0.8406 - val_loss: 0.8514 - val_accuracy: 0.7543\n",
            "Epoch 77/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5358 - accuracy: 0.8452 - val_loss: 0.8271 - val_accuracy: 0.7650\n",
            "Epoch 78/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5377 - accuracy: 0.8415 - val_loss: 1.0828 - val_accuracy: 0.6859\n",
            "Epoch 79/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5417 - accuracy: 0.8395 - val_loss: 0.7554 - val_accuracy: 0.7724\n",
            "Epoch 80/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5371 - accuracy: 0.8412 - val_loss: 0.7885 - val_accuracy: 0.7746\n",
            "Epoch 81/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5211 - accuracy: 0.8485 - val_loss: 0.8333 - val_accuracy: 0.7372\n",
            "Epoch 82/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5172 - accuracy: 0.8515 - val_loss: 0.8533 - val_accuracy: 0.7895\n",
            "Epoch 83/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5158 - accuracy: 0.8470 - val_loss: 0.9890 - val_accuracy: 0.7575\n",
            "Epoch 84/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5122 - accuracy: 0.8495 - val_loss: 0.9612 - val_accuracy: 0.7425\n",
            "Epoch 85/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5043 - accuracy: 0.8512 - val_loss: 0.7973 - val_accuracy: 0.7703\n",
            "Epoch 86/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5112 - accuracy: 0.8509 - val_loss: 0.8363 - val_accuracy: 0.7735\n",
            "Epoch 87/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.4899 - accuracy: 0.8583 - val_loss: 0.7864 - val_accuracy: 0.7895\n",
            "Epoch 88/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5030 - accuracy: 0.8535 - val_loss: 0.8043 - val_accuracy: 0.7895\n",
            "Epoch 89/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.5102 - accuracy: 0.8494 - val_loss: 0.8416 - val_accuracy: 0.7618\n",
            "Epoch 90/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4978 - accuracy: 0.8584 - val_loss: 0.8066 - val_accuracy: 0.7564\n",
            "Epoch 91/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.4858 - accuracy: 0.8588 - val_loss: 0.8102 - val_accuracy: 0.7585\n",
            "Epoch 92/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.5039 - accuracy: 0.8577 - val_loss: 0.8576 - val_accuracy: 0.7724\n",
            "Epoch 93/100\n",
            "244/243 [==============================] - 11s 43ms/step - loss: 0.4836 - accuracy: 0.8594 - val_loss: 0.7566 - val_accuracy: 0.8045\n",
            "Epoch 94/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4998 - accuracy: 0.8583 - val_loss: 0.8838 - val_accuracy: 0.7810\n",
            "Epoch 95/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.4710 - accuracy: 0.8629 - val_loss: 0.8109 - val_accuracy: 0.7778\n",
            "Epoch 96/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.4815 - accuracy: 0.8592 - val_loss: 0.9362 - val_accuracy: 0.7628\n",
            "Epoch 97/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.4813 - accuracy: 0.8570 - val_loss: 0.9702 - val_accuracy: 0.7404\n",
            "Epoch 98/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4873 - accuracy: 0.8575 - val_loss: 0.8442 - val_accuracy: 0.7746\n",
            "Epoch 99/100\n",
            "244/243 [==============================] - 10s 43ms/step - loss: 0.4786 - accuracy: 0.8617 - val_loss: 0.8050 - val_accuracy: 0.7853\n",
            "Epoch 100/100\n",
            "244/243 [==============================] - 10s 42ms/step - loss: 0.4740 - accuracy: 0.8628 - val_loss: 0.8434 - val_accuracy: 0.7714\n",
            "Training completed in time:  0:17:24.445097 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "id": "0uQq18p1XMuP",
        "outputId": "7b5a4247-4183-4fc8-998e-4e391ced8131"
      },
      "source": [
        "show_results(history5)"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 80.4487 %\n",
            "\tMin validation loss: 0.74084\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UrLd7ScwXPGN"
      },
      "source": [
        "### fold-6 <a name=\"fold-6\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "miiUaxFLXUNG",
        "outputId": "1c3d0ac9-38fa-49f5-869d-363709db1b7a"
      },
      "source": [
        "FOLD_K = 6\n",
        "REPEAT = 1\n",
        "\n",
        "history6 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history6.append(history)"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 13.0012%\n",
            "\n",
            "Epoch 1/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 2.0831 - accuracy: 0.2235 - val_loss: 1.9002 - val_accuracy: 0.2855\n",
            "Epoch 2/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.9173 - accuracy: 0.2588 - val_loss: 1.8165 - val_accuracy: 0.2989\n",
            "Epoch 3/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.8124 - accuracy: 0.3117 - val_loss: 1.7709 - val_accuracy: 0.3402\n",
            "Epoch 4/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.6966 - accuracy: 0.3523 - val_loss: 1.6361 - val_accuracy: 0.3706\n",
            "Epoch 5/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.6020 - accuracy: 0.3976 - val_loss: 1.5109 - val_accuracy: 0.4265\n",
            "Epoch 6/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 1.5254 - accuracy: 0.4362 - val_loss: 1.3793 - val_accuracy: 0.5176\n",
            "Epoch 7/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.4366 - accuracy: 0.4706 - val_loss: 1.3044 - val_accuracy: 0.5310\n",
            "Epoch 8/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.3646 - accuracy: 0.4998 - val_loss: 1.2692 - val_accuracy: 0.5261\n",
            "Epoch 9/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.3177 - accuracy: 0.5291 - val_loss: 1.3107 - val_accuracy: 0.5030\n",
            "Epoch 10/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.2578 - accuracy: 0.5554 - val_loss: 1.2473 - val_accuracy: 0.5225\n",
            "Epoch 11/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.2281 - accuracy: 0.5629 - val_loss: 1.2438 - val_accuracy: 0.5723\n",
            "Epoch 12/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 1.1808 - accuracy: 0.5787 - val_loss: 1.1231 - val_accuracy: 0.6063\n",
            "Epoch 13/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 1.1514 - accuracy: 0.5943 - val_loss: 1.0822 - val_accuracy: 0.5978\n",
            "Epoch 14/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.1013 - accuracy: 0.6179 - val_loss: 1.1841 - val_accuracy: 0.5857\n",
            "Epoch 15/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.0937 - accuracy: 0.6184 - val_loss: 1.0960 - val_accuracy: 0.6039\n",
            "Epoch 16/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.0663 - accuracy: 0.6246 - val_loss: 1.0862 - val_accuracy: 0.6075\n",
            "Epoch 17/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.0810 - accuracy: 0.6281 - val_loss: 1.0266 - val_accuracy: 0.6197\n",
            "Epoch 18/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.0130 - accuracy: 0.6512 - val_loss: 1.0428 - val_accuracy: 0.6403\n",
            "Epoch 19/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.9882 - accuracy: 0.6639 - val_loss: 1.1028 - val_accuracy: 0.6124\n",
            "Epoch 20/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.9767 - accuracy: 0.6680 - val_loss: 1.1679 - val_accuracy: 0.6221\n",
            "Epoch 21/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.9534 - accuracy: 0.6786 - val_loss: 1.1685 - val_accuracy: 0.6173\n",
            "Epoch 22/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.9478 - accuracy: 0.6807 - val_loss: 1.0308 - val_accuracy: 0.6488\n",
            "Epoch 23/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.9231 - accuracy: 0.6868 - val_loss: 1.1808 - val_accuracy: 0.6367\n",
            "Epoch 24/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.9113 - accuracy: 0.6965 - val_loss: 1.1599 - val_accuracy: 0.6318\n",
            "Epoch 25/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8938 - accuracy: 0.7060 - val_loss: 1.2654 - val_accuracy: 0.6452\n",
            "Epoch 26/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8741 - accuracy: 0.7187 - val_loss: 1.2110 - val_accuracy: 0.6586\n",
            "Epoch 27/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8560 - accuracy: 0.7192 - val_loss: 1.3261 - val_accuracy: 0.6586\n",
            "Epoch 28/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.8282 - accuracy: 0.7308 - val_loss: 1.2204 - val_accuracy: 0.6513\n",
            "Epoch 29/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8189 - accuracy: 0.7376 - val_loss: 1.3307 - val_accuracy: 0.6221\n",
            "Epoch 30/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8163 - accuracy: 0.7368 - val_loss: 1.1970 - val_accuracy: 0.6051\n",
            "Epoch 31/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.7816 - accuracy: 0.7476 - val_loss: 1.4635 - val_accuracy: 0.6343\n",
            "Epoch 32/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7828 - accuracy: 0.7507 - val_loss: 1.2103 - val_accuracy: 0.6537\n",
            "Epoch 33/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7707 - accuracy: 0.7593 - val_loss: 1.3049 - val_accuracy: 0.6744\n",
            "Epoch 34/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7715 - accuracy: 0.7540 - val_loss: 1.2139 - val_accuracy: 0.6719\n",
            "Epoch 35/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7565 - accuracy: 0.7633 - val_loss: 1.3352 - val_accuracy: 0.6646\n",
            "Epoch 36/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7308 - accuracy: 0.7718 - val_loss: 1.3693 - val_accuracy: 0.6610\n",
            "Epoch 37/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7475 - accuracy: 0.7670 - val_loss: 1.4425 - val_accuracy: 0.6914\n",
            "Epoch 38/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7210 - accuracy: 0.7767 - val_loss: 1.4860 - val_accuracy: 0.6987\n",
            "Epoch 39/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7065 - accuracy: 0.7790 - val_loss: 1.4297 - val_accuracy: 0.6513\n",
            "Epoch 40/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7076 - accuracy: 0.7872 - val_loss: 1.2937 - val_accuracy: 0.6707\n",
            "Epoch 41/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6972 - accuracy: 0.7835 - val_loss: 1.4697 - val_accuracy: 0.6659\n",
            "Epoch 42/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7032 - accuracy: 0.7784 - val_loss: 1.1814 - val_accuracy: 0.6938\n",
            "Epoch 43/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6781 - accuracy: 0.7942 - val_loss: 1.3537 - val_accuracy: 0.6804\n",
            "Epoch 44/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6550 - accuracy: 0.8025 - val_loss: 1.2901 - val_accuracy: 0.6999\n",
            "Epoch 45/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6714 - accuracy: 0.7987 - val_loss: 1.2675 - val_accuracy: 0.6974\n",
            "Epoch 46/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6649 - accuracy: 0.7983 - val_loss: 1.3897 - val_accuracy: 0.6974\n",
            "Epoch 47/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6432 - accuracy: 0.8039 - val_loss: 1.5045 - val_accuracy: 0.6938\n",
            "Epoch 48/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6599 - accuracy: 0.8021 - val_loss: 1.3776 - val_accuracy: 0.6780\n",
            "Epoch 49/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6398 - accuracy: 0.8101 - val_loss: 1.5348 - val_accuracy: 0.6804\n",
            "Epoch 50/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6494 - accuracy: 0.8036 - val_loss: 1.5032 - val_accuracy: 0.6974\n",
            "Epoch 51/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6099 - accuracy: 0.8203 - val_loss: 1.5870 - val_accuracy: 0.6974\n",
            "Epoch 52/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6118 - accuracy: 0.8220 - val_loss: 1.3177 - val_accuracy: 0.7047\n",
            "Epoch 53/100\n",
            "248/247 [==============================] - 11s 43ms/step - loss: 0.6086 - accuracy: 0.8170 - val_loss: 1.5759 - val_accuracy: 0.6902\n",
            "Epoch 54/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6278 - accuracy: 0.8124 - val_loss: 1.4686 - val_accuracy: 0.6464\n",
            "Epoch 55/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6047 - accuracy: 0.8212 - val_loss: 1.3398 - val_accuracy: 0.6841\n",
            "Epoch 56/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5881 - accuracy: 0.8200 - val_loss: 1.5352 - val_accuracy: 0.7035\n",
            "Epoch 57/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.5819 - accuracy: 0.8250 - val_loss: 1.5199 - val_accuracy: 0.6817\n",
            "Epoch 58/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5889 - accuracy: 0.8265 - val_loss: 1.4992 - val_accuracy: 0.6902\n",
            "Epoch 59/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5812 - accuracy: 0.8237 - val_loss: 1.4365 - val_accuracy: 0.6865\n",
            "Epoch 60/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5785 - accuracy: 0.8344 - val_loss: 1.4252 - val_accuracy: 0.6889\n",
            "Epoch 61/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.5728 - accuracy: 0.8328 - val_loss: 1.6695 - val_accuracy: 0.7060\n",
            "Epoch 62/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5777 - accuracy: 0.8332 - val_loss: 1.6438 - val_accuracy: 0.6744\n",
            "Epoch 63/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5641 - accuracy: 0.8315 - val_loss: 1.3724 - val_accuracy: 0.6902\n",
            "Epoch 64/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5672 - accuracy: 0.8288 - val_loss: 1.6424 - val_accuracy: 0.6853\n",
            "Epoch 65/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5618 - accuracy: 0.8383 - val_loss: 1.7800 - val_accuracy: 0.6707\n",
            "Epoch 66/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5498 - accuracy: 0.8347 - val_loss: 1.3257 - val_accuracy: 0.6610\n",
            "Epoch 67/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5428 - accuracy: 0.8411 - val_loss: 1.3731 - val_accuracy: 0.6646\n",
            "Epoch 68/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5379 - accuracy: 0.8385 - val_loss: 1.6322 - val_accuracy: 0.6403\n",
            "Epoch 69/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5335 - accuracy: 0.8420 - val_loss: 2.0199 - val_accuracy: 0.6865\n",
            "Epoch 70/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5338 - accuracy: 0.8418 - val_loss: 1.6345 - val_accuracy: 0.6962\n",
            "Epoch 71/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5325 - accuracy: 0.8476 - val_loss: 1.6250 - val_accuracy: 0.6707\n",
            "Epoch 72/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5331 - accuracy: 0.8469 - val_loss: 1.5700 - val_accuracy: 0.6817\n",
            "Epoch 73/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5223 - accuracy: 0.8508 - val_loss: 1.4113 - val_accuracy: 0.6804\n",
            "Epoch 74/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4995 - accuracy: 0.8528 - val_loss: 1.8265 - val_accuracy: 0.7108\n",
            "Epoch 75/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5018 - accuracy: 0.8567 - val_loss: 1.6028 - val_accuracy: 0.6707\n",
            "Epoch 76/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5037 - accuracy: 0.8564 - val_loss: 2.0065 - val_accuracy: 0.6950\n",
            "Epoch 77/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.5171 - accuracy: 0.8502 - val_loss: 1.9528 - val_accuracy: 0.6355\n",
            "Epoch 78/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5054 - accuracy: 0.8561 - val_loss: 1.5071 - val_accuracy: 0.7047\n",
            "Epoch 79/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4980 - accuracy: 0.8595 - val_loss: 1.4329 - val_accuracy: 0.6829\n",
            "Epoch 80/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5151 - accuracy: 0.8505 - val_loss: 1.3208 - val_accuracy: 0.6634\n",
            "Epoch 81/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4891 - accuracy: 0.8604 - val_loss: 1.2553 - val_accuracy: 0.6756\n",
            "Epoch 82/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4723 - accuracy: 0.8670 - val_loss: 1.4207 - val_accuracy: 0.6744\n",
            "Epoch 83/100\n",
            "248/247 [==============================] - 11s 42ms/step - loss: 0.4911 - accuracy: 0.8573 - val_loss: 1.4813 - val_accuracy: 0.6768\n",
            "Epoch 84/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4741 - accuracy: 0.8670 - val_loss: 1.6197 - val_accuracy: 0.6744\n",
            "Epoch 85/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.4935 - accuracy: 0.8576 - val_loss: 1.5059 - val_accuracy: 0.7060\n",
            "Epoch 86/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4882 - accuracy: 0.8561 - val_loss: 2.0864 - val_accuracy: 0.7035\n",
            "Epoch 87/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4814 - accuracy: 0.8652 - val_loss: 1.4029 - val_accuracy: 0.7120\n",
            "Epoch 88/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4706 - accuracy: 0.8633 - val_loss: 1.6757 - val_accuracy: 0.6792\n",
            "Epoch 89/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4705 - accuracy: 0.8688 - val_loss: 1.5185 - val_accuracy: 0.7084\n",
            "Epoch 90/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4650 - accuracy: 0.8667 - val_loss: 1.6752 - val_accuracy: 0.6683\n",
            "Epoch 91/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4628 - accuracy: 0.8689 - val_loss: 1.6410 - val_accuracy: 0.6561\n",
            "Epoch 92/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4613 - accuracy: 0.8686 - val_loss: 1.5146 - val_accuracy: 0.6780\n",
            "Epoch 93/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4479 - accuracy: 0.8724 - val_loss: 1.6748 - val_accuracy: 0.7047\n",
            "Epoch 94/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4411 - accuracy: 0.8756 - val_loss: 1.7428 - val_accuracy: 0.7084\n",
            "Epoch 95/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4643 - accuracy: 0.8657 - val_loss: 2.5097 - val_accuracy: 0.7181\n",
            "Epoch 96/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4338 - accuracy: 0.8784 - val_loss: 1.7848 - val_accuracy: 0.6756\n",
            "Epoch 97/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4458 - accuracy: 0.8736 - val_loss: 1.7483 - val_accuracy: 0.6695\n",
            "Epoch 98/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4415 - accuracy: 0.8772 - val_loss: 1.8713 - val_accuracy: 0.7096\n",
            "Epoch 99/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4462 - accuracy: 0.8686 - val_loss: 1.6449 - val_accuracy: 0.7047\n",
            "Epoch 100/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4398 - accuracy: 0.8737 - val_loss: 1.4914 - val_accuracy: 0.7145\n",
            "Training completed in time:  0:17:05.559936 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "id": "Ppgte1bCXZFs",
        "outputId": "7a8a5471-6921-4ac8-cdbc-98a3a3b6c431"
      },
      "source": [
        "show_results(history6)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3gAAAE9CAYAAABZZMC4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOydd1xUV/r/35feFbCgoIKKBVAUiUgkllSTbIrZZNM31eymbDbJliS7+9tks8l+s5uy6cmm9943MZtqjUhEFDugggJKkSKMwMAw5/fHYYYyF2lzmRk479eL1zAzZ8498znn3jPPfZ7zHE0IgUKhUCgUCoVCoVAoPB8vVzdAoVAoFAqFQqFQKBTOQRl4CoVCoVAoFAqFQjFEUAaeQqFQKBQKhUKhUAwRlIGnUCgUCoVCoVAoFEMEZeApFAqFQqFQKBQKxRBBGXgKhUKhUCgUCoVCMUTwMbJyTdOWAY8D3sCLQogHu7w/CXgZGA1UA1cIIUqOV+eoUaNEbGzsgNrV3NyMn5/fgOoYiihd9FG66KN00Ufpok9/ddm8efMRIcRoA5o0JFFzpHEoXfRRuuijdHFEaaKPEfOjYQaepmnewNPAaUAJsEnTtM+FELs6FHsYeF0I8ZqmaScD/wdcebx6Y2Njyc7OHlDbGhsbCQwMHFAdQxGliz5KF32ULvooXfTpry6aph0woDlDFjVHGofSRR+liz5KF0eUJvoYMT8aGaI5H9grhNgvhGgG3gXO61ImAfih7f9VOu8bQknJcZ2Ewxaliz5KF32ULvooXfRRungOqq/0Ubroo3TRR+niiNJEHyN0MdLAiwaKOzwvaXutI7nABW3/LwdCNU2LNLBNAJSWlhp9CI9E6aKP0kUfpYs+Shd9lC6eg+orfZQu+ihd9FG6OKI00ccIXQxdg9cLfg88pWna1cBaoBRo7VpI07QbgBsAxo8fz+rVqwGYPHkyoaGh5ObmAhAZGUliYiJr164FwMfHh4yMDHJycqirqwMgNTUVs9lsryM+Ph5/f3927NgBwJgxY5g2bRrr168HwN/fn/T0dLKzszGZTACkpaVRUlJi75Dp06fj7e3Nrl0y+jQqKoq4uDgyMzMBCAwMJC0tjaysLBobGwFIT0+nsLCQsrIyABISEmhtbSUvLw+A6OhoYmJiyMrKAiAkJITU1FQyMzMxm80AZGRkkJ+fT0VFBQBJSUmYzWYKCgoAmDBhAmPHjrWH64SFhZGSksL69euxWCwALFq0iJ07d1JVVYXJZKKmpob6+nr2798PyHCfiIgIcnJyAAgPDyc5OZk1a9YghEDTNBYvXkxubi41NTUApKSkUF1dTVFR0YD6qby8nOLiYpf3U1NTk328uEM/ASQnJ7u8n0wmEyaTyW36yV3Op8bGRvt4cYd+Avc4n0wmE2azuc/9pFAoFAqFom9oQghjKta0dOBeIcQZbc/vBhBC/F835UOAPUKImOPVm5qaKrquL2hpaaGkpISmpqZeta21tRVvb+9elR1O2HQJCAggJiYGX19fVzfJLTh8+DDjxo1zdTPcDqWLPkoXffqri6Zpm4UQqQY0aUii5kjjUHOkPuqap4/SxRGliT5GzI9GevA2AfGapsUhPXOXAJd1adgooFoIYQXuRmbU7DMlJSWEhoYSGxuLpmk9lm9paVEXZh1aWlrw8fGhqqqKkpIS4uLiXN0kt0D90NFH6aKP0kUfpYvrUHOkc1BzpD7q3NZH6eKI0kQfI3QxbA2eEMIC3AJ8DewG3hdC7NQ07T5N085tK7YEyNM0LR8YCzzQn2M1NTURGRnZq4nLVl7hSFNTE5qmERkZqTTqgC0ETdEZpYs+Shd9lC6uQ82RzkHNkfqoc1sfpYsjShN9jNDF0DV4QoiVwMour/21w/8fAh8641i9nbgUPaO0VCgUiqGFuq47D6WlQqFwd4zMoum2+Pg4166tra3lmWee6fPnzjrrLGpra53aloHgbF2GClFRUa5ugluidNFH6aKP0sVzUHOkPmqO1Eed2/ooXRxRmuhjhC7D0sDz9/d3an3dTV62zHrdsXLlSkaOHOnUtgwEZ+syVFDrLPRRuuijdNFH6dKOpmkTNE1bpWnaLk3Tdmqa9ludMks0TTuqadrWtr+/6tVlBGqO1EfNkfqoc1sfpYsjShN9jNBlWBp4x44dc2p9d911F/v27WPOnDmccMIJnHTSSZx77rkkJCQAcP755zNv3jwSExN5/vnn7Z+LjY3lyJEjFBUVMXPmTFasWEFiYiKnn366Pf37YOJsXYYKtrTvis4oXfQZSrrUHGvmhz3ltFoHnm15KOniBCzA74QQCcAC4GZN0xJ0yq0TQsxp+7tvsBqn5kh91Bypjzq39VG6OOLxmphNcMD538EIXYalgedsHnzwQaZMmcLWrVt56KGHyMnJ4fHHHyc/Px+Al19+mc2bN5Odnc0TTzxh38usIwUFBdx8883s3LmTkSNH8tFHHw3211AoFMOQ4uoGsouqHV63tFpZ8Xo2176azen/XsOX2w5jtQqaLVa+2HaIK17M4pRHVrO3ot4FrfZshBCHhRA5bf/XIxORRbu2Vcah5kiFQjEkyH0HXj0Lmo66uiU9MuQCyv/2353sOlR33DJ93eMnYXwY95yT2Ovy8+fP7+RufeKJJ/jkk08AKC4upqCggMjIyE6fiYuLY86cOQDMmzfPvqnxYKIWjusTGBjo6ia4JUoXfTxBl7KjTXyx7RD/3XaY3GK5xun/LpjFpfMn2ss8/n0B2QdqWHFSHKvzKrn57RymjQ2hytRM1bFmokcGYra0ctFzmbx27Xxmx8hQOiEEX24/zKo9lTx80Wz7dcUTdHEFmqbFAnOBLJ230zVNywUOAb8XQuwc6PHUHNl/1Bypjzq39VG6OOLxmjTWgrBKT17ACKdVa4QuQ87A6w1G78MRHBxs/3/16tV89913ZGZmEhQUxJIlS3TTK3eM7ff29nZJ+ElISMigH9MTSEtLc3UT3BKliz7uqovVKlhTUMlbGw/ww54KrAKSosO468wZbNxfxZ8+2U6Ivw/nJI9nw74jPLVqLxfOi+HPZydw15mCz3NLeXl9Eamx4Vw6fyInxY+muLqBK17K4tLnN/LCVal4aRr/t3I3uSVHmREVSk1DCxHBfoD76uJKNE0LAT4CbhNCdLW6coBJQgiTpmlnAZ8C8d3UcwNwA8D48eNZvXo1AJMnTyY0NJSmpibq6+vl3CekAWfD29sbq7UVIWzPvdA0zV7Gy0sG+litVtux8PLy6lQHyPBFW5ng4GCam5tpaWkBoLm5GSEE9fX1NDQ0EBAQgNVq5dixY6xbt46vv/6azMxMrFYrZ555JtXV1VitVoQQmEwmTCYTfn5+NDc3YzabsVgsNDU12euwtTM4OBiTyYRo+zIhISE0NTXZ1/oFBAQghMBsNgPg6+uLr68vDQ0Nneqor2/3SoeEhNDY2Gj/vhaLhdbWVpqamli9ejWxsbFERESQk5MDQHh4OMnJyaxZswYhBJqmsXjxYnJzc6mpqQEgJSWF6upqu5Fq66fc3FwAIiMjSUxMZO3atYBM7pKRkUFOTg51dXKYpKamUl5eTnFxMQDx8fH4+/uzY8cOAMaMGcO0adNYv349IH9jpKenk52djclkAuQ5WVJSQmlpKQDTp0/H29vbnrI9KiqKuLg4e/hYYGAgaWlpZGVl2X+jpKenM2LECPuYS0hIoLW1lby8PACio6OJiYkhKyvLrmdqaiqZmZn2fsjIyCA/P5+KigoAkpKSMJvNFBQUADBhwgTGjh1LdnY2AGFhYaSkpLB+/Xp73y5atIidO3favb/JycnU19ezf/9+AJf1k8lkcpt+KiwspKyszOX9tH37drfrp96eT6X79jAJ2LJpA3NPvcip/bR69eo+99NxEUJ41N+8efNEV3bt2uXw2vGor6/vU/meOHLkiJg4caIQQohVq1aJs88+2/7ep59+Kn72s58JIYTYvXu38Pf3F6tWrRJCCDFp0iRRWVkpCgsLRWJiov0zDz30kLjnnnuc2sbe0FGXvmo6lNm4caOrm+CWKF306ahLY7Ol23JNLd2/1xv2VtSLW9/JEb99J0e8tqFQbCuuFeV1jeKbnWXiH1/uEhc9u0Gc/cRasfzp9eLi/2wQCx/8Xky68wsx7+/fiH9+tVvsq2g/3xvMFnHRcxvElLu/FB9kF4sT7v9WLH14lTA1tfTYjsO1jeLUR1aLyXd/KSbd+YVIe+A78f6mg8LSau1Urr/jBcgWbjD3OPsP8EXuE3tHL8sXAaN6KqfmSONQc6Q+ai7QR+niiMdrsvJOIe4JE+LQVqdWa8T8OCw9eEIMPGFARyIjI1m4cCFJSUkEBgYyduxY+3vLli3jueeeY+bMmUyfPp0FCxY49djOxNm6DBVc4U31BJQu+jQ2NlLb0MzdH2/nhz0VvPDLVBZNG21/XwjBA1/u5q2sgzx35TwWd3hPjw37jlDfZCE5ZiRjw/wxmS08+cNeXl5fSICvN4F+3ny69VCnz/h5e5EUHcbY0ACaW62YW6zEjwnhzmUzOCMxCj+fzsuvA/28eemqVC57IYvff5CLn7cXr1xzAsH+PU8RUSMCeP9X6fz1853MHBfKtQvjCPB1jJJQ46UdTcb6vQTsFkI82k2ZKKBcCCE0TZuPXDPvuDjNANQcqY+aI/VR57Y+ShdHPF6Tloa2R+d+DyN0GZYGnhG8/fbbuq/7+/vz1Vdf6b5ncyePGjXK7gYG+P3vf+/09ikUisFhV1Urdz22jqpjZsaGBXDDG9m8fm0a8+MiAHj023xeXF/IiEBfVryezYtdDMCOvLhuP/d/udv+fHSoP1aroOpYMxfNi+GPy2YwKsSPQ0eb2HKwhrKjTcyZMJKk6BG6RtbxCA3w5bVr53Pbe1s5N3k8ieN7v74gPNiPJy+d26fjDXMWAlcC2zVN29r22p+AiQBCiOeAC4EbNU2zAI3AJcKDLQw1RyoUCo/H0rbEymbouTHD0sDruEZO0Y7SRZ/09HRXN8EtGY66CCHYU1ZPVFgA4W1ryzq+98g3+Tyd3UTcqGBevGohUSMC+MV/Mrn21U28vSKNDfuqePKHvVxywgT+uGwGl7+YxYrXs3npqhPIiB/Vqb7n1uzjwa/2cNasKK7LiGN7yVG2lR6lvsnCzUunMmdC+/5g0SMDiR458EXaEcF+vH7t/AHXo8dwHC/dIYRYDxw3Y4cQ4ingqcFpUWfUXKCP0kUfdW7ro3RxxOM1sRl2zc418IzQZVgaeGaz2fMz+RiA0kWfwsJCZsyY4epmuB2erktxdQOVJjNBft4E+noTHuxHWICvbtnDRxv5dMshPsopYW+FiYkRQby9Io2Y8CB7mYe/yePpVfs4PT6Ux69cSKCf9KC9dX0aFz2XySXPb6ShuZVzk8fzwPJZeHtpvHV9Gpe9sJHrXtvETUumMmNcKPFjQli5/TAPf5PPOcnj+fcvkvHx9mLepIhB0cUoPH28DCfUXKCP0kUfdW7ro3RxxOM1sYVmOjlE0whdhqWBZ8vmo+iM0kWfsrIyz74gGYQn6mK1CtbkV/LqhiLW5Fd2es/HS+OmpVO5ZelU+xq1+qYWHvo6jzc3HsAqIHVSOH9cNp3nVu/j4v9s5K3r04gdFcxza/bx9Kp9XDp/IqeHH7EbdwDjRgTy9vULuPSFjSyKH8Ejv0jG20s6byKC/Xjr+jRWvJ7Nv7/L79Se5XOjeejC2fh4D43tSj1xvAxX1Fygj9JFH3Vu66N0ccTjNbEbeM714Bmhy7A08BQKxfBACEFRVQNbDtaw5WAta/IrOVjdwOhQf247NZ7kmJE0trTS0NzK+oJKnvi+gG93lfPwRbMpO9rEXz7dQVldE79cMIlrFsYRO0qGaC2KH82VL2Xxi/9kcvEJE3jyh72ckzye+89PYt3aNQ7tmBgZxNo/LrUbdh2JDPHn45sWUt/Uwr7KYxSU1yME/HxejG55hUKhUCgULsCeZEWtwXNLAgICXN0Et0Tpok9CQoKrm+CWuLMuNceaeXdTMW9uPEBprbzjFuznTcqkcP5wxnTdTJIXzovh7Nnj+dMn2znnyfVYBUwfG8ozl6cwd2J4p7JJ0SN494Z0Ln9xI0/+sJeTZ4zh0TbPXHe69GSshQb4MmfCyE5r64YS7jxeFJ1Rc4E+Shd91Lmtj9LFEY/XpMWYJCtG6DIsDTwPTkRmKEoXfbpu6KuQuIMuZUeb+Nf/9lBc00BEsB8Rwf40Nlv4akcZZouV9MmR3HLyVOZOHEn8mNAejazTEsZyQmw4D3+Tx7gRgaw4abKDIWhjelQo7/8qnS+3HWbFosn4toVSuoMu7ojSxXNQc4E+Shd91Lmtj9LFEY/XxKAkK0boMjQWd/QRs9ns0uOHhIQAcOjQIS688ELdMkuWLCE7O/u49Tz22GM0NLQPsrPOOova2tp+t8vVurgreXl5rm6CW+JKXVqtgtc2FHHqo2v4cvthvDSNwiPH+HZXGT/sqeDCeTF8c/si3rlhAZfOn8iMqLBehzuODPLj/vNncXOHtXjdMXl0CL85Jb7TlgRqvOijdPEcXD0XqDnSs1Dntj5KF0c8XhODkqwYocuw9OC5C+PHj+fDDz/s9+cfe+wxrrjiCoKCZCa/lStXOqtpCsWgU3TkGCazhYRxYXh1MMZ2H67j062lHKptwsdLw9tLI6+snu2lRzkpfhT3n5/EpEiVvlyhGGqoOVKhULgVBiVZMYJh6cHz9dVPhd5f7rrrLp5++mn783vvvZf777+fU045hZSUFGbNmsVnn33m8LmioiKSkpIAuYv9JZdcwsyZM1m+fHmnXe1vvPFGUlNTSUxM5J577gHgiSee4NChQyxdupSlS5cCEBsby5EjRwB49NFHSUpKIikpiccee8x+vJkzZ7JixQoSExM5/fTTOx3H2boMFaKjo13dBLekP7ocPtpISU3nC2NLq5XHvsvn1EfX8LMn15P6wHf85p0tPP5dAWc9vo4zH1/HS+sK2V5SS/aBajL3VdHQbOGxi+fw+rXz3c64U+NFH6WL56DmSDVH9gV1buujdHHE4zWxGGPgGaKLEMKj/ubNmye6smvXLofXjkdra2ufyvdETk6OWLRokf35zJkzxcGDB8XRo0eFEEJUVlaKKVOmCKvVKoQQIjg4WAghRGFhoUhMTBRCCPHII4+Ia665RgghRG5urvD29habNm0SQghRVVUlhBDCYrGIxYsXi9zcXCGEEJMmTRKVlZX249qeZ2dni6SkJGEymUR9fb1ISEgQOTk5orCwUHh7e4stW7YIIYS46KKLxBtvvKGrS181Hco0NDS4ugluSV912VFaK2bf+7WYdOcX4tLnM8UnOSVie0mt+NkT68SkO78Qv30nR3ycUyxuf2+LSL3/WzHpzi/EuU+uE6/+WCiqTGaDvoXzUeNFn/7qAmQLN5h7POVPzZFqjhxs1DVPH6WLIx6tiaVZiHvC5N+7lzu1aiPmx6EXovnVXVC2/bhFrK0WvLz78NWjZsGZD3b79ty5c6moqODQoUNUVlYSHh5OVFQUt99+O2vXrsXLy4vS0lLKy8uJiorSrWPt2rXceuutAMyePZvZs2fb33v//fd5/vnnsVgsHD58mF27dnV6vyvr169n+fLlBAdLz8YFF1zAunXrOPfcc4mLi2POnDkAzJs3j6KiIvvnjh07RmhoaK9lGS5kZWWxZMkSVzfDpeSX1/PgV3sYEejL1DEhTB0TQmHeTqYnzqKxuZVWq2DK6BDix4bYk410JK+snitezCLYz5ur0ifxydZSbntvKyD3gnv28hTOnDUOgOVzYxBCUNvQQniw36B+T2egxos+Shc3Qc2Rao50Murc1kfp4ohHa9LRa+fkNXhG6DL0DDwXcdFFF/Hhhx9SVlbGxRdfzFtvvUVlZSWbN2/G19eX2NhYmpqa+lxvYWEhDz/8MJs2bSI8PJyrr766X/XY8Pf3t//v7e3dKfxEodCjvK6Jq1/+iWPNrQT7efPJltL2N3/a1Kmsn7cX8WNDSJ0UzpLpY1gwOZLS2gYuf3Ejfj5evL1iAbGjgrnt1GlsLKxia3EtF82bwOhQ/071aJrmkcadQqHQR82RCoXCo+lo1Dk5i6YRDD0D7zh3EW2Yjx3DJ9i563YuvvhiVqxYwZEjR1izZg3vv/8+Y8aMwdfXl1WrVnHgwIHjfn7RokW8/fbbnHzyyezYsYNt27YBUFdXR3BwMCNGjKC8vJyvvvrKbuWHhoZSX1/PqFGjOtV10kkncfXVV3PXXXchhOCTTz7hjTfe6PE7eHkNyyWZPWLL6DYcOWa2cO2rmzja2MJ7v0onKXoEJrOFvRUmcrbtZM6sBAJ9vdE06aXbdaiOnYfqeD+7hNcyD+Dn44W/jxf+Pt68df0C+0bhXl4aJ04ZxYlTRvXQAs9jOI+X46F0cRPUHKnmSCejzm19lC6OeLQmnTx4zjXwjNBl6Bl4vSDYyRMXQGJiIvX19URHRzNu3Dguv/xyzjnnHGbNmkVqaiozZsw47udvvPFGrrnmGmbOnMnMmTOZN28eAMnJycydO5cZM2YwYcIEFi5caP/MDTfcwLJlyxg/fjyrVq2yv56SksLVV1/N/PnzAbj++uuZO3dup1ATPYzQZSiQmprq6iYYxuGjjfxUWE1BuYmCinoOH20iZWI4p8wcQ+qkCG55O4c9ZfW8eFUqSdEjAAjx92nbkHthp7pmRIVx3hy5ULippZWfCqtZnVfJwepj/HHZDKaO8eALex8YyuNlIChdPAc1R+qj5kh91Lmtj9LFEY/WxLbJuebl9BBNI3TR5Bo9zyE1NVV03ftm9+7dzJw5s9d1mEwmz76LYBAddemrpkOZzMxM0tPTXd2MTgghqGuyMCKwf1ndjja08PTqvbz6YxHNrVZ8vDRiRwUzKsSPLQdrMVus+HprtLQK/rF8FpelTXSowx11cQeULvr0VxdN0zYLITz4V8HgouZI41BzpD7qmqeP0sURj9akZDO8eDIEjQLfQLh9h9OqNmJ+HJYePE8zagcLpYs+7rK57faSo7y76SB7yurJL6un3mxh6fTR/OGMGSSMD9P9TKtV8M//7SG7qJqJEUFMjAxGA17dUERdUws/T4nhuow4powOsW/q3djcyoZ9R/h+TwVTRofoGnfgPrq4G0oXfZQunoOaC/RRuuijzm19lC6OeLQmtrDMoEhoOOLUqo3QZVgaeAqFp3HMbOHa1zbRYLaQOH4Ey1OiCfH34c2NBzj7yXWcmzyeO06b1mlPuJZWK7e/t5Uvth0mecJINhXV8FnuIYSAJdNHc+eyGcwc52gYBvp5c8rMsZwyc+xgfkWFQqFQKBQK98QWlhkUCbUHXduWXjAsDTwVeqKP0kWfjIwMVzeB59fup7LezMc3nUjKxHD7679aNIXn1u7jlR8L+XLbYX6eEsMtJ09ldKg/t7y9he92l3P3mTP41eIpAJgtrRxtaGFMWMCA2+QOurgjShd9lC6eg5oL9FG66KPObX2ULo54tCZ2D16E3PDcagUnJV4yQpchkxKqL6ETA0mhPJSx6aLCUDqTn5/v0uOX1zXx/Nr9/Gz2uE7GHcCIIF/uXDaDtX9YyhUL5P5ySx9ezdlPrOO73eX8/bxEu3EH4O/j7RTjDlyvi7uidNFH6eJa1Bw5cNQcqY86t/VRujji0ZpY2q6LQZGdnzsBI3QZEgZeQEAAVVVVvb7oWiwWg1vkmVgsFoQQVFVVERDgHCNgKFBRUTFox/psaykvrtuPpdVqf+2Rb/JotQruXNZ9lrkxYQHce24ia/+wlMvSJnLE1MxDF87myvRYw9o6mLp4EkoXfZQurkPNkc5BzZH6qHNbH6WLIx6tScc1eB2fOwEjdDE0RFPTtGXA44A38KIQ4sEu708EXgNGtpW5Swixsq/HiYmJoaSkhMrKyl6Vb2pqUhdnHWy6BAQEEBMT4+rmDCusVsFD3+Tx7Op9AHyx7TD/vngOTS2tfLC5hOsz4pgQEdRjPVEjArjvvCTuOy/J6CYrFAoPQc2RzkHNkQrFMMa2Bi+4bV9NJ++F52wMM/A0TfMGngZOA0qATZqmfS6E2NWh2F+A94UQz2qalgCsBGL7eixfX1/i4uJ6Xf7IkSMOG58qlC7dkZTkXGOpoq6J4ppGZkSFEuzvQ1NLK797P5cvtx/msrSJpMVF8P8+3cFZj68jOjyQsABfblka79Q2OANn6zJUULroo3RxHWqOdA5KF33Uua2P0sURj9bEwYPnvL3wjNDFSA/efGCvEGI/gKZp7wLnAR0NPAHY0viNAA4Z2B47Hp2m1UCULvr0RpeiI8e4/8vdzJsUzi9SY4gM8Xesx9LKi+sKefKHApparGgaxEUGgwaFR47x57Nmcv1JcWiaxvy4CP7wwTbW7z3C//tZAiOC+rffnZGo8aKP0kUfpYvnoPpKH6WLPkoXfZQujni0Ji2NcpPzgBHyefMxp1XtadskRAPFHZ6XAGldytwLfKNp2m+AYOBUA9tjp6CggOjo6ME4lEehdNGnJ122lxzlmld/or7Jwne7y/n3t/mcOSuK0xLGEuzvQ5CvN9XHmvnX13kUHjnGssQozp87nrwyEzsPHeXQ0UaevTyFZUnj7HWOGxHI69fOZ+ehOpKi9fe4czVqvOijdNFH6eI5qL7SR+mij9JFH6WLIx6tSUsT+AbJP3CqB88IXVy9TcKlwKtCiEc0TUsH3tA0LUkIYe1YSNO0G4AbAMaPH8/q1asBmDx5MqGhoeTm5gIQGRlJYmIia9euBcDHx4eMjAxycnKoq6sDIDU1FbPZbK8jPj4ef39/duyQO9KPGTOGadOmsX79egD8/f1JT08nOzsbk8kEQFpaGiUlJZSWlgIwffp0vL292bVLOiejoqKIi4sjMzMTgMDAQNLS0sjKyqKxUQ6I9PR0CgsLKSsrAyAhIYHW1lby8vIAiI6OJiYmhqysLECmZ05NTSUzM9Nu6WdkZJCfn29fnJmUlITZbKagoACACRMmMHbsWLKzswEICwsjJSWF9evX2xfRL1q0iJ07d1JVVYXJZKKmpob6+nr2798PQGxsLBEREeTk5AAQHh5OcnIya9asQQiBpmksXmTnacQAACAASURBVLyY3NxcampqAEhJSaG6upqioqIB9VN5eTnFxcUu76empib7eOnaT2WM5L4fygj0snLvggACgoLINYXy3k8H+GxrZ4f02CCN383z59KlkzCbzQQcKSV5oq2fQu3HsPXThg0/YrFYWLO3cz8BJCcnu7yfTCYTJpPJbfrJXc6nxsZGe1+6Qz+Be5xPJpMJs9nc535SKBQKhcLltDSAb2AHA8+91+BpRqX7bTPY7hVCnNH2/G4AIcT/dSizE1gmhChue74fWCCE6DadTGpqqrD9wOov+/btY8qUKT0XHGYoXfTpTpf/7SjjN+/kMGV0CK9dO5+xHbYfaGxupajqGI0trTQ2t9JqFaRNjsDfx3swm24oarzoo3TRp7+6aJq2WQiRakCThiRqjjQOpYs+Shd9lC6OeLQmH/8KDm6AS9+FZ0+EX7wOCec5pWoj5kcjPXibgHhN0+KAUuAS4LIuZQ4CpwCvapo2EwgAepfmawCMHTvW6EN4JEoXffR0qW1o5u6Pt5EwLozXr0tjRGDnNXKBft7MHOeeoZXOQo0XfZQu+ihdPAfVV/ooXfRRuuijdHHEozVpaWgL0Qxse+68EE0jdDFsHzwhhAW4Bfga2I3MlrlT07T7NE07t63Y74AVmqblAu8AV4tB2EF0oHc3hypKF330dPn3t/kcbWzhwZ/PdjDuhgtqvOijdNFH6eI5qL7SR+mij9JFH6WLIx6tSUsj+AS0h2g6McmKEboYugavbU+7lV1e+2uH/3cBC41sg0LRF1bnVfC793OZHTOCS+ZP5OQZYxzK7Cmr442NB7hiwaQh76VTKBQKhUKhGPZYjEuyYgSuTrLiEsLC1I9yPYa7Lh9kF3PXx9uZFBnEzkN1/OqNzYwJ9efkib6cYLYQ7O+DEIJ7P99JWKAvd5w2zdVNdinDfbx0h9JFH6WL56D6Sh+liz5KF32ULo54tCYtDRAw0pAQTSN0GZYGXkpKiqub4JYMV12EEDz1w14e+TafjKmjePaKFAJ9vVmVV8nbWQd4d2cl3z+8mt+dNo0gfx827q/m/vOTGBnk5+qmu5ThOl56Qumij9LFc1B9pY/SRR+liz5KF0c8WpOWRggdB96+4OULLc4L0TRCF8PW4LkztlTgis4MJ12OmS2sya/koa/3sPyZDTzybT4XzI3m5atPIDTAFx9vL05LGMsr18zn3oXBTAgP5K6Pt3PrO1uYOS6MS+dPdPVXcDnDabz0BaWLPkoXz0H1lT5KF32ULvooXRzxaE1sSVYA/IKc6sEzQpdh6cGz7Vml6Mxw0SVrfxXXvrqJY82teHtpzIoewV/Onsl1GXFomuZQPjYUPrrxRFZuL+P1zCLuPmsm3l6O5YYbw2W89BWliz5KF89B9ZU+Shd9lC76KF0c8WhNWhrBt207LN8gp+6DZ4Quw9LAUwwPKuvNRAb74dXBGNtaXMt1r2UTNSKAe85JZN6kcIL9ez4NNE3j7NnjOHv2OCObrFAoFAqFQqFwN1qa2j14voHQPEw3OjcKZ2ziarVa8fIaltGpx2Uo6fJW1gH+8ukOpo0J5Y7Tp3F6wlj2lNVzyfMbCQv04YNfnUjUiICeK2Jo6eJMlC76KF306a8uaqPzvqHmSONQuuijdNFH6eKIR2ty3yg48RY49V54NgNGToRL33ZK1UbMjx6q8sDYuXOnq5vglgwFXYQQPPl9AX/+ZAdpcRE0t1r51RubOf/pH7nypSwCfb15+/oFvTbuYGjoYgRKF32ULvooXTwH1Vf6KF30Ubroo3RxxGM1aW0Ba0tnD54Tk6wYocuwDNGsqqpydRPcEk/XxWoV3PfFLl7dUMQFc6P554Wz0YCPc0p5/PsCAN68Po0JEUF9qtfTdTEKpYs+Shd9lC6eg+orfZQu+ihd9FG6OOKxmtgSqti2SHBykhUjdBmWBp5i6NFqFfzhw1w+zinluow4/nzWTPvau1+cMIHlKdG0tFoJ8lNDXqFQKBQKhULRS2zGnE+HJCsN7m2sDssQzeTkZFc3wS3xBF32VZr4347DWK3ta0etVsGdH23j45xS7jhtGn85e2anxCoAvt5e/TbuPEEXV6B00Ufpoo/SxXNQfaWP0kUfpYs+ShdHPFYTi82DZ0ySFSN0GZYGXn19vaub4Ja4uy7ZRdWc//SP/PrNHM57+kc2FVVjtQru+ngbH24u4bZT47n1lHjdrQ4Ggrvr4iqULvooXfRRungOqq/0Ubroo3TRR+niiMdq0jVE09e5IZpG6DIsDbz9+/e7ugluiTvrsjqvgiteymJUiD9/Pz+JynozFz2XyZmPr+P97BJuPSWe206dZsix3VkXV6J00Ufpoo/SxXNQfaWP0kUfpYs+ShdHPFYT2553dg+ec/fBM0IXtSBJ4dYIIfg89xC//yCX+DGhvH7dfEaF+PPzlGieW7OfF9ft59aTp3L7qfGubqpCoVAoFAqFYqjh4MELdKqBZwTD0sCLjY11dRPcEnfRRQhBbslRvtp+mJU7DlNc3cj82AhevDqVsABfAIL8fLjjtGncdkq8w3o7Z+MuurgbShd9lC76KF08B9VX+ihd9FG66KN0ccRjNXHIohkMrc3QagHvgZtSRugyLA28iIgIVzfBLXEHXYQQ3PF+Lp9sKcXXW2Ph1FHcsnQq582JJsDX26G80cYduIcu7ojSRR+liz5KF89B9ZU+Shd9lC76KF0c8VhN9Dx4IL143mEDrt4IXYblGrycnBxXN8EtcQdd/v1dAZ9sKeWmJVPI/vNpvHrNfC4+YaKucTdYuIMu7ojSRR+liz5KF8/goa/38NCnWa5uhluixrA+Shd9lC6OOEWTty+BHx8feD19oUUni2bH1weIEWNlWHrwFO7Jp1tKeeL7An6RGsMfzpju9GyYCoVCoTg+6wuO0NxgcXUzFAqFQp/irHYDa7CwJ1mxefCCO7/uhgxLAy88PNzVTXBLBlOXo40t7Dx0lMhgf0aH+rO3wsQfP9xG+uRI7j9/llsZd2q86KN00Ufpoo/SxTOYERXGym31CCHc6jrsDqgxrI/SRR+liyNO0aTZNPiG1fFCNJ2AEWNlWBp4HrvRosEMli4b91fx23e3UF5n7vT65FHBPHtFCn4+7hU5rMaLPkoXfZQu+ihdPIPpUaG8l22lst7MmLAAVzfHrVBjWB+liz5KF0cGrImlWSY3aT7mnAb1Fpsh59MhyQo4LURTbXTuJNasWePqJrglRuvSahU89l0+l72wkWA/H56/ch5PX5bCPeckcMdp03jj+jRGBvkZ2ob+oMaLPkoXfZQu+ihdPIMZUaEA7Cnz0A2JDUSNYX2ULvooXRwZsCbNJvk42B48SxOggY+/fO5kD54RY2VYevCEEK5ugltipC6Nza1c99omNuyr4oK50dx3fhIh/p4x/NR40Ufpoo/SRR+li2cwvc3AyyurZ9G00S5ujXuhxrA+Shd9lC6ODFgTc9uNp2YXhGj6BoEtbN1m4DmpHUaMFc/4he1k1LoCfYzSRQjB3R9vI3N/Ff/8+SwuPmGiIccxCjVe9FG66KN00Ufp4hlEhvgzwl9THjwd1BjWR+mij9LFkQFrYvfguSBEs2NiFycnWTFirAzLEM3Fixe7ugluiVG6vPJjEZ9uPcTtp07zOOMO1HjpDqWLPkoXfZQu7WiaNkHTtFWapu3SNG2npmm/1SmjaZr2hKZpezVN26ZpWspgtW/2xEj2lNUN1uE8BjWG9VG66KN0cWTAmpjbDDxXefBsOHmbBCPGyrA08HJzc13dBLfECF0y91XxwMrdnJYwlluWTnV6/YOBGi/6KF30Ubroo3TphAX4nRAiAVgA3KxpWkKXMmcC8W1/NwDPDlbjIr2bKKgwYWm1DtYhPQI1hvVRuugzaLqU74K/j4Hq/YNzvAEwYE2a2yILBj2LZgP4dkg6ZTP2nNQOI8bKsAzRrKmpcXUT3BJn6HK0oYXS2kZMZgs1Dc386ePtTIoM4tFfJOPl5ZnhCmq86KN00Ufpoo/SpR0hxGHgcNv/9Zqm7QaigV0dip0HvC7k4oyNmqaN1DRtXNtnDWWUbzPNFitFVQ1MHRNi9OE8BjWG9VG66DNoulTuhlYzHCmAiMnGHKNwHfiHwPi5A6pmwJqYOyRZsVrBa5D8VC1NnUM0/Zxr4BkxVoalgacwhj1ldfz8mQ0ca261vxYa4MPzV6YSGuDrwpYpFIpeUXcY3r8S0n4Nsy50dWuGBZqmxQJzgawub0UDxR2el7S9ZriBFxMifzTtKatTBp5C4e40VLc9Vhl3jJV/gKaj8JvN7caNK7CtwQOwNLZvV2A0LQ2dQzR9nJtkxQiGpYGXkjJoSxk8ioHo0tBs4Za3txDk78O/LkxmRKAvIQE+TIoIIjzY/bY+6AtqvOijdNHHY3U5dgTeOB8q90DI2IEZeIe2wo6P4NS/2e+weqwuBqJpWgjwEXCbEKLfi940TbsBGcbJ+PHjWb16NQCTJ08mNDTUHv4TGRlJYmIia9euBcDHx4eMjAxycnKoq5OHT01NZc7ksWiZB/g6awdzIwX+/v7s2LEDgDFjxjBt2jTWr18PgL+/P+np6WRnZ2MyyR9faWlplJSUUFpaCsD06dPx9vZm1y7poIyKiiIuLo7MzEwAAgMDSUtLIysri8ZGuaYlPT2dwsJCysrKAEhISKC1tZW8vDwAoqOjiYmJIStL2sUhISGkpqaSmZmJ2Sz3WM3IyCA/P5+KigoAkpKSMJvNFBQUADBhwgTGjh1LdnY2AGFhYaSkpLB+/XosFgsAixYtYufOnVRVVWG1WqmpqaG+vp79+2U4XGxsLBEREeTk5AByw+Lk5GTWrFlj3yx+8eLF5Obm2u/Sp6SkUF1dTVFR0YD6qby8nOJieQ8gPj7eZf0UHh5uH3Pu0E8g9xVzdT9ZrVZMJpPh/ZR4JI/RwN7tmyipHW/I+XRi7SH8Wo5S/t+/sTvyzH7305w5c9i+fXu/+ymmdCu2xT4/rv6WFr+Rg3I+za+rJmDkGNa2jXN/f3/SfQIoK97PnrbXBnI+Wa1WVq9e3ed+Oh6ap6VxTU1NFbbB01+KioqIjY11ToOGEAPR5a6PtvFedjFvXJtGRvwo5zbMxajxoo9H6tJQDUERhh7Cqbq0WiDrWYhbBOMM3DS3sRZeOweO5MOoaWAqh9/ltaeE7gsVu+GVs8AvBG5YBcHyetBfXTRN2yyESO17Q9wbTdN8gS+Ar4UQj+q8/x9gtRDinbbnecCSnkI0nTVHXvdREZNHh/DCL4ec9P3GI695g4DSRZ9B0+WrOyHrOci4A069x/n1CwF/HwVWC/iHwa1bITiyX1UNWJO1D8EP98v/f5sL4QOoqy88cyJExMElb7W/9s84SPo5nP3wgKs3Yn4clklWbFa+ojP91eW/uYd4d1MxNy6eMuSMO1DjpTs8Tpddn8HD8VBdaOhhnKZLSxN8cBV88xf44QH9MkdLBx4i0lgLb10kDbOL34K5V0oD72jJ8T9Xuhm+uB0q9rS/VrUPXj8fvP3gqs/sxh144HgxEE3mxH4J2K1n3LXxOfDLtmyaC4Cjg7H+DmRfzYgKU5k0u6DGsD5KF30GTRdbaGZjtTH1m+ulcTf3Chkiufahflc1YE3MHUI0BzM8sqUBfAI6v+Yb5LQ1eEaMFUMNPE3TlmmalteW5vkunff/rWna1ra/fE3Tao1sj8L5HKxq4E8fb2fuxJHcfto0VzdHMVjUFOHfVOnqVvSNHR/JSapovatb4kj5TjicKxeNAzTVwVsXwp4vYPQMKFonDb6OtDTCs+mwqhvj73gIAcU/wac3w6MzpbF20SsQfyrEtN0MLO3BC/TTC5D9smzDJ7+GAxvg9fOgtRl++Zlxi/2HBguBK4GTO8yBZ2ma9mtN037dVmYlsB/YC7wA3DSYDZwRFUpxtUyYpVAo3Bij1+DZDMeJ6ZDyS9j0ousydnZcgzeYmTQtXZKsgFyLONjZPPuAYWvwNE3zBp4GTkMuDt+kadrnQgh7ljAhxO0dyv8GudDccCZPVj889OirLuV1TVz1yk9oGjxxyVx8vYemQ9gtx8ve72D1P+Gsh2D8nN5/ztoKe76E6WeBdz9P/5ZGeOVsUvwjgIv6V8dgYzHD3u/l/8UbIeXK3n/20Fb4+s9w8Ru9Cu/s83ixmOHlZWCug8AIiDtJTp4Vu+GCFyBgJLx9ERzMhClL2z+3b5Vc9F64pm/Hs1rhjfOgcK3crHXWhZB6Xfs4Gpsk71SWZEPi8u7rOZgJk5dA1Gz46XnIfQf8R8BVn8OYGQ7F3fI8chFCiPXAceNf27Jn3jw4LerM5MmTaa6Xa6fzy+tJmRjuima4HWoM66N00WfQdLEZYA0GZe20GZCBEbDkbtj2Pnz/d3lTsI8MWJNOHrxB3Oy8a5IVkAafk/bBM2KsGPmLfD6wVwixXwjRDLyLTPvcHZcC7xjYHjuhoaGDcRiPoy+6VNabueyFjVTUNfHKNfOZEOHCrEoG02tdBmM9q9UqwyPevBBKfoLPboHWlt5/Pud1mSVxy+v9b0PWf6CuBP/KbWCq6H89g0nROnnnL2AkHOx5cXIn9q+CA+thyxu9Kt7n68uBH6Vxl34LTFsmDauaA3DJOzD7FxC7ELz9pVHfkT1fyMfynTKEprcUZ0nj7qTfwe/z4NwnO98k8PGDcXOgZFP3ddSXQ00RTD0VTv+7XJNx0u/gl592e8NBXXc9h9DQUGZEhQGw53AfxtYQR41hfZQu+gyaLjbPnWEevDbDMTAcQqMg/WbY+bGcp/rIgDVxlQevpdHRg+cb5DQj04ixYqSB112KZwc0TZsExAE/GNgeO2pTTn16q0uVyczlL27kUK007uZN8rC7u01HYf2/2+9K9UCvdDmyF/4RDU/Nl+mEd38hvTA1RfLvaEn/DMCG6vY6juyF966QC4xnXQgXvAjl2yHzqd7V1WqBHx+X///0Qv/bs+5RGTYIUPBt3+twBXlfyYtx2q+hqkBmjOwttjV7m16UHtAe6PP1Jf8b6TFb+mdY/izcvhPuLIJpp8v3/YJh0omdDbxWC+SthLAYENbjG2Nd2fGhTPGccQf4dzOpxKTKkFFLs/77xRvl44QF8jFsHJzyV4juPlOmuu56Drm5ucSEBxLs502eWodnR41hfZQu+gyaLjbPnVFr8GwGni2CZdKJ8rHuUJ+rGvhG5yYIGNH2/yB58FotcumBgwcvyGkevKG80fklwIdCCN1fT85OAW02m+11uDK1sLulgDaZTD2mgPYNHsH9G0wUVpq4fV4AjQe3Q5znpID2O7qfwM+uJ6jxEBWVRxhx1l977KempqYeU0DXfvhHxrS2YNLCGLHlTRmy1gVT8CQOjzuNiCU30qQF9thPuV88z6zNf8JLtHvorJo3+6Zez6ilf6beZCJwVDoRP/yDw2GphMUmHz8FdEQl1BRSOSqN0RVZHNv1DZXB0/vWT8e+QpjryZ50H0m19+K76wsOhC0Y1FTd4dYqkq3b2RS0lGPNwt5P3Z5PQrBw53/xil3MtqNhzAX2rnqLqT/7bfv5JAQZyVPILzM5nE9B+7cwQvPBq/Ygjds/I6sm4rjnU2Njo3289CoFdMHXVIUlsn3DT+2purft6HQ+WSJTiNj/CJn/+4DohDTCa7cT2ljDnolXMr3uKTiQyZpirx7Pp5IDRaRv/QDLxMU01DexIzNbt5/GNYQx3dLErtUfUOET7dBPU/Z+SIy3P5U+49jV9l17uu6ZTCbMZnOfr3sK1+DlpTEtKpQ9ZcqDp1C4LZZmaK4HzUt68IToX/bj42H34LUZeAEj5WOTC9JmmE1yG5+mo4PnwbO0GXG+XZOsBEJ92eC0oT8IIQz5A9KR6Z9tz+8G7u6m7BbgxN7UO2/ePDFQtm3bNuA6hiI96VLf1CLOe2q9iP/TSrE6r2KQWuVEdnwsxP3jhPjXVCEejBXi/at79TG7LmaTEB9eJ8Th7Z0L1BwU4m8RQnz5B/m8pUmIwvVCbHmr/S/zGSH+s0SIe8KEuG+0EN/dd/yDWq1CvHCKEA9P71zP4S59dPSQEP+IEeLVc+RnuqO1VYinFwjx1HwhzMeEeHCSEO9e0avvb6e6SIj7Rgnx6U1CCCGqXrlciAeihWgx962egVBdKMTDM6SOOW/07jOHtraXb26U3+Hrv3Quk/OmEPeGC1Fb7Pj5f88S4r1fCvHITCFeO6/Hw/Xp+lJZINuW9fzxy1XskeU2vSyff/l7If4+Ro7JZxbK/u8NBd/JenZ9fvxytcWy3Mb/6L///FIhXlrWu2O20d/rLpAtDJqnhuKfM+fIuz7KFbPv/VpYj3dtGUao3w76KF30GRRd6g7La/VjyfKx8ajzj7HqQVm3pVk+r9onn295u89VDViTp+YL8crZ8viZzw6srt5SX6E/T394vRCPzXbKIYyYH40M0dwExGuaFqdpmh/SS/d510Kaps0AwoFMA9vSicTExME6lEdxPF2aWlq5/rVNbC89ylOXzWXxtNGD2DInkP0yfHA1jE2AX62R65oO984lbtdlz0rY/gF8eG1nt/yGJ+Xjib+Rjz7+sv45l7X/LbhR7gn26/Uw7QxY9zAU/dj9QXd/LsPulv65cz1RszqXCxsHp94rE23kHmcJa8HXULFLhuX5BclMWHu+7JwKv7VFfsfuQld/uF/eJVzyJwBGpl0i7xwe7HLqVuZB2Y7u29Jf6g7JLI0txyB0PGx7r3efy/sK0CD+DHkHbtwcuQ6tI9kvg2iV69k60toiNYqcCqnXyPV4lfnHPVyfri/5/5OP8acfv9yoaTBiggzTFEL23ZRTZPjmxDS5bq+1F9kOd3wk9zGaetrxy4VFQ0iUfuhnc4M8dyam9Xy8Dqjrrudg66sZUWEcbWyhrK6ph08MD9QY1kfpos+g6GKbr0fFtz03YB1eY7WcN7x95fMBePAGrInZBCFj5P8tgxSiafMU6mbRdE6IphFjxTADTwhhAW4BvgZ2A+8LIXZqmnafpmnndih6CfBumyU6KNhCzhSd6U6XZouVX7+5mazCah79RTKnJ0YNcsucwE8vQPQ8uHolhI2Xm0ZX75Pp6HvArsuuT+XmzUfy4Pv75GumSsh5DWZfAiMn9NyOqFmw/D9y7dRXf9T/Ud7aAt/9DUbPlEZdT8y7BsantBuaXREC1j4MIyfKTTlBZk1ESMMGZCz7O5fCu5fCIzPgw+tkIo6KPTKpyjuXwfb3YcFNMEKG7P1Y6i33Oyv4pv1YjbXwypnw3EJ470pp7DmDY0ekcXfsCFzxMcy7CgrXyX3geiJvJUyYDyFtNyUmpsGhLe3bDlTsbt8S4EgX4+1oiTT8IuIg5Wr5fTe9cNzD9en6UvC1XM8YPun45TQNpp4C+9fI7Q3qSmHmOfK9CQvkRFfeg1Hd0gS7/wszfuYYaqJ3vJhU/a0SDuXI7SYmph+/ji6o667nsHaNzMxqy56Zuc+g5A0ehhrD+ihd9BkUXWwGXWSbgWfEOrzGGggc2f7ctgause8G3oA1aa6HoFGANnj74NmMON0kK85pgxFjxdC89kKIlUKIaUKIKUKIB9pe+6sQ4vMOZe4VQjjskadwDwrK67n4+UxW51Xyj+WzOG+Obp4c96a6UHqvki6UGQJBenEAyrb3rg5zvUwoMudymH8DbHwG9q+GrGdlmvuM23rfHr8gOON++YM851XH93Nek8bnqfeCl3fP9Xl5yeyLFbv1sykWrZM/1Bfe1r41QvgkmHYmbH5VxpC/fj7s+x5OuQfmXQ17v4XXzoFn0qQhWr4DTrheZkpso9UnEGIzIP/r9mOtflDeUZx/g0zj/8wCudfaQC+Cn/wKaovhsvel4TH7F4CQRufxOFoqvU3Tz2x/bcICuWD68Fb5fMub4OUj71B2NfBq2hKshMdKAzHxAtj6dq9uDPRIU53cO27aGb0rP/VUObl9+1fQvNs/N7Et0UlXr2RX9n4ns3XO+nnvjhdzgkwUdKzLj3ubxzbmhN7Vo/AchIA3L2TKPnnjJ3F8GJHBfqzJ97A9LxWK4YLNoIucIh97mTyuTzRUt6+/A/m7xD9s8NfgCSE9eP4hMnplsNbg2T14etskNAxOBvV+MDQ3LusBHx93yS3jXnTUpaXVytOr9nL2E+spOnKMJy+dy6XzJ7qwdQPAFgY3fVn7a+OS5WMvwjR9fHykEdNqhsTz4dS/ybtln94kPYMJ57WHR/SWhPMh9iQZ9tjxgmyul0bSpIze//AH6Z1EyD3burLxOQgeI43TjsxfIe/+PZ0mjZ2LXoWT7oCz/gW/y4OfvyRT6P82F27bBmc/Ii+sbfj4+Miwx6qCtn3b9sjkMvOulvvz/TZXevy2vgmr/68v6nSmtUV66064Toa+gtxEe0Ia5L7neHGtL2s3KPO/ko/Tz2p/f0JbaOHBjbLube9JA3lsEhwp6FxXTZF8DI+Vj2k3yCxeue9229xeX1/2r5KesPhe9nPcYmmIFm+UhrUto9nICTKk8uDG439+x4cQFAlxS3p3vO42PD+YJb2OvdgTsCPquusBaBr4BjC2Yi1YW/Hy0lg0bTTrCo5gtbrnj5jBRI1hfZQu+gyKLvYQzWmdnzuTxmrH633ASJnopI8MSBNLk4yo8Qtx6hYFPWLz4Pl0TbISJNvTl62qusGIsTIsDbyMjAxXN8EtmTY7lZXbD3Pff3dxxmNreejrPE5LGMu3dyzmnOTxrm5e/8lbKcMdIzpsJBkyBkLH9crAy8jIgJ2fyDVJExZID9wF/5GGhLlOGkV9RdPgzH9KL86qB+RFeeen8NEKOFYJp93Xt0xYtvT0XX+Mt1pkqOVMnbC8yUvkD3WrBS7/QBqqNnwD5VYMKb9sN266kJGR0Z7KP/8b6enzD4GT/598LTgSzngAUq6CzKd7vebRgco90rgeP7fz67MvhsrdULat/bXV/4RHpsM/xsHD06QBHTG56WmBUgAAIABJREFUffID6YmLmCI9XgXfSL3nXiGNdAcPXpEMywwdJ59Hz5Oeq8ynul3z1uvrS/43MtRlQi/XsgWEtW9LYAvPtDEh7fgePLMJ8v4nbyz0doP78XPlmsuO6/CsVrn/Ym/b3AF13fUQZl2EX3OtvG4Ai6aNovpYMzsO9f3H3FBDjWF9lC76DIou9hDNqZ2fO5PGGrkHXkcCR/QrRHNAmtg2OfcPbVv/NthZNHW2SQCnrAU0YqwMSwPPlqJc0c6DX+1h/j++56a3cnj7pwOMDvHnuStSePryFEaF+Lu6ef2nsUYmM+kYomdjXHKvjI6tP/0ow9sSzpXhkCB/6J/zOCz6Y7s3sK+MTZRhj5tegn9Nhg+ukuGUJ/0OYub1ra6gCAiPg9LNnV8v2ybD+iYtdPyMpsGVn8BNG6Wx10dycnKk8RQZL5PGFK6BpX+Rhl1HTvub9Bx9fmvvEoF05dAW+djVwEtcDl6+0osHsO0DWP0PmHkunPwXmbgkapbUs6uxPHGBNIi2vCm9m1NPk0ZgQ1XnO6A1RTByUudQ2Yw7oPaA3Oi1O1260twgfzAfzpVGktUqjcspp/Te4AI5jr18YcbZjt+nrlSGseqRt1JOUrMu7P2x/ILlGC3pcNOgco+8a2sLC+0D6rrrIcSfTqtPEGz/EICT4uXa1bUqTFON4W5QuugzKLo01khDI2SsvCFnxBq8riGa0ObB67uBNyBNmtuWoPiFgG+w69fg+QV1fr8rJdm93nPXiLEyLP3qtr2hFJJXfyzkuTX7OHG8D384fz6J40fg5+Omtn9tsfRSTUyH0F4ke9n7vXShdwzRszEuWf7Ibm5oP1F18C36QYYGJJzf+Y2UK/vYeB2W/kmGZUbEyRC86JT2TFV9JSZVrunqyIG2TJ16Bh7IhDP9xH4eTTtDerTGJELqtY4FA8Nl2OcHV0PWc3DiLX070KGtMt4/PK7z60ER8tg7PpQeys9ukqGtP3+pfa1ld0xIg61vScPnxFulkWXz8h0paM8QWVPk6MGctgzGJMgN35MubDf6Lc3w6Y3EF22BvUnSMPQNlH1SnCXX/YGcKMfPhWMVfQvDBblR+8yfOfabzaNWnKWf7Cf7ZdmeCX00zGJOkAZ08U8yUY19g/O+e/DUdddD8A2kMnIBUbs/h7MfYVRIAEnRYazJr+SWk/sYij7EUGNYH6WLPoOiS0OVvIHq5SXnWmd78Kyt8qZe1xDNwJFwZG+fqxuQJnYPXkibB2+QQzS79eDpGHjNDfDKWfIm/rJ/9HgII8aKm/6KVwwW3+ws429f7OLUmWO5fpYfcyeGu69xB/DZzdJQeGS6XDv21Z1Qe7D78nu+hODRbWvUujAuGYTVMTV+lzVdYyp+lHfH+uG16JHAkbD8WVj8R2lU9Ne4A/kd60qh7nD7awc2SC9b2LiBt7U7EpeDt79cd9edNyrhfJnUZdUDUHOgb/Uf3ir7yktnXM6+GEzlMknMiBi4+I2ejTvonAFy7hXycVRbiIstTFMIqC5yNPC8vKQXr3J3+xo/kMlPdnyIxSdIjqms52DNP+XkmPYruOwDmUHVlhDHL7Tn7Qq64u2jHzI7Nkne0ey6ZQXA4W3y9fkr9DU8HieskD8aXjod/vcnmVgoeHTncGfFkKN87CIZft6WIXfxtNHkHKylrmnga00UCoUTaahuD58MinT+Grymo4BwDNEMGDH4SVaa2ww8+xq8wU6y0jWLZttzvbWApdlyaUn1PmPbdhyGpQcvNTXV1U1wC7YW13Lru1uYHT2CJy+dS2uzc/bzMIzD22QYYNqvpQdj/2qZBXLb+3DhSzDl5M7lLc1toZXn6f+wtSda2QoT2jIC1hTB80ukx+vk/wcjJzCqdgvMvbJ3GS1dic2IPZQDYWfLMMADGxzXazkJ+3kUkwp/Kj2+cappcPbD0ij/eAVc+elxvaZ2WlvknnppN+i/P+2Mtj15hMyw2dvEH6Pi5WQYHgejp8vXRk6S6+1sBl5jDZiP6htUicth1f2w7hHpHd71qcyomnYjvif9BUJCpP6WRhnq2JHkS6TxaLUMzKDviLePHMMHddbh/fQ8+AS2G7J9YWwC3JQJ390LG5+Wr808p2/rQ9tQ113PYcqp18C+p6V3POFcFsWP5ulV+9iw9wjLkgy8WeTmqDGsj9JFn0HRpbFazmUgo0Oc7cGzGYx6IZr9WIM3IE06rcELNma9oR7H2yah4/sdOdB2s9WWqK0HjBgrbuyqMY7y8nJXN8GlHKpt5MGv9nDli1mMDvXnxatOINDP2/11yXxaeimW3AULfyvXj924QXrX3vy5DJnr6H078KO8C60Xngky82BQZOd1eD8+Li8i+9fAs+nwxnI0S5PMnunuRM2SWRZta6Yqdso7bLHGLPTuNF56Y6iMiIHznpJJO967XG4v0RMVu+VdMNu2Fl3x8YcrP4brvutbJlNNg0velh41G17ecqG6LZOm7cIcEefwcbx95Bgs3SzDHz/7jQxnPO2+dl28vByNu47Hd5ZxZ2Niuuzzjlt/NFTD9g/kthJd78D2loAw+NmjcPWX8jvOvqRf1bj99UVhp7yySm4Jkvc/aKojZVI4If4+rMnv3XqSoYoaw/ooXfQZFF0aqtpvbAZFyhuTzsRWn0OSlZHyBmZv5vEODEgT+xq8YNdk0ezWwNNpx8G25TK1B3u1jYIRY2VYGnjFxd0kIhjilNQ0cPPbOZz0r1U8v3YfJ00bxZvXpTE6VCZRcWtd6g7Ju8kpV3a+0EROgeu/kyGA3/8N3rxAeu2sVsj7Sqa1nbxEv05N65xopb5MJt2Ye4XcFiD9Zjici9kvss+bOrsE30CZFMOWaKWoh/V3A6Rf4yVxudx6Yd8P8ME10kMnhPQ0fnYLbHmrc3nbXnVdE6x0JHoejJ7W/fvdMXFBe1imjY6ZNLtukdCV5MtkZtUv75AG34WvgI+f686j1OvkzY73f9mevnrLm3L96PxuPKB9ITZDnmszf9avj7v19UXRieLiYph1kby5sucLfL29WDg1krX5lQg33fNpMFBjWB+liz6DokvHBChBBqzBsyVt0dsmAfrsxRuQJuYOIZqDmUWz220SAju/b6PVAsWbpAHY0iAzdfeAEWNlWBp4w5Gmllaufy2b1XsquC4jjrV/XPr/2XvvuDiuc///fegICdQbIIF6x5aQERaWZMu9xTVxmu0kN7bTndiJ49wkTnKvf9+bnjh24sSJ7SQuce+9qNqybCQbCaFeAQkQQqIIRD2/P54ddlmG7cvuwnm/XryWnZ2ZPXzOmWWfeRp//vwiJo/qw8MQbWz8q+TLLfla79eSh8I1D8KF/ycFOR65Gv6YJ56LKWd7DgWckCdeoo5WKRTS1QFLvy0fZuf/L3xnC5sX/jL6wzMtMhdJ1cmuLji4HoZPsi+6EUlO/wJc9GvY+YoY5Pcuhocugo//DW/9REJrLQ5/DMkZvQushIvRM8Sw62h1NjkfPtl+38QUaZGh4uCqByKv89AxYmQePyjGclcnfPSAGPjj50V2bIbYIytf1v7WpwBYNmMMlSda2Hu0n+6aGwwGz3R2SJSOFaJp5eAFehPmiS/AK7f13NanB8/xPIBeeAHT5lJkpV+raDaLseaemmBF6LgbeFUl4tWbfbk897fuQIgYlAbe9OmDrxLY/722gx1Vjfzpc6fzo4tnkzWit9ETEl3KXpAwR+un7IXAP2wsWptg00OS+9OXN0UpMf5u2yHG3qgpcufJW1n4CXnQ1S4epI8ehHlX9ywgMWwck+afGdz4+5PMfAlLPbZb/qbJ4evDE9R6KbgJVt4lrQOGjIRP/Rmu/afcfbQa04MY7BMW+F8cJFBGTZeqq3X7xdBLG9OjuXsvzrgJbt8N053FUiL6+TK5EM69C7a/KP+sTxyS4ipRwGD83I1Vpk+fLp+p86+RXOcTh1jmaJewZhC3SzBr2B6jiz1h18UqcmJ511JHitc90NDFqq2987i7c/Dci6wM7zkGHwlKk24P3jBnFc3+iChob+kdngl9F1mxqpnnOdIZTng38MKxVgZlkZXk5Bju6xYAb5dV8/D7B/jS0hzOmTWuz/2C1mXr0/DMV3pvn38tXHaPb0U17PjkUblLVOhDef2EZDHS5l0tHwZ95UBZWIVWXrlNPiyKvttrl5haL1ahlU8eFWMpJzzhmRACXc76Hiz+ilTjArkbOWyCjH3O5eLJqy6VCpT9hZXHd2y3o0WCF8+hUpA2usemiK+XM78Nhz6QFhDDJsKswEIqQ03EdTH4TPdcLfoSfPAXeOU2sj/3JFPGpLFu91G+UtRPHvUow6xhe4wu9oRdF/cCKJYnr6XO843Jvmg57vQAWt6qljqJUrEMOgvr/7afIZpBadLWKGGS8QniUdNdEm2TmOL92GBob5FCZe70VWTl4Ab57pB9hjz3odBKONbKoPTglZaWRnoI/UZ1wym+/3QJsyek88OLZnncNyhd6islFynrDLizAn50GO6slIbTW5+Gf5wnXhF/6eqED/4s57UuFl9JHuq92t+IXAkBrNsrxVjGze21S0ytl9HT5e5W8UPyPEz5dxAiXax/EiAf2nnXSWn2xippQ9DZ5jn/LtRYBl7tLvseeD4Q8fWiFFzxZ+lVd/adoS/mEiAR18XgM91zNTxbqgnvfhO2Pk3hlFF8tL+Ojs6uyA4wQpg1bI/RxZ6w62Ll23UXWRnZc7s/dHbIjfTWhp6FWlqOi3HnHkWTGpgHLyhNWpsk/w5cwiP7IUyzoy8PnmXguYyhq0vaEk0+U8aYNsYnAy8ca2VQGniDBa01tz1Zwqn2Lv702dNJTghTHllXlzSZ7uyAK+93lrBNHgrLvg+ffxrqK+BvyyXkzh/KnpeL48xvhWXoUmhlgfxe9L3wvEd/EhcPE0+TD+n0zIAMlIhy2hfkrlzJf5xrpa8KmuEgeZh4varLZM3Gmn4WqSPgK2/CwusjPRJDrFNws4R+v34Hy7LiONnWydbKfsy7MRgM9rgXQLE8eIH0wnPNpXM1SFz77LkSYJGVoGhrcnomLeOqPypp9hWimZAMqJ4GXu0umRerMN/wyT6FaIaDQWngjR07NtJD6Bc+Lj/B+j213H7BTKaN9e6u79alqwuevQl2veHbG330gORpXHC3VLV0Z/q5cNNq+X3jX3u/3hdaS+uD0TPCG2aW/yUJ/7R64bkRc+vFCtOcvDSgfmW+EhZdRk+D7CVS/dEqsNLfTbVHT4d9q8TQDMDAi7n10k8YXWKHHnMVFy+Vb0/Vs3z/7wDYsK+f+k9FGWYN22N0scdvXTra4MMHpLq0L7iHaFqPgRh4rl47VwOvpc6+v2yAHryg1krbSYlQgv714FlFVtxRStoIHSlx5gJa7REmO2o3jJjsU5GVcFxDg9LAmzEjgJLqMchjGw+RlhTPZxb7Vt2vW5cjH8OWJ+Dpr8CxvZ4POroT3vopTL8AFt3Y934jc2HGRZIX5OuH1+43JQer6HvhLbIx72oxTvsg5taLZeCFMf8OwqjL6V+QHLjSZ8W7GkYj1ZbR050hLnY98LwQc+ulnzC6xA695mrcHCj6HillT/P5kTvZsHdwGnhmDdtjdLHHb112vQav3g4H1vm2f3eI5qiejy1BGniuHqeW472bnIOE/iem+e3BC2qttDZGlwcPpG7D7jfh/Xvk+cENkDbWeWN6RI5EA3V2eHyLcFxDg9LAW79+faSHEHbqW9p5ecthLj8tk6HJvtXS6dZl15uSVBsXLz213BNILTrbxdOXOETu8Hr7Ij77Urnbc/A974PRGtb+BjImea+EGWZibr1MPw/Oul16zoWRsOky9wpZU631/Zt/ZzHa5YM2AA9ezK2XfsLoEjvYztWy22H4JG5IeIPiA8dp6xh8eXhmDdtjdLHHb10qiuXR12blLXUQn+T0ZqUOB1RgOXh9efCaj9uHaFrv56cHL6i10uaag2eT/xYuPBl4S2+VPsxv/wz2vOPIvyt0fh8ePlkqczdUenyLcFxDg9LAGww8t7mCU+1dfGvYGji6y7+Dd78BWYvh6r+LB+3V79vvt+ZX0oj6sj/CsL6rc3YzdaVUItr+svd9D74HFR9KT7ooKRIRMySmwsqf9CxgEkskD3MapxP7Mf/Owiq0Ep8sjcwNBoPkm+QuJ+fUdlraO9hS0Y+5NwbDYKByszz62luu+Zh41yxjIi5ejK5gQjRThruFaB63D9G09u3PPnitrjl4DqO2P3rheTLwrKJmY2bDkzdAfTlMcmmtNcLRRzcCeXiD0sAb6CV9tdY89uEhrhl3hInv/Tes/v98Oi45ORkaqyX3afr5Tk/Qx/+Gjx/tuXNFMaz7LeR9Vkra+0LSEJi2Ena8LHl+nlj3W6k+dPoXfDt3GBno6yVQwqpLwS0wfj7knBW+9+gLy4M3YnJAocFmvdhjdIkd+pyrrMUktdWTG1c1KMM0zRq2x+hij1+6dHXKdy/wPeyx+bgzLNMidWRwHryJpztzxjrapDVBXx68lIx+bpNg58GLcIgmiAf1ukfFwAbx4FkMdxh4XippmjYJIaKwsND7TjHMpoPH2VXdxLdTHUVSdr3h012OwsJCiSUGmHGBPJ79I8hdBi98A569WVodtJ2U0Mz0iXDRL/0b3OzLoPEIHN7c9z6HP4a970LhNzxfVP3EQF8vgRJWXSYsgFvWw9AIJO8Pmyh3B731wOsDs17sMbrEDn3OVZYUorp4eOWgLLRi1rA9Rhd7/NKlZrvTWPHVK2ZXAGXIqOBy8CYsEC9UV6dzWwhDNINaK61NEuEDLjl4YfbgtZ+SvzHJS6HCkblw3WOw8AYYN8+5PSNLUp68FFoJxzU0KA284uLiSA8hrDy28RAzkuvIrn5b/iG3N8Oet7weV1xcLOGZ6ZnOBRoXD595REIly56HexfDgxdI37gr/ux/GOCMCyAuAba/1Pc+Hz4AyemQb9M0PQIM9PUSKANWl7g4yTdadENAhw9YXYLE6BI79DlXY2ZC0jBWpB1g08HjtHZ09u/AIoxZw/YYXezxS5fKTfKo4nw3mpqP2Rh4QXjwUjJg5FTo6pCcMcvA8xSi6acHL+C1orWbB6+fqmiWPiPvMfMi7/vmLIXL73F68kBSjDKyvIZohuMaGpQGXlNTU6SHEDZONLfx8tYj3DV2LUrFwdX/kDs6ZS94PfZkw3HYu1pCM10LpqRkwHm/gG9/In21arbDmd8Wz56/pI6QsLvtLznLyrrS2S6VNmdeLOVno4CBvF6CYUDrctb3YNYlAR06oHUJAqNL7NDnXMXFQ+ZCZrbvpLWji48PDa48PLOG7TG62OOXLpXF8v1o5BQ/QjTrele4HDJKQjf9pcVRTGWES0ih5QkMoQcv4LXSdhLQ/VtFU2vYeL/k1+UuD/w8w723SgjHNTQoDbyBSEtbJ899XMGXH/6IlI5Glpx4BeZeJRfr7Mtg5+t9V8N0kFFfJvHW0y+w3yF9Alz6O7jjgBh8gTL7UvEAHt3R+7WD78kHzeww9r0zGAwGQ2Bk5TOsYSepqnVQ5uEZDGGhcrO0OPK1cElXl30BlNQRgXvwUkc4K0cfP+gSounBg9fW5Hvrq2BocxhASW4GXjg9eIc2QNUWKLg5uHZNIyLT7HxQGngFBQWRHkLIqG9p52cvbuOMu9/mu0+UUNvUxoMLthPfflJy2EBKuLafhD1vezzXvMRKqRw4xcudiuRhwS32WZcCyr6a5vaXpdLm1JWBnz/EDKT1EkqMLvYYXewxusQOHucqazGqq4PLxtQMujw8s4btMbrY47MurU1QUyYGnq9esdZ6Kb/vXmRlyCjoaPE/N62lTgy89CxQ8eLBs6px9hWi2d3s3PdKmgGvlVaHgWfl4MXFyXfFcHrwNt4vRuyCzwR3nuE50FTtcU7CcQ0NSgOvoqIi0kMICat31nDB79fyrw0HOG/OOP5z0xJWf3cp+VVPSBikVWI+5yy5A7PteefB7S3w6g+g+MHuuy9dO1+DnCJnbHO4GDZecgO3v9AzTLOrC3a8IpU2rQpJUcBAWS+hxuhij9HFHqNL7OBxrjLzATg/vZxPDp3gVPvgycMza9geo4s9Puty5BPQXXJtpWT4ZjBZxlevEE3Hc38LrVgevPgEyRnzJUQzxX8DL+C10tYoj67FTpKGhM+Dd6JcHA6Lbgj++6jlFT1xqM9dwnENDUoDr7LSc8PBaKSptYM9NU18Un6C9btruePpLdz40EcMS0ngua8v5XefOY0lU0YRV/qkJMcWftN5cHyChDzucoRpdrRJv44P/wovfxfuzYf37iGp8ZCzema4ybsOqrbCbpfiL4c3Q+NhCSmNImJxvfQHRhd7jC72GF1iB49zNXQMjMghT+2mrbNrUIVpmjVsj9HFHp91sQqsZC70vXBJX941y6Pnby+8luNOY3FEjoQUthyHuMS+K0haRfb8KLQS8FqxPHiuDojEtPBV0fzo74CGxV8N/lw+9MILxzWUEPIzGkLOqfZOVvx6FbVNbd3b4hTcsnwqt547nZRER8We2j3w2g/FOzb9/J4nmXMFbP6XtEEofVaqZV7yO7lT887/wFs/kf2mn9c/f9TpX4QN98Lbd4nHLi5eCq/EJfSfkWkwGAwG/8nMZ/TB90lLiufNsirOnhWBdiYGw0CholiMqrTRjhDNeolu8pQKY3nX7PrggX95eF1dYqRZnroRk2HnazB2jmzraxzdIZoBFHXxFysHL9ndg+cWotnZLpVIXStZWnR1ymveUozammHTw5JONDw7qGEDLr3w+jcPb1AaeDNnzoz0EPzitdIj1Da1cceFs5g5fihDkxOZODyFrBEubuP2FnjqBvHWXfNQ7wbNucvkQn3+G+LqPv9uWOxoQzDtPCh7jobKnaSPnNI/f1RCEqz8KTx1I5T8B077nDRAzynqOxwgQsTaeukvjC72GF3sMbrEDl7nKmsxqvRprpiieKOshruv0MTFBZGXHSOYNWyP0cWNri7Y8zYzp/moS+UmmLREfk/JkNy6Npeeb3Y09xE+aRl8/oRottYD2sXAy4GTR6G+ou/8O3CGaPrhwQt4rXR78Fw0SRzS24P395VSw+Hcu3qf428rYPQMuOqB3t+RXdn+ouRBFtwS2FjdGTpW8gU9ePDCcQ0NyhDN+Hgbyz6KefzDcnJGDeGW5VM4Z9Y4zsgd2dO4A3j1+1BdKgvX7o5DfKLcjWhrhBV3wpkuIZxxcTDvak6d/l/h/UPcmXOFJBWvulti0I/tibrwTIi99dJfGF3sMbrYY3SJHbzOlaPh+afGHKa2qZWPywdHuwSzhu0xurhR8SE8di2jXv2K91DJhiOSVuPIbfU5r83y0Nn1wQP/QjTdG5pbOWNHPum7gia4ePB8v/4DXitWDl4PD15azxy8rk6oKoX9a3sf31wnFTFLn4Z3fu75vSqKpRfz5DMDG6s7SsHwSZLX2AfhuIYGpYFXVlYW6SH4zL6jTXy4v45PL85G9eVW/vhR+PjfcNbtnkMsz/05XPc4LL/D9uV+10UpabfQUCk5gQAzA+s9Fk5iab30J0YXe4wu9hhdYgevczV+PsQnk6d2kxCneKusun8GFmHMGrbH6OKG44t84uFiePACj1/snfl3i+TR17y2ljqpdpmc0XO7ZaQFY+ANz3Gc45jniKoAPHgBr5VWtzYJ4PDguYRonjwq3s/qbWLsuVK1VR4nLoT3/iAhmH1RXQrj5gZXLd4dK6+xD8JxDYXVwFNKXaiU2qmU2qOU+mEf+3xaKVWmlNqmlHosnOOJet79Xyh9psemJ4rLiY9TXLMwq+e+bc0S2vjgRfDC16VS5tk/8nz+tFEw6+LQLtpgySmCGRfKws9aLL32DAaDwRC9JCTBhDySqz5myZRRvFlWFekRGQzRw4lyALYsuEvK4//9XDj8sf2+lcVSe2DCAnnua+sBy/hyDzWMTxSjz58cvL48eABDPBh4iSmQkOJ3s/OAcO+DB72raDY4CpV0tEDdvp7HWwbeZx+HaefCy9+Dve/2fh+txUAcNy90YwfJazx+sGfl+DDj1cBTSl2mlPLbEFRKxQP3ARcBc4DPKqXmuO0zHbgTWKq1ngvc6u/7BML48eP7423840Q5rP01PP1leP9PALR3dvHMpgpWzhrL2PQU57573obfzoLnboamKjj3Z3DdY/ZJpX4QMV3O/ZlUapp7VWTe3wtRuV6iAKOLPUYXe4wusYNPc5WVD4c/5vxZI9l39CR7aprCP7AIY9awPUYXN+oPQdoYkudcBF95Swy41++03/dIiRQzSUyV55YHz5vR1FzXu8CKxZCRfhp4jveyDLwhI52GlKcQTZDx+uHBC3ittDZJ1UxXg9a9imbDYefvVVt6Hl+1FYZNkDZd1zwEY2bBkzfCqYae+504CK0N4sELJSOnyHlPHrV9ORzXkC+G22eA3UqpXymlZvlx7jOAPVrrfVrrNuA/wKfc9vkqcJ/W+jiA1rrGj/MHTG5ubn+8jX/sWyWPk5fCmz+Gt3/GO2XV1Da1cd0ZLjl1nR3w2h1SbemGl+Fbm6Hou5CSHvQQIqbL2Nlw6xYouDky7++FqFwvUYDRxR6jiz1Gl9jBp7nKyoeOFi4aUwswKMI0zRq2x+jixolyyMgWXcbMlKIffYVpniiHkS76+Rr22HK87wIowyZINNiDF8KG+7o9ih7PBU4DTymnF89b0buU4X71wQt4rbQ19sy/g95VNBuOOH+3PHauz8fPl99T0uGCu6W4TPmHbvuVyqO1b6iwPIJHtti+HI5ryGsVTa31F5RS6cBngYeVUhp4CHhca93o4dBMwHVVVQDurdpnACil3gPigZ9prV93P5FS6ibgJoCJEyeyevVqAKZMmcKwYcMoKSkBYNSoUcydO5e1ayXBMiEhgaKiIjZv3kxDg1jp+fn5vPfeeyQnJwMwffp0kpOTKS2VSR07diwzZsxg/fr4fruwAAAgAElEQVT1ACQnJ1NYWEhxcTFNTXKHsqCggIqKiu6+FTNnziQ+Pr47hnb8+PHk5uayYcMGAFJTUykoKGDjxo20tLQAUFhYyP79+6mqktCWJeWvkThkLOtybmNG21Amrv89yWk7GJH8RfThMoqbysnPz2f38/+P6cf2UDr3h8zKLmTX9u3U1IhdPG/ePFpbW9m9ezcA2dnZjBs3juLiYgDS09NZuHAh69evp6OjA4Bly5axbds2jh07RlNTE0uXLqWxsZF9+8S9nZOTw8iRI9m8eTMAI0aMIC8vjzVr1qC1RinF8uXLKSkp4fhx+ZBYuHAhdXV1HDhwIKh5qq6upry8POLztGrVKlJSxIM6Z84cOjs72blzJwCZmZlkZWWxceNGAIYOHUp+fj4bNmygtbUVgKKiInbt2hWyeQLIy8uL+Dw1NTWxYsWKqJkn1+spkvP07rvvkpqaGjXzBNFxPTU1NXHeeef5PU+G/mfDhg2sWLHC807ZUvVvzPFPmJ85n7fKqvjaiqnhH1wE8UmXQYjRxY36chg316lL+kQJ1ezskErnFlqL18m1dkK3B8+L0dRwuG8j5Kq/wiePSeupN34Eb90FN74Ck9y/gjvorsg53LltRI7konmqomkd40eIps9r5eAGqT45yvGZ0naydz8+9yqaDZUSDTZmZk8Dr/0U1O6EmRc6t2UtlhzGQxtg+rnO7dWlgBLHQyix5qqqpOf7OQjLNaS19ukHGIWEUB4AXgN2A9/ysP81wN9dnn8RuNdtn5eB54BEIBcxCId7GseiRYt0sKxatSroc4SUzg6t/2+y1s/eIs+7unTDyz/W+q50veqhnzj362jT+g95Wv9lqdadnSEfRtTpEiUYXewxuthjdLEnUF2AYu3j/ynz08//I38/T+snvqj/+PYunfPDl3V1Q0vQ7x3NmGvbHqOLC11dWv/PWK1f/5FTl4/+ofVd6VqfqOi5b/Nx2f7ePc5tnR2y7d27fXoPr9Tu0frXM7T++/lynB2v3qH13Zk9t712p4xj2/Oez//ItVr/pcj7OBz4tFa6uuQ78b+v7vk+95/Vc7/Vv5QxdrTJ82e+Kp9Jz96i9a+nO/er/Fj2K3225/H3L9P6oUt6bnv8c1rfs9DXP8c/fj9f6yeut30pHP8ffcnBu1wp9Ryw2mGInaG1vgjIA27zcGgl4FqvP8uxzZUK4EWtdbvWej+wC5jubUzBYt1djxqOlIiLfOo58lwp/tB1Ha91Lmb5ofvg0AeyveQ/cHw/rPiR5x4eARJ1ukQJRhd7jC72GF3sMbrEDj7P1aRCOPQB588Zi9bwzvZ+ybKIGGYN22N0ceFkLXScgoxspy7pmfLomiPm+jx9onNbXLyU6PfkwWuuk/ewzuuJUVNh+Q+g/APY9Yb9Pi3He4didodohtaD59NaqdsnYzr4vjQuBymykuTWFzDR0S7MKrTScBiGTRRvWVM1NDk+jyxv3vgFPY+ftERaIljvAY4KmiEusGIxYUHv3EAH4biGfLESrgZ+r7Wer7X+tXbkyWmtm4GveDjuI2C6UipXKZUEXAe86LbP88AKAKXUaCRk0630TegpKOjDTR0prEo+U1YAsLu6kX9uOMgH83+BysiGp74kC3ftr2DCaTDzorAMI+p0iRKMLvYYXewxuthjdIkdfJ6rSUugqZqZSceYPGoIL2857P2YGMasYXuMLi7UH5LH4dlOXSwDrtHt+rCeuxtqKcM95+A1VMhjhg8GHsDC66XIxzs/790+ABz5fG4GXvZiCRcdNc3zuVOGQ4vvOXg+rRWrdUT7SaiUdAZa+8jBA2eYZsNh0bo7HHKr8zExDUa45blNWiIVN628uFMNkis5PkwG3vg8MV7dC7sQnmvIFwPvZ0B3FqJSKlUplQOgtX6nr4O01h3AN4E3gO3Ak1rrbUqpXyilLnfs9gZwTClVBqwCvq+19qP0T2BY+R1Rw77VsiCHjkFrzV0vbiMtOYHvXJIPn/6XVEP62wo4cQjO/u+wtTmIOl2iBKOLPUYXe4wu9hhdYgef52pSIQCq/AOuOj2L9/ceo7yu2ctBsYtZw/ZEhS6dHbDhz35VdAwLVkGTjGynLsMcBp4vHjyA1AzPHrx6RzBcelbf+7gSnwjn/BhqymDrU71ft/PgTTwdfnjIe+uq1OFSrMTOcLTBp7VSuUnaL6CcTcvbmmxy8NLksb3Zmc+YPtFpoLkaeOPn9Y58c+QRU+6Ikqtx9KIbF+ICKxZWK4zq0l4vheMa8sXAewrocnne6djmFa31q1rrGVrrqVrrux3bfqq1ftHxu9Zaf09rPcfhIfyPv39AIFiFGaKC1iYJwXSEZ76y9Qjv7z3G7efPYGRakiyIi38l7uasxZ4bmQdJVOkSRRhd7DG62GN0scfoEjv4PFejZ8od/EMbuHqReBOe2VwRxpFFFrOG7YkKXXa9Dm/cCTtetn+9bn//9CCrd6z/4dlOXYaMhPhkZ582C8vAG+pWIj/FS9ijdR5fPXgAc66ECXmw6m7oaO35mp2B5yu+FoWx3sqXtVJRLA3Jx8+H/WtkW2uTBw/eSdGro0UMvNQRkJEthl1XV88Kmq6kT4Dhk6XQCriEcobLg+cw8GwqaYbjGvLFwEvQ0uYAAMfvSSEfyWDl4HvQ1Q5Tz+Fkawd3v7KdORPS+VzBZOc+C2+AK/4CV/41upqUGwwGg2HwEhcnYU4HN5A1YghF00bzVHEFXV3918zXYACcnil3LxlIWNw9p0PZC+EfR325eJpSXCpSKiWGRy8PXiWkjYUEt6/UKd48eBVSLTJtrO/jiouDlXdJJFjxQz1fC8rAsxqzh8hz2tEmeWqZCyF3mbQxaG/pw4PnkoNnaTvM4XEcP18MthMHpcVCXxVHJy2BQxudDc5TMnzLbQyEYeMhbUyfeXihxhcD76hLSCVKqU8BteEbUvgpLCyM9BCc7F0lrujsJdy3ag9H6k/xi0/NJT7OxZBTCk77nLNcbJiIKl2iCKOLPUYXe4wu9hhdYge/5mrSEji2G07Wcm1+NpUnWtiwL+yZFhHBrGF7Iq5La6N48KC3lwygdg+gnfuEE0cPPJTqqUt6Zs8+beAMKXTHaw5epXif/C22N/Uc8YyVPu3cpnVwBp7VWsFHD57XtVJdCp1t0mczdzl0toqHrb0Zkt2KrCQ5QjTbTjq1tYyz8fPlc6niI+dzO7IL4GSN3ASoLpXwzHA5UpQSL56NBy8c15Avq+MW4EdKqUNKqXLgDiA6O1L7yP79+yM9BCd734XJS9lf38nf1+3nqoWZ5Od4qVoUJqJKlyjC6GKP0cUeo4s9RpfYwa+5cuThSTXNcaSnJPDER14aK8coZg3bE3FddrwiVSUTh9h78KyiJHvfDX+YZv0hGC4F5Hvokj7BPkTTzlvkrTJlfaXv+XeuKCWpPtVlzpy51kbQncF78HzMffS6VqwCK5mLYHKh9Krb+Zps8+jBs/ISHQbz+Pmgu2DLE6DiYOwc+/fr/vzaILqEKzzTYsICOLq9V5hsOK4hrwae1nqv1noJMAeYrbU+U2u9J+Qj6UesJrsRp75Smi9OPYf/fbmMpIQ4fnjhrIgNJ2p0iTKMLvYYXewxuthjdIkd/JqriadLftGhDaQkxnPF6Zm8vq2K+uZ278fGGGYN2xNxXbY+BcMnicfHzsCzipI0VTsLaYSL+grx4OGmS/pEaDzS08Ds04OXIUZLR1vv10AMVn/y71wZP1+qU9Y5DIqW4/IYtAfPNwPP61qp3CRhjBnZ4rHLXOQ08Hrl4FkevGbRFiVhkOBsdbD3XRg9AxL7aEMwZpboXfIf0SVcLRIsxi+Arg6o2d5jcziuIZ/8u0qpS4CvA99TSv1UKfXTkI9kMLJvNQAfxuXxzo4avr1yGmPTUyI7JoPBYDD4hVIqTSkV5/h9hqN/bGKkx9UvJCRLvoyjX+un87Np6+jixRKbUDmDIVgOrIf2U87nJ2sl1WXe1WL02IVo1lc4vT9WW6q+OH7QmZPlL61NYjBl2HjX0jMl9LDZEb5sFQbpK0QT7MMeu7okHDHQPDHLQ1XtKCjSUieP/eTB80rlJjHqrDDJ3GWS1wgePHgnZd6HjpWKoSDFU5LTxYvXV3gmSJhrdgEcWCfPw+7By5PHfsjD86XR+f3AZ4BvAQq4Fpjs8aAoZ86cPly1/c3hzejkYdyxroMpo9O48cxc78eEkajRJcowuthjdLHH6GLPANdlLZCilMoE3gS+CDwc0REFgd9zNWkJHPkE2pqZOzGd2RPSebJ44FXTHOBrOGD6TZeaHfDwJfDkF53erW3PSYjh/GvFWGo57uyLZtFQCePmirfGm4H36u3w4Pnw8KXdNy18xjJEhk8C3HSxin9YBqh7zpgrnipTnqyRwnx2RqQvjJktYY9VjlL9/ezB87hWTtVD7S7IzHduy13m/N3dwHPtg9dw2KkxiOFmeeM8GXggBh6ILmNme943WEbkSsN2tzy8cFxDvnjwztRaXw8c11r/HChEGpLHLJ2dvvXrCDtVpVSlTGP/sWZ+etkckhL8TJgNMVGjS5RhdLHH6GKP0cWeAa6L0lo3A1cBf9ZaXwvM9XiAUg8qpWqUUr2bIsnrK5RS9UqpTxw//RY54/dcTSqUsKPKTSil+HR+Flsr69lV3RieAUaIAb6GA6bfdDnyiTzufhOev0XyyLY+LflV4+Y6+801uhUzqa8QQ2rqOXDwfanK2BdHd0hIX+0uePACePRaKcDhCy498MBNF8uQs0JIu3PGbPrMeTKaunvgBejBS0yRv89qCRCsgZc4BOISfPbgeVwrhz+Wx8yFzm3ZZ0gIOPQO0XTtg2fn1bQMO28GnpWHN3q66BNO4uLES+jmwQvHNeSLRWH5wpuVUhOBdsBL58PoZufOnZEeAnR10VW1lXdOjOPc2WNZMdOPcrdhIip0iUKMLvYYXewxutgzwHVRSqlC4PPAK45t8V6OeRi40Ms+67TWpzl+fhHkGH3G77nKPkMeHR6PSxZMQCl4desRDwfFHgN8DQdMv+lStVWqjp/zYyh9Bp7+sjSpnne1vG6FO7qGaVoNsDOyxMDrOCVGnh3tLWKkzbsavvMJnPszCde8/yz4+BHvYZvdHjwx8Hroku7W7Ly7ybknD56N0WQVjAk0Bw/EwKh29+AFWNxPKe99+1zwuFYqiuXR1cBLTHV+vrh78OITID7JUUWzsrexPGUFJGfAhNM8DypzobSdCHf+ncX4BeJBdWkOH45ryBcD7yWl1HDg18Bm4ADwWMhHMsjYt7uUuPaTbO+axI8vMWEfBoPBEMPcCtwJPKe13qaUmgKs8nSA1notUNcfgws7qSMktKlcDLyxw1JYnDNywBl4hn6g/RQ8/w04trf3a1VbxFu37Puw9DtQ9rxsn3+NPLp7yUBy9DpbxcCbfKYYBH2FadbtBzSMmiYFPIq+C197TwoJvfANCQ1t9nDJ1peLoeDeuBwkP0zF9/bgDbPxl3jKa+v24AUYogni0WqolL+l28Ab7vkYTwwZ2bsFRCBUbhbt3b2Jucvl0b1NAogHsfmYfT7jrIvhjgMyPk8kpsLVf4dltwc8dL+YsMBR6MZHz3CAeDTwHEnj72itT2itn0Fy72ZprWO6yEpmZpiaGPrI66VH+OOjzwHwhSsuI2d0WkTHYxFpXaIVo4s9Rhd7jC72DGRdtNZrtNaXa61/6fi/Wau1/nYITl2olCpRSr2mlPIY8hlKApqr7MXSc6qrC4CL541nV3UTe2oGTpjmQF7DwRBSXQ6sh08ekfL2rmgtXg8r3O7cn8OZ34aF18OIHNmW7pbnBk6PV3qmGG2TCqUoix3HdsvjqGnObcOz4foX4LxfwM7X4d9X9u3JO1EuRoajP10PXeLipcKjZeA1HhFDxsojc8VTb7mGSvFiejNaPGF5qqq2ihGZmCbFkgJl6jmwbxWc9N7/ss+1ojVUFkuBFXfyvwzn/y+MnNL7taQ0580AO2+or70C514BY8Ocf2cxfoE8uoRphuOzJcHTi1rrLqXUfcDpjuetQKunY2KBrKwg7nwEyb3v7uY3b+7iVyMPo1vimb3gjIiNxZ1I6hLNGF3sMbrYY3SxZyDropR6DOkZ2wl8BKQrpf6otf51EKfdDEzWWjcppS4Gngem9/H+NwE3AUycOJHVq1cDMGXKFIYNG0ZJSQkAo0aNYu7cuaxduxaAhIQEioqK2Lx5Mw0NDQDk5+fT2dnZfY7p06eTnJxMaamEdI0dO5YZM2awfv16AJKTkyksLGR/x1hyT9Xz4WuPMv+ca5iZdgoF3PfiBn5wyXzi4+MpK5MS9ePHjyc3N5cNGzYAkJqaSkFBARs3bqSlRfKjCgsL2b9/f3f58Dlz5tDZ2dkdypSZmUlWVhYbN24EYOjQoeTn57NhwwZaW+VrSlFREbt27aKmpgaAefPm0drayu7d8kU+OzubcePGUVwsoWHp6eksXLiQ9evX09HRAcCyZcvYtm0bx44dQ2vN6NGjaWxsZN8+ufuek5PDyJEj2bx5MwAjRowgLy+PNWvWoLVGKcXy5cspKSnh+HHxlixcuJC6ujoOHDgQ1DxVV1dTXl7u1zwVFxfT1NQEQEFBARUVFVRWikE0c+ZM4uPjOfDRGwxpriBh/pU+zVNra2v3egl2nqbsfYRJwKnda/hAndk9T+OHdJLWUseuxhSaNm+WeRpyAR1JHbB6tczTzn3MShhKzfaPGDLvOI2NjTRsfIN5wOGTcQxtaOCoymFqzRq2bXyHuQUre85TnHQAW1d2mM5dJ9zmKY/5Z/6EUet/ypZnf0fdqEW95mlh+TbSR0zqnietNRMmTOiep9MZSnLtfppqa2HfFlLiMjhUVtZ7nvIlpHBf2cccOrm6xzzN2fMxI9PGUXf0aMDXU2JbI0uBht3vc3J/GSPiUinfvTvg6+lw5xwWd7ZR++69pJx9m8frafHixWzdupVjx8QYzMvLo7GxkcqyDylsquZYag6JDQ29r6f2Beg1a3pdT2e0Q0rtLuKAT/bVcOL46qi7ntzn6YPt1ZylEqja/AYT513Nxo0baW5uprKy0u/PPY9orT3+AL8BrkaSyL3uH+6fRYsW6WBZtWpV0OcIhKr6Fp3zw5f11x4p1h2PfFrrewsiMo6+iJQu0Y7RxR6jiz1GF3sC1QUo1lHwv8fTD/CJ4/HzwG+BRGCLD8flAKU+vscBYLS3/SL2P7Jmh9Z3pWu96V/dm675y3v6gt+vCXo80cKguLZb6rX+3TyZyyNbfDokpLr8dbm89/9O0Lqj3bl9x2uy/eAGz8ffV6j1Y9c5n39wvxzXWCPPD5fI848f633sc1/T+tcz+j53e6vWv52t9YMX2b/+m1laP3tL99NeujzxRa3/lC+/33+W1o9c0/d7/WKM1m/+pPf2B1Zq/fClfR/nK7+eLmN97Dqt/7w0+PM9cK78bV1dHnfrc62UPCHzUv6Rf+97/1ly3F3pWh/d7d+xkWTbCz3GG47/j774Lm8GngJalVINSqlGpVSDD8cZ3Hip5DBaw23nzyS+ujT8/TYMBoPB0B8kOvreXQG8qLVuBwJopOVEKTVeKWkGpZQ6A0mp8B4DFSlGTZfcoYoPuzddNG8CO6oa2Xe0KYIDM/jFmz+WsMakofDu3Z731Rre/xNJrSFKJW05AUdKYPRMyVGyerWBs+rjOC+RyukTe4Zo1ldIFca00Y7j50kjbbs8vNrdUkmxLxKSoPCbcPA9KP+w52ud7RJ26SiwYsuwiZJDZxV+seuBZ5GS0XcOXjD5dxbj5jlCNI8Hl39nsegGqTzqb2sJED023CshmN4KoriT6JLiZFeRNFqZczmMnuZ9vyDwauBprYdpreO01kla63TH8/SwjirMDB061PtOYeCFTw4zPzODqWlt8gHqrXRrPxMpXaIdo4s9Rhd7jC72DHBd/op42NKAtUqpyYDHG6FKqceBDcBMpVSFUuorSqlblFK3OHa5BihVSpUA9wDXOe7Yhp2A5iouDrIW9/jie9F8KTbxWmlVqIYWUaJiDXd1wRNfgP1rQ3/u3W/B5n/Cmd+SAiO7XoPyj/re/+hOePPHZJ8I4Eu9HQffl8bUVrGLQy5haFVbxACwK7ThSvrEngU/Giplm9U4Oy5OinbsX9s7l+7YHhg11fP5F14vNzLW/6Hn9oZKQPfoT9drvaRPFMO1+RicPOps62BHqk1lys4OaKoKroKmxfh50hKiqTrwFgmuzL1SGotv/qfH3WyvoT1vi2Ff9D2pjOkPVg5jSobk48Uo4fhs8aXR+TK7n5CPpB/Jz8/3vlOI2Xu0ia2V9XzqtInO8rRRZuBFQpdYwOhij9HFHqOLPQNZF631PVrrTK31xY7ImYPA2V6O+azWeoLWOlFrnaW1/ofW+n6t9f2O1+/VWs/VWudprZdorfuo7R56Ap6r7DPkS6PD8zAhI5WFk4bzypaBUU0zKtZwUzVsfwl2vRHa87Ychxe/Jc3AV/wICm4RT9e7HrpzHN0OQPbwxNCMYf9aKSAy51PipSp3MRyrS337zpSeKc3ArUbo9ZW9m4JPWiKG0olDzm3NddBSJ55oTyQPhYKbYecr0njdwq0HHtisF8tjV7mp53M7UjJ6F1lpPCIGcKA98FwZv0AaptftC42Bl5Qm1Uy3PeeszGlDL020hjW/Et0WfMb/9010GHiejOUYIByfLb6EaH7f5ecnwEvAz0I+kn7ESkTtT1785DBKwaULJrqEGkSXgRcJXWIBo4s9Rhd7jC72DGRdlFIZSqnfKaWKHT+/Rbx5MUnAc5W1WB4ri7s3XTx/AmVHGjhQezIEI4ssUbGGrSqM9RWhPe9rP4SmGrjyfmn2nDwUzrpNjK59q+2PcRg4tfu32r/uLwfWifGVkAyTCsSDpzW0Nooh4pOB59bsvKGyt0GUXSCPrmGWx6TASo8Kmn1xxs2QkArv3+Pc1t0Db1L3pl7rxRpHxUc9x2pHyvDeIZpW6Km7wRoIrj3fQmHgASy6UfoMbnmqz116aXJgnYR1L/2OhMD6i+W186RlDBCOzxZfQjQvc/k5D5gH9G2exwBWRaD+QmvNiyWHWZI7ivEZKWLgDR0PQ8f06zi80d+6xApGF3uMLvYYXewZ4Lo8CDQCn3b8NAAPRXREQRDwXGUuAlSPsL6L5ktezKulse/Fi4o1bJX9D6WBV70NtvwHim6Vnm8Wi74knrR3fmHfGuCoGHgJLbXBj+FkrXjpcs6S59lLoPGwGE7V22SbLzfFXRuKd3U6m5y7MnaO5BiWu4SA1jpaJHjKwbNIGyU5Z1ueEGOm/ZRzPlyMyV7rxcoRsxp6e/LE2XnwbN4jYEZNk9xECJ2BNyFPcug2/7PPVhK9NFn7Gxg6Dk7/YmDvaXnwYin/zoZwfLb42CCiBxVAPzWLGBhsraxnf+1Jrjjd8cFTZQqsGAwGwwBiqtb6Lq31PsfPzwGbpk0DnJR0+fLsUmglc3gq+ZNH8HRxBf2UQjiwcW+UHQo2PSxf9gu/2XN7YgqsuENCCve83fs4h4GX1BaCIisHpPw8uY4MoEkOL9uhD5xRT76GaILo01gFurN3zlp8gtyMcA0BPbYH4hJ6eOA8cua3Jazw2f+C386ETx6FtLGiWV9YTc0rNzvG6mcOXrcHLwQGXnwCjJsjvwfTU8+dhdeLoW4Z5Z4o/xD2rxEtPenmCSsHLxRG7wDDlxy8Pyml7nH83AusQ/rzxCxFRUX9+n4vfHKYpPg4Lpw7QeLCj+6Iuvw76H9dYgWjiz1GF3uMLvYMcF1alFLdf6BSainQEsHxBEVQc5W9GCo2dTc8B/hcwST21Z5kw97oLQLqC1Gxhq0v+Y1VzjyzYGhrhpInJO/N7ot+3mfFS+Ju4HW0dYc1pnbU993821f2rxWvmuVBHDsXkoY5DLwtkDrStzA8Vw+epZVd1cnsAjFCWhvl+bE90jA93sd8woxM+NZm+OJzMG2lvJ9bhc9e6yUhWfIaW+vlb0vxUK/Q8uC56lpf6Tguw7cxesMK0wyVBw9g5kXy2EdYbw9N1v1O5jX/S4G/X+LACNEMx2eLLx68YmCT42cDcIfW+gshH0k/smvXrtCesKNV7kTYfMB1dmleKjnMipljyBiSCLU7JbE1Cg28kOsyQDC62GN0scfoYs8A1+UW4D6l1AGl1AHgXqTFUEwS1FxlnSFfYGt3dm+6eP4Ehg9J5NGNhzwcGP1ExRq2PHhoCWEMlm3PynwtutH+9fhEMboqintur9sHXR0wfj6qvRlag+yedWAdTCp0GljxCZCVL2GUVY4CK1YlTE+kpIuh2HDYGdJo5/GaVCAFS6yCJ8f2eC+w4k5cHEw9B655EL6/Bz7zSI+XbdeL5cXzZpCkDJfxWQYoiMEaCu+dxfgF8hhKAy99oui4f43ty92adHXJPvOvCa76ZdLAKLISjs8WXwy8p4FHtNb/1Fo/CnyglBoS8pH0IzU1NaE94UvfgX+cB39bAXve6WHobdx3jJrGVj51muOijNICKxAGXQYIRhd7jC72GF3sGci6aK1LtNZ5wAJggdb6dOCcCA8rYIKaq+wz5NGlgEVKYjzXLsrijW1V1DScCnJ0kSMq1nDDYQklBPHoBMumh6Xv3OQz+94nc5F40Tpc8oQcFTSZskIeG4NohdFwRHqo5boVaJ+0RLxs1dv8uylu9cLr9uDZGEWZ+Ui+6IdibBzb671FgidSMqQwjQu268Uai1cDz+Glc83Dq68IbSjijAtgytneewv6y5Tl0vKis73XS92aHN8P7c09i70EQncOXmwbeOH4bPHFwHsHSHV5ngrYBGMPUspegJLHYfblUmL3kavg4UuhVkIXXtl6hCFJ8aycPVb2r9oq1ZeC+SAxGAwGQ9ShtW7QWluujO9FdDCRYtQ08Qi45OEBfK5gMh1dmieLyyM0sAFCQ6XT8xJsoZWqUqnouOhGz96xrFISpMIAACAASURBVMXQ2Sb7W9TsAOXoKQfOqpWB0J1/d1bP7dkFgIbO1gAMvMOOkMah9iGNqcNh7GwJAa0vl/fwpcBKsFiGiDeDxGo+7pqHF2oP3ojJcP3zofXggayJtiand9SOmjJ5DNbAm7wUZl7sW/XTQYYvBl6K1rrJeuL4PaY9ePPmhajASWMVvOSoOnXNg/DNYrj4N1C9FV77Pl1dmrfKqlkxcwwpifFyTNVWSWyNiw/NGEJIyHQZYBhd7DG62GN0sWcQ6uJDPFl0EtRcKSVhmm4NsnNHp1E0bTSPf1hOZ1dsFluJ+Bru6hJvl+UlrQ/SWN70kBRXybvO835Zjh5dFS5zenSH5KyNyJXnwXjwDqyF5Ayn4er6vsrxXckvAy/TkYPn8Hj1ZbxmF8jfZFXQDLGRYLte0v0I0QRnq4SOVmmObpdPGG3kFAEK9vUO0+zWpLpM9hk7K7j3GjcHPvt44EVaooRwfLb4YuCdVEottJ4opRYRw8njEKJypFrDC98UF/OVf5O48YRkOOOrsOQbsPddtpVtoaaxlfPnjJdj2prhSElU5t9BlJSAjkKMLvYYXewxutgzCHWJTSuGEMxV9mLJwWvuWV3x8wWTqDzRwuqdURDqGAARX8Mnj0oO/6hpUpwimEqabSdhy5Mw9wrvVRTTJ0qOk0t/Q47ukKbowxzfb4Lx4B18HyYX9r7xnTxMKo7HJ8HoGb6fL32is5G5J49XdoHkDu54SZ77m4PnBdv1EmiIZigraIabISNhwgLbPLxuTapLYWRucPl3A4hItUm4FXhKKbVOKbUeeAL4ppdjoprdu3cHf5JND8Get+C8X8AYtw+e078AKo7G9/9BQpzi7FmO8MzN/5IPkwVe7pZFiJDoMgAxuthjdLHH6GLPQNRFKdWolGqw+WkEYjYpJOi5mn6+PG59usfmc+eMY+yw5JgtthLxNdydUzZRersFE6K55Qn5PtJXcRV3shY5C61YFTTHzILkoXTEDwncg9dUI+fqKwcw/8vSj8/X6pYg+uguyd3zlLNmeUJLn5XqlEPH+v4ePmC7XqyefBnZng92D9G0Wit4Oy5ayF0u+Y1tJ3ts7takpkxaqhiA8Hy2+NLo/CNgFvA1pFLYbK21h8DaQcLq/4PJRbD4q71fy8hEz7iAWUdeoGhKBhmpifKB+P49MOlMuVNlMBgMhphGaz1Ma51u8zNMa50Q6fFFjAl58uPW8DgxPo7rFmezamcNO6qCrLo4GLEqaKZnyhf9QA28wx/DG/8tobSTfPw+krVYCmOcPOasoDlWWiK3Jo8M3IN38H15nNSHgbfoRrj4V/6d0zLqujo8G0Qjp8CQ0WLojprqW5XOYJlcBFc9IMVNPOHqwWttgrfuEoPIUzGcaGLKcvE2H9rQ+7W2ZilqE2z+ncEjvvTB+waQprUu1VqXAkOVUl8P/9DCR3Z2kHdAOlqhqVqqR8XZS3h46nWM1Cf48mhHIumWJ+Tu21m3BffeYSRoXQYoRhd7jC72GF3sMbrEDiGZK6vh8eGPe2z+0tJcRgxJ4kfPbqUrxnLxIr6Gexh4WYFV0ayvgMeugyGjpKy/r0ZNpiMPr7LYWUFzzEwAVPqE3h68jlZ46kY4/Inn8x7aIJUQJ+T5/jd4wzX80VNIo1JSqRPCUmDFdr3ExcGCT0sbCE8kZwBKcvDW/J/kE176B/88mZFkUiHEJfbKw8vOzpbwXrSz0bohLJ8tvoRoflVr3V3GR2t9HLBxW8UO48aNC+4E1geZFXtuw3MNs6jQoyk4/hJ0dcL630sC8bSVwb13GAlalwGK0cUeo4s9Rhd7jC6xQ0jmav61UjF68z97bB6RlsR/XzybzYdO8NiHsRWqGfE13FAp+WhDRonh0lrfs4y+N041wKOfltoBn3sShvnx90w8TQqeVBRLBU1Ud15c4sjs3h68qlLY9hy883PP5z34nhRTSUjyfSzecA3L9NZWwArTDEMVxqDWS1wcJKeLAbzhz7DwBundFyskpYm2bnl448aNk9BZMB48F8Lx2eKLgRevlPMWj1IqHvDpSlRKXaiU2qmU2qOU+qHN6zcqpY4qpT5x/PyX70MPnOLiYu87eaLbwJvQ5y5vbK9l7dALST64Bt77A9TtFe9df4QABEjQugxQjC72GF3sMbrYY3SJHUIyVykZMPdKycNrberx0lULMzlz6ih++fqOmOqLF/E13FAp3zvi4py5XL568bSGZ74ixW8+/U//vSdJaRIiWFnsrKCZKB20jjQi34tcwnG7vXx734UjW+zPeapeDMHJS/0bizdSR0CCo6pihpeqk9Z7O8JNQ0nQ6yU1QxrApw6Hc38WiiH1L7nLZe5dii0VFxdL/l1CqqwhAxCezxZfDLzXgSeUUiuVUiuBx4HXvB3kMATvAy4C5gCfVUrZfaI8obU+zfHzdz/GHjmsO1V9ePAqT7SwtbKezrwvyB2vd34h1ZlmX9aPgzQYDAaDIYIsvF76YZU932OzUoq7r5xPa0cXP3+5LEKDiwJKn7HPo2s/Bb+bC5v/3XN7w2GnR8rKLfO1kmbtLtj9JpzzY5h6TmDjzcqHik1Qs72HQdSaPFL65LUcd+5bs11aMCQNk/oDdhzaCGjf8wB9RSmXfnNePHhZ+XDjqzDr0tCOIRRYeXjn3+290mk0MmU5oGH/2p7bq7fJ+onCdmEDCV8MvDuAd5ECK7cAW+nZ+LwvzgD2aK33aa3bgP8Anwp0oKEkPT09uBN48eC9uU1eL1q4AGZcKBuLbo36xRy0LgMUo4s9Rhd7jC72GF1ih5DN1aQlEsa36Z+9Xsodnca3zp7GK1uOsGpHbLRNCOkabqyCp78Ma3/d+7XyjZJztev1nttdG11bhouvvfAOviePsy8PbLwgxlBrvXgBxzj7l8UPd4zFyhEE8dKMmQmLbpAqlccP9j7fofchLkEKuISa9Ezx5CX50LY5Z2lYvp8FvV4m5MH0C7z3KYxWMhdJo/m973ZvSk9PFwPP5N/1IBz/H32potkFbAQOIEbbOcB2H86dCbh+8lQ4trlztVJqi1LqaaVUv2QwL1y40PtOnmg8IsmjfdxReWNbFdPHDiV3dBqs+KHETs//dHDv2Q8ErcsAxehij9HFHqOLPUaX2CFkc6WUePEqPhSPjhs3L5vCHelv8dpLT8ZEwZWQruH96+Rxz7s9QxsB9q2Wx4qPnK9p7fDgOTxTw8ZLhJCvlTQPvg9Dx0vlyECxCq1ADwNvygJHZUfXQis1OySkc8nXZR188GebMW2Aiaf7ZoT5S+5ymHZu6M/rB0Gvl0/dB597IqpTezwSnyie0dJnuhu2L5yRBc21Jv/OjXD8f+yzjI9SagbwWcdPLdL/Dq21l9qufvES8LjWulUpdTPwT8SAdB/LTcBNABMnTmT16tUATJkyhWHDhlFSUgLAqFGjmDt3LmvXijs4ISGBoqIiNm/eTEODlGTOz89nw4YNJCZKJaLp06eTnJxMaWkpAGPHjmXGjBmsX78egOTkZAoLCykuLqapSfIIik5UoFNH894aSR6dOXMm8fHxlJWV0dim+XB/MzedldM9ztRxn6cgIYmNGzfS0iI94gsLC9m/fz9VVfKBOGfOHDo7O9m5cycAmZmZZGVlsXHjRgCGDh3aPXarIWJRURG7du2ipkbufs6bN4/W1tbufhrZ2dmMGzeuO7Y3PT2dhQsXsn79ejo6OgBYtmwZ27Zt49ixYzQ1NbF06VIaGxvZt28fADk5OYwcOZLNm6UHy4gRI8jLy2PNmjVorVFKsXz5ckpKSjh+XMIzFi5cSF1dHQcOHAhqnqqrqykvLw94ngoKCqioqKCysrLXPAGMHz+e3NxcNmyQMr6pqakUFBT0mqc1a9aQlJQUNfMEkJeXF/F5ampqYsWKFVEzT9FyPb311lskJydHzTxBdFxPTU1NnHfeeX7Pk6H/Wb9+PUVFRaE52YLr4O2fw3t/hCvv7/FSUtlTfK3tISpaR/NW6RVcsMBLvlSECakuVvGJ+kPSB861iqNl4DVViwE3PBuaj0kYpOW5i4uX311z8Nqa4W8r4Ow7Jf/RQmtHM/EzgzMWRs+Qwh+tDTDWaeB9tLOCxeBMX2k5AY2HJQwvI1NucG/+Fyy/w3ljvL0FKjfBkq8FPh5PLP9+eM7rByFZL7Fq3FkUfgO2/Ac2PQxFt1L6zhPMA9MDz42QfrZYaK1tf4AuYA0wzWXbvr72tzm+EHjD5fmdwJ0e9o8H6r2dd9GiRTpYVq1aFdwJHr5M6wfOtX3piY8O6cl3vKy3lJ8I7j0iQNC6DFCMLvYYXewxutgTqC5Asfbx/475iZL/ke689TOt70rXeuPfnNvq9mt9d6bu+s0sre9K17/81S90V1dXaN83xIRUlz8s0PovS0WXDX9xbm+u0/quDK0f/Yy8tvVp2X74E3le9qJz339coPWDFzufb39Z9nn4sp7vVXegt/6B8s/LZXxtzd2b1rz9hpx/9a9kw8EN8nzn6/K8apvj9V86z7N/nWzb8VrwY4pSzP8CBw9fqvVvZmnd0aZ3/+tWmfemo5EeVVQRjv+PnkI0rwKOAKuUUg84Cqz4cyvhI2C6UipXKZUEXAe86LqDUso1ie1yfAv9jDyNVX0WWHlzWxWZw1OZl2nyTQwGg8Fg4Jwfw4yL4LU7JB+nswOevRmUQn35NRqGTuHSxqd4d3t1pEfaP5w4BMcPwGmfh5FTYe87ztf2rwO0eD4SUqQtATg9da5FQzKyeubg7XhFHg+s71G5sLuZeCiaZC+8HvK/1F1BE6ArPkny3SwPXo2jcI5ViGXcHKlHsO638PGjLmNSsVX63xAYhd8Sj+6250g7eQCGjoO00ZEe1YCnTwNPa/281vo6YBawCrgVGKuU+otS6nxvJ9ZadwDfBN5ADLcntdbblFK/UEpZWb7fVkptU0qVAN8Gbgzuz/GNZcuWBXeCxirbAisnWztYu7uW8+eOQ8WgWz1oXQYoRhd7jC72GF3sMbrEDiGfq7h4uPoBydt66kZ49TYo/wAu+S2MyGHI2d9jTtxB1r3xpBXRE5WETBcr/y7nLOmNu3+dVM4ECc9MGioFaiaeLnl44KyW6W7gNRyGri7pt7vrdQl9052w81Xnfgffg5ThMCYErQDmXQ2X/r7HpmXLlsl3IisHr2a7/A0ZLmUVLr9X+qK98HV48Vvyd46bK4bhAMV85jmYdi6Mngnv/4nx6pgJz7QhHGvFlyIrJ7XWj2mtLwOygI+Ryppe0Vq/qrWeobWeqrW+27Htp1rrFx2/36m1nqu1ztNan6213hHE3+Iz27ZtC/zgtpNSRcrGg7d651HaOrq4YG7fDdCjmaB0GcAYXewxuthjdLHH6BI7hGWukofBZx+XAmWbHhZDYf61ACTkfYbm5LGcX/c463bXhv69Q0RAutTukXwzVw6sk2blY+fA1JXQ0SINrUEMn5wiKVCRlQ9HSqCjVQy5uARIG+M8T3omdLXDyRqpvNl8DJbdDhmToMwlYMrKv4vzpXC6/2zbts1h4FkevO1izLve6B46Br74vPQD3vwvMTpD3R4hyjCfeQ7i4qDw61C1BXWkRAx7Qw/CsVb8utq11se11n/TWq8M+Uj6EatQRUB4aJHwxrYqRqYlsTgnBvuVEKQuAxijiz1GF3uMLvYYXWKHsM3ViMlSFXDBdXDJ75wGQEISSUXf4Mz4Ml55/eWo9eL5rUtrI9xfBC9/17lNa/HY5RTJF9+cIjF6974joZt1e2GKo5Zd1mIprFK1VQy8YRN7GmmWh6y+UsIz45Ng2nnSc3ffKjjVIN9Z6vaGJjyzD44dO9bbg2fXODwuHlb+FD77Hxg1TYz8AYz5zHNhwXUwxBGWaQy8XoRjrYTnds5AptvA6+mla+voYtWOGs6dPZb4uNgLzzQYDAaDIexk5cNVf4XU4T02Jyz+Mm0Jw1hx9FF+9Upp1Bp5frF/rXjnSv4DVVKxluP7pcddzlnyPNkRjrnnXWf1zCkr5DHrDHks/1BCNK0WCRYZjqqj9YfEwMtdBinpMOdyMQx3vxna/DtPDBsvVT8bq6UMvqcwvJkXwbc2weSB7cEzuJCYAmd8VX4fPz+yYxkkDEoDLy8vL/CDrRAENw/e+3traWztiNnwTAhSlwGM0cUeo4s9Rhd7jC6xQ0TmKiWdxCVf5aL4j7j9o7No/H8z0Q9eCLve7P+x9IHfuux5BxLTxOh6+2eybb+0MyHXJedm2kqo2SaG4NDx0iAcIH0CpGdJHp5rk3ML6/met8VwnHmxPM86QwpZlL0gBl5iGowP35zm5eWJgac7JfwUerRRGKyYzzw3lt5K02V/NwaeDeFYK4PSwGtsbAziYHsP3hvbqklLimfptNitDBSULgMYo4s9Rhd7jC72GF1ih0jNlTr7R+gr/sKa8TfwVst0Wg6XoT+4LyJjscNvXfa+I4bcWbfBnrfEuNu/Toyv0TOc+1kNuQ++J94719y1rHyHgXe4twcvZbgUM9n6tDy3DLy4OGkwvedtqVo6qQDi+2x7HDSNjY3Om95735VHU0jDfOa5k5hC3RhTNdWOcKyVQWngWU2HA6LxCCSkQkpG96bOLs1bZdWsmDmWlMT4EIwwMgSlywDG6GKP0cUeo4s9RpfYIWJzFZ+IOu1zrLj5DxSf/v944tQSOg58AJ3tkRmPG37pcmyvtEKYthLOuEkKorx1l3i4cs7qacSNmydGHzjDMy2yFksrhI5TPStogpwjI0tey1wkHj+L2ZdBe3PY8+/AoUu3gbdKKmNaf88gxnzm9cZoYk84dBmUBl5QWD3wXD6cNx86Tm1TKxfMi93wTIPBYDAYooG4OMXdV8yndWIBiV2nOLJ9Q6SH5D97HL3tpq2UnnFn/zcc3ix5arln9dxXKZh6jvw+ZXnP17LPcP7u7sEDZx7erEt6bs8pcrYgmLw0sL/BH6yopsbD0o4hBltFGQwDiUFp4OXk5AR+sE0PvHd31JAQp1gxc0wfB8UGQekygDG62GN0scfoYo/RJXaIhrmKi1NcfaW0UVjz1gtRUXTFL132vgMjcmHkFHmed52zD12uTc+rs26Hy+7pbcSNXyBVNkHy8dyxvHoz3Qy8+EQJ00wcAhMX+j7uAMjJyYGhYwGHUWdXQXMQEg3XUbRhNLEnHLoMSgNv5Mgg2hg0HumVf7d+dy2nTxpOekpikCOLLEHpMoAxuthjdLHH6GKP0SV2iJa5GjNhEifSchlbV8xLW45Eeji+69LRJrl201w6SsXFw6fuk3y8Ebm9jxk9DRbd0Ht7YgpMWCC/23nw5l0NBbc4C7O4cv7/wFfeknOEkZEjR4pBafXoMwYeED3XUTRhNLEnHLoMSgNv8+bNgR2odS8P3vGTbZQerqdoWmx77yAIXQY4Rhd7jC72GF3sMbrEDtE0V+kzl1OQsIu7X9pKfUtkc/F81qX8A2g/KU3MXclaJH3g/A1fnFQonrihY3u/NmU5XPRL+3OmjoDx8/x7rwDo1sW6+W0MPCC6rqNowWhiTzh0GZQGXsC0NsqHtosH7729tWgNRdNjt3qmwWAwGAzRSFzOUtJ0M2Oa9/CbN3ba73SiHNb9Drq6+ndwFidrxWtnsedtCat0z7ULlOU/gC+/Ll7AaMa6+T3GGHgGQ6QZlAbeiBEjAjuwu0WC04P33p5ahqUkkJeV0cdBsUPAugxwjC72GF3sMbrYY3SJHaJqrhwVIL85pZpHNh7kqeLy3vt88Bd45+fiOQsjtrp0tsO9i+H+IjiyRbbteVealycPC80bp2TAhOjtqdaty5iZknOYNiqyA4oSouo6ihKMJvaEQ5fwNUaJYgJuKNjd5Fw8eFpr1u2upXDKKBLiY99WNk057TG62GN0scfoYo/RJXaIqrnKyILhkzkvbS9F0y7hB8+IEXVtfrZzn92OZujbngtdS4AjJbD2N9DRCp1toDvJO+u23vsd3QktdRLh8/eVsPRWqN4KK+8KzThigO71cvZ/Q9F3IzuYKCKqrqMowWhij2l0HiLWrFkT2IFuHryDx5qpON7CWQMkPDNgXQY4Rhd7jC72GF3sMbrEDlE3V5OXEn/ofR744iKKpo3mB89scXry6vbDsd0QnwxlL4YuTPOtu6TVQVM1tDVBxSZqXv917/2OlMjjjS/DtPNg7a/k+bSVvfcdoHSvl8QUGGKKaFhE3XUUBRhN7AmHLoPSwAu43HK3B08aeK7bUwvA0mkDw8CLhjLU0YjRxR6jiz1GF3uMLrFD1M3V5DOhpY6UE3t44Pr8biPv5S2HJd8NYNnt0FQVmjDNqlLYt0rOefMa+K+3Yfp5DGvY1XvfIyWQNBSyzoDrHoXL/gj5X4Zx84MfR4wQdeslSjC69MZoYk84dBmUBp4KtAFnYxUkDeuOq1+/+yiZw1PJHZ0WwtFFjoB1GeAYXewxuthjdLHH6BI7RN1c5TgadR9cT0piPA9cn09e1nB++sI22ne8LnlfS74OCSkSphksG+6DxDTI/5JzW1Y+qadqoOloz32PlDh61cVJJctFN8Klv5fng4SoWy9RgtGlN0YTe8Khy+D5BHJh+fLlgR3o0gOvo7OL9/ceo2ja6AGzYAPWZYBjdLHH6GKP0cUeo0vsEHVzNSJXUiMOvg9ASmI8d185j5bmJti/FqafD8lDYdq5wYdpNhyBrU/B6V+QNgMWmYvksbLYua2rE6q2RHUBlP4g6tZLlGB06Y3RxJ5w6DIoDbySkpLADnQx8LZU1tN4qmNAtUcIWJcBjtHFHqOLPUYXe4wusUPUzZVSEqZ58H3pRwvMnZjBHbNqSdRtVIwukv3mXhl8mOaHfwXdCUu+1nP7hNPQKh4qNzm3HdsD7c2D3sCLuvUSJRhdemM0sSccugxKA+/48eOBHdh4pLvAyvrdtaj/v707j4vquhs//jkM+yqLgGwBFDWgEgmKRKImzZ406dOkaZa2SZo2SVvb/LrlSZ+uaZsuWfp0SZpmaZa2adZu6dPsCxoSRBGDigoqokBkEVBAFmE4vz9mJOAcUJRhZrjf9+uVV5jhMh4+9+L1MHdR0+f8OziFLtOcdDGTLmbSxUy6+A6vXFeZqxz74Kq/Dz91TfQOegnkex9EOc5hmXvRqR2m2d8N5Y/D6R+HmIzRnwsMpTssDRpGvIN39AIrFp/geeX24gWkiytpYuaOLpac4J0UrR3n4DnfwSvZeYCcpEhiwgI9PDAhhBBimsu9DlKWwL+/Dgf3ARC85y3aZhZQvLuL16qaTv0wzU1/gb5DcNbXjJ/uipgLH1Z89Nr7Kx0Tyri5J/tdCSGEW1hygpeXlzfxL+rtcNwLJ2IWfQN2NtV3sHz29Hn3Dk6yiwVIFzPpYiZdzKSL7/DKdWXzh08+CnoI/n6L4/5zHXuYlX858xMj+MG/qnjg7Z3sSbzg5A7TtA/AugchdRmk5BsXiV54nmMC2L7b8cT+SkhY4BibhXnl9uIFpIsraWLmji6WnOC1t7dP/ItG3OR8076DDNg1BZnT634vJ9XFAqSLmXQxky5m0sV3eO26ismAS++HfaXw/OcAsM29gHuvyiUuPIj7Xq/h0lfD6dFB9P35GvRLX4Pd74B98PivveUFxzuDRf9vzEXaQ5yHbTaUO97F219p+cMzwYu3Fw+TLq6kiZk7ulhygldXVzfxLxqe4M1i/Z52x9WQT5teE7yT6mIB0sVMuphJFzPp4ju8el3lfhoWfgpadzgOjYxOZ2FKFC/ffjYbv3ce9153Fr9NuodX+3M48sEL8OdPwANnQm8HWms+qD/IoZ6B0a85ZIe190HiQsd5fGPYedDPcc+7xo1wsA76O2WCh5dvLx4kXVxJEzN3dLHkBO+kdDU5/h+RwIa6duYnRhIVEuDZMQkhhBBWc+n9MPN0WHT1qKdjw4O4dNEs7vjiDWxeej+Len7P08nfg4469r58P1f9oZRPPPge//PPLaNfb+vfHYddrrjDccXOsSgbJC123CpBLrAihPBiljxwPDMzc+Jf1LYL/PwZCE9m494ars5PmfyBedhJdbEA6WImXcyki5l08R1ev66Co+DLpWNOxvz8FN+/7HSiQgL47puBJIUUcObmP9IZtJSzZsfz6tYmPjzYS9KMEMehlmvvhfhsmH/ZuH9sZmYmDJzpuBF6/XrwC4D4093xHfoUr99ePES6uJImZu7oYsl38CIiIib+Ra3VEDuHquZeegfsLM2InfyBedhJdbEA6WImXcyki5l08R0+sa7Ge6cNUEpx+3lZ3HV5Di+EXUuk6uGVwu3cc9UitNb8ed1ex4Lb/wUHquHsb4Lf+P8kioiIcFyAZWgANj8HCdngHzRZ35HP8ontxQOkiytpYuaOLpac4J3UDQVbtsPM+azf0wbAkozoSR6V58kNKM2ki5l0MZMuZtLFd0yndXXDWen8/ls3wdyL8S/7PSmhdi7ITuSZ9fvo7R9wnHsXm+W4SfpxVFZWQvKZjgc9bXJ4ptN02l4mk3RxJU3M5EbnnjLQCx11zgleBxlxYcRHBHt6VEIIIYQ4ESvvgL6DsP4RblyeTm/PYXa/8F1o3gorvgV+thN7ncgkiEhyfCwTPCGEl7LkBC82doKHVx7YCWiGZs5nQ107S9Kn37t3cBJdLEK6mEkXM+liJl18x7RcV8l5kHUBvP8ABQdfZm3ot1mw62H0/MtgwVUn9BLDXZKd96yadYabButbpuX2MgmkiytpYuaOLpac4OXk5EzsC1p3ALDXL5VDvQPT8vw7OIkuFiFdzKSLmXQxky6jKaUeV0q1KKW2jvF5pZT6rVJql1Jqs1Jqyu4QPG3X1Yo7oLcd9dJqAsPjuPbIdynN/80J36h8uMvscyEk2nFhFjF9t5dTJF1cSRMzd3Rx6wRPKXWRUqrauYO6c5zlrlRKaaVUvjvHc9TatWsn9gWtO8DPn/c7ogBYmj697n931IS7WIR0MZMuZtLFTLq4BiX/jQAAIABJREFUeBIY+6ZrcDGQ5fzvFuChKRgTMI3XVeoSuOgX8MlHCVn9LtUhi3ni/boT/vLhLmfeBF/fBoGh7hmnj5m228spki6upImZO7q4bYKnlLIBD+LYSWUD1yqlXH7dpZSKAG4Hytw1llPWsgNiZrNuXzeJkcGkxoR4ekRCCCF8mNZ6LdA+ziJXAH/SDuuAGUqpWVMzumls2Zdg0dUEBwbwmYI03tjWzO+Ld6G1PvHX8POTyZ0Qwqu58x28pcAurXWt1voI8CyOHdaxfgL8Euhz41hG8fef4O3/WnegZ85j/Z42lmTEoI5zeWZfNeEuFiFdzKSLmXQxky4TlgzUj3jc4HzO7ayyrr58zhwuz03inler+faLmzkyODTu8lbpMlHSxUy6uJImZu7o4s7Spp1TwcgFnOcUpGqt/6OU+rYbxzJKUVHRiS880Acde+icfQXNnf0szZieh2fCBLtYiHQxky5m0sVMuriPUuoWHIdxkpSURHFxMeC4eW5ERMTwJbhjY2PJyckZPhzI39+foqIiKioq6OzsBCA/P59Zs2YNv0ZWVhZBQUFs3eo4VTA+Pp65c+dSUlICQFBQEIWFhZSXl9Pd3Q1AQUEBDQ0NNDY2AjBv3jxsNhvbtm0DIDExkYyMDEpLSwEICQmhoKCAsrIyent7ASgsLGTPnj00NTUBkJ2djd1up7q6GoDk5GRSUlIoK3Mc/BMeHk5+fj6lpaX09/cDjm2upqaGlpYWABYsWEB/fz87d+4EIDU1lbsvm4Pf4VZe3NjAtn0tPPHFs9m5uZzBwUEAVqxYQVVVFW1tjlskdXR00NXVRW1tLQDp6enExMRQUVEBQHR0NLm5uaxZswatNUopVq5cSWVlJR0dHQDk5eXR3t5OXV3dKa2n5uZm6uvrPb6e4uLihrcXd62nhIQEysvLAYiMjCQvL4+SkpIx11Nubq5XrKfu7m6vWU9T8fN0Iutpy5YtXreePP3zNDg4SHFx8YTX03jUhA5LmACl1FXARVrrLzgffxYo0Fqvdj72A94GbtRa1ymlioFvaa3LDa81cud15tNPPw2c/ErcuHHj8OEYx1uJ0f2N5JZ+mX+n3sFXd57BT5eHcOV5hdPqh+3oX4o9PT0UFhbKD9sx6+n9998fftfWG9YTeMdfij09PaxYscJr1pO3/Dy988472Gw2r1lP4B0/Tz09PXzsYx+b8HpasmTJRq31lJyfPdWUUunA/2mtFxg+9zBQrLV+xvm4Gliltd4/3mvm5+fro9voyaqoqCAvb8qu6eIV/vVBI99+cTP2Ic3H5sdz7dI0Vsydic3voyN2rNjlREgXM+niSpqYnWwXpdSY+0d3TvAKgR9prS90Pv4OgNb6587HUcBuoNv5JYk4zke43DTJO2oydl7FxcWsWrXqxBbe8iL87WYemP8nHtoWxJYfXYif3/Q8RHNCXSxEuphJFzPpYnayXcbbgfm640zwLgVWA5fgOPrlt1rrpcd7zSnfR04jdQcO88z6fby4sYG2w0dIiwnld9cuJjd1BmDdLscjXcykiytpYuaO/aM7z8HbAGQppTKUUoHANcBLRz+ptT6ktY7TWqdrrdOBdRxncucRrTtA2Sg+EMWC5KhpO7kTQggxdZRSzwClwDylVINS6mal1G1Kqduci7wM1AK7gEeBL3toqJaRHhfGdy45ndLvfIyHrs/DPqT51MOlPL+h/vhfLIQQXsRt5+BprQeVUquB1wAb8LjWukop9WOgXGv90viv4D75+RP4ZXDLdnRMJpubernxrAT3DcoLTKiLhUgXM+liJl3MpMtoWutrj/N5DXxlioYzitXXVaC/HxcvnEVBZixffaaCO/62mcqGg9y+crGnh+aVrL69jEW6uJImZu7o4tbL2WitX8bxW8iRz/1gjGVXuXMsIzU3NxMeHn5iC7dW0xkxmyONQyxMjnLvwDxsQl0sRLqYSRcz6WImXXyHrCuHmLBAnrppKfe+Vs3Da2t5Zv0+MuLCyE6KYllmDNcuSZOjepDtZSzSxZU0MXNHF7fe6NxbHb3QwHEN9kN7LfX+aQAsSpneE7wT7mIx0sVMuphJFzPp4jtkXX3E3+bHdy45neduWcbHMwPInBnOxrp2vvuPrXzp6Y30HBn09BA9TrYXM+niSpqYuaOL3JBiPG27QNvZ2j+LyGB/0mLkxqZCCCGE1RRkxtK7L5BVq/LRWvPEe3X89D/b+NQfSnnshnxmRYV4eohCCDHMku/gZWVlndiCrTsAKOmMY1HKjGl7g/OjTriLxUgXM+liJl3MpIvvkHVldrSLUorPF2XwxxuXsLethyseeI9Xt+5n0D7+jdKnK9lezKSLK2li5o4ulpzgBQUFndiCLTvQyo+3WyOn/eGZMIEuFiNdzKSLmXQxky6+Q9aV2bFdzpkXz9++dBYhgTZu+0sFy3/5Nr96vZoPD/Z6aISeIduLmXRxJU3M3NHFkhO8ozf3Pa7WHfRHnEbPUIAlJngn3MVipIuZdDGTLmbSxXfIujIzdZmXGMFb31jJI589k9NnRfK7d3ax8t53+OWrOyxzfp5sL2bSxZU0MXNHFzkHbzytO2gJTgdgYcoMz45FCCGEEF7H3+bHBTmJXJCTSH17D79+cycPFe/mn5sa+e6lp3PpwlnT/hQPIYR3seQ7ePHx8cdf6FAjHKhhK1nEhgWSFBXs/oF52Al1sSDpYiZdzKSLmXTxHbKuzE6kS2pMKPdfncuLtxUSHRrI6r9u4qYnN9B0qG8KRugZsr2YSRdX0sTMHV0sOcGbO3fu8ReqeRWAf/SewcKUKEv89u2EuliQdDGTLmbSxUy6+A5ZV2YT6ZKfHsO/v1rEDy7Lpqy2nfP/dw0vbmzAcf/66UW2FzPp4kqamLmjiyUneCUlJcdfqPoVhqIzeetAFIsscnjmCXWxIOliJl3MpIuZdPEdsq7MJtrF5ue42uYrt5/N/MQIvvVCJTc/VU5DR4+bRugZsr2YSRdX0sTMHV0sOcE7rv5u2LOGlqRzGNKKRcnT/wIrQgghhJh86XFhPHdLId+/LJvS3W2c/6u1PLJ2NwMWva2CEML9LDnBO+7lSHe/BfYjVAQXArDQAlfQBLl87Viki5l0MZMuZtLFd8i6MjuVLn5+ipuLMnjzmytZPieOn728g4//roTt+zsncYSeIduLmXRxJU3M3NFF+drx4Pn5+bq8vNy9f8g/boPqV/hm+t8oqT1I2f+c594/TwghhJFSaqPWOt/T4/AVU7KPFKfs9aomvv+vrfQcsfPkTUs587RoTw9JCOFjxts/WvIdvHF3fvZBqHkN5l7IpsZuFiZb4/w7OE4XC5MuZtLFTLqYSRffIevKbDK7XJCTyN+/vJzYsEA++8cy3tt1YNJee6rJ9mImXVxJEzN3dLHkBK+7u3vsTzash952DqefT23rYRanWWeCN24XC5MuZtLFTLqYSRffIevKbLK7JM8I4fnbCkmLCeWmJzbwn837ffIqm7K9mEkXV9LEzB1d5Ebnx6p+GfwCKA/IA3bIYRNCCCGEcIv4iGCevWUZNzyxga/8tYKkqGDnTdMTmJ8YyYyQAPz8pv9tmoQQk8uS5+D19vYSEhJi/uTvzoQZadwb/3MeXlPLlh9dSEig7ZT+PF8xbhcLky5m0sVMupidbBc5B29i3L6PtDB3duk9Yuc/W/bzWlUTa2ta6R90XGHT308RExZIbuoMfv3pMwgL8r7fy8v2YiZdXEkTM3fsHy15iGZDQ4P5Ewd2QtsumHcJ5XUd5CRFWmZyB+N0sTjpYiZdzKSLmXTxHbKuzNzZJSTQxlVnpvDo5/LZ9IPzeexz+fzw49ncujKTs7Nm8tb2Zm5/9gPsQ973S3nZXsykiytpYuaOLpac4DU2Npo/setNAAZmn09lw0HyLHZ45phdLE66mEkXM+liJl18h6wrs6nqEhroz3nZCdy0PINvXzif+6/O5Ycfz+HN7c384pXtUzKGiZDtxUy6uJImZu7o4n3v9XtSfRlEprC9dwZ9A0Pknxbj6REJIYQQwuJuOCud3a3dPPruHjJnhnPt0jRPD0kI4cUsOcGbN2+e+RP1GyB1KeV1HQDknWadK2jCOF0sTrqYSRcz6WImXXyHrCszT3f5wWXZ7G3r4fv/3Mq/Kz+kb8BOzxE7kSEBXJ2fymWLZhEcMPWnlXi6i7eSLq6kiZk7ulhygmezGf4CPNQInQ2QupqNtR0kzwhhVpS1TgQ1dhHSZQzSxUy6mEkX3yHryszTXfxtfjxw3WLu/NsWmjv7CAvyJzY8iNrWbr71QiU//c82rsxLITU6hCP2IfoHhogND+KTeclunfh5uou3ki6upImZO7pYcoK3bds24uPjRz/ZsB4AnbKEjW93sDTDeodnGrsI6TIG6WImXcyki++QdWXmDV0iggN48Pq8Uc9prSmtbeMv6/by1Pt1DB5zIZZfv1nD6nPn8OklqQT5T/4/JL2hizeSLq6kiZk7ulhygmdUvwH8g/kwJIumzhK5/50QQgghvJ5SirNmx3HW7Di6+gYYsGuC/P0I9Pdj494OfvV6DT/4VxV/KN7NL65cxIq5Mz09ZCGEm1nyKpqJiYmuTzash6TFbGw4DGDJCZ6xi5AuY5AuZtLFTLr4DllXZr7QJSI4gJiwQMKC/Amw+bEsM5bnbl3Gn29eSniwPzc9uYHnNuwb9TWNB3v52cvb2dxw8KT+TF/o4gnSxZU0MXNHF0tO8DIyMkY/MdAHH34AqUvZWNdOaKCN+YkRnhmcB7l0EYB0GYt0MZMuZtLFd8i6MvPVLkopzs6ayd++dBbL58Tx33/bwq9er6a7f5D7Xqvm3PuKeWRtLZ9+eB3F1S0Tfn1f7eJu0sWVNDFzRxdLTvBKS0tHP7G/EoYGIGUpG/d1cEbqDPxt1kvj0kUA0mUs0sVMuphJF98h68rM17tEBAfwxxvyuTo/hd++vYuld7/JA+/s4qIFifzrK8vJiAvjC0+V8/eKid102de7uIt0cSVNzNzRRc7Bg+ELrBxOyGP7/g/4yqrZHh6QEEIIIcTkCrD58csrF5EeF8a62na+fl4Wi9Mcp6Q8d+sybv3zRr7xfCWNHb3csjLTLRdlEUK4n/XepgJCQo65/UF9GUSnU9keiH1Ik2fB8+/A0EUA0mUs0sVMuphJF98h68psunRRSvHlVXP40+eXDk/uwPEO3xM3LeHjuUnc/0YNK+8p5qn36+gbsI/7etOly2STLq6kiZk7uiit9fGX8iL5+fm6vLx88l5Qa7h/PmSs4A9xd/KLV3ZQ+YMLiAoNmLw/QwghxElRSm3UWud7ehy+YtL3kcJytNa8t6uN37xVw4a6DuLCg0ieEUzvgJ2+gSEiQ/y5dGESl5+RRPIMxz9Mu/oGqPqwk/AgfxYkR3n4OxDCGsbbP7r1HTyl1EVKqWql1C6l1J2Gz9+mlNqilPpAKVWilMp253iOKisr++jBoXroboLUpVQ3dTErKtiyk7tRXcQw6WImXcyki5l08R2yrsys0kUpRVFWHM/fWsgzX1zGkvRoosMCyYwL58zToh2Heb66g+W/eJtPPPgeZ/30VRbd9TrXPLKOjz9QwqNra/G1Nw/cwSrby0RIEzN3dHHbOXhKKRvwIHA+0ABsUEq9pLXeNmKxv2qt/+Bc/nLgV8BF7hrTUb29vR89qHecf0fqUnaUdjLPglfPPGpUFzFMuphJFzPpYiZdfIesKzOrdVFKUTg7lsLZsS6f29fWw0uVjby5vYWEULimcC4LU6J4obyeu1/ezp62w9x1eQ4BFrxg3VFW215OhDQxc0cXd15kZSmwS2tdC6CUeha4Ahie4GmtO0csHwZM/a986tdDQBgDcaezu+UtVsyNm/IhCCGEEEL4irTYUFafm8Xqc7MoLi5m1aosAFZmzeTe2GoeKt5NfXsPD16fR2SwNY+KEsKT3PmrlWSgfsTjBudzoyilvqKU2g3cA3zNjeMZVlhY6PhAa9j7PiTnUdfezxH7kCXvf3fUcBcxinQxky5m0sVMuvgOWVdm0sVsZBc/P8V/XzSfe65cROnuNq544D2qm7pGLd90qI//faOGP6/bS9WHhxi0D031kKeEbC+upImZO7p4/DYJWusHgQeVUtcB3wNuOHYZpdQtwC0ASUlJFBcXA5CZmUlERASVlZUAxMbGkpOTw9q1awHw9/enqKiIiooKOjsdbxbm5+ezZcsW+vv7mfXh68xr3kL3qp/w97cdx78GHG5lcDCRkpISAIKCgigsLKS8vJzu7m4ACgoKaGhooLGxEYB58+Zhs9nYts3x5mRiYiIZGRnD97UICQmhoKCAsrKy4bdhCwsL2bNnD01NTQBkZ2djt9uprq4GIDk5mZSUlOHjcsPDw8nPz6e0tJT+/n4AioqKqKmpoaXFcWPSBQsW0N/fz86dOwFITU0lISGBoyfcR0ZGkpeXR0lJCYODgwCsWLGCqqoq2tra6Ovro6CggK6uLmprawFIT08nJiaGiooKAKKjo8nNzWXNmjVorVFKsXLlSiorK+no6AAgLy+P9vZ26urqTmk9NTc3U1/v+B1BVlYWQUFBbN26FYD4+Hjmzp07Jetpw4YNw728YT0B5Obmenw99fX1UVRU5DXryVt+nsrKyhgaGvKa9QTe8fPU19fHOeecM+H1JKbenj17mD9/vqeH4XWki5mpy9VLUsmYGcaXn67gEw++xz1XLeLc+fE8vLaWR9bupm/go0ldSICNc+fHc9cVOcSFB416ne7+QXqP2JkZMfp5XyDbiytpYuaOLm67iqZSqhD4kdb6Qufj7wBorX8+xvJ+QIfWetzLL03GFcKKi4tZtSgNHiqClHz47D+5742dPLRmN9t+fKFl7/viOMxilaeH4XWki5l0MZMuZifbRa6iOTGTto+UbdiFdDEbr0tLZx9ffrqC8r0dRIUEcKh3gMsWzeKOC+ejFFTs62Dj3g6e3VDPjJAAfnPNYgpnxzJoH+LZDfX86o0aeo4M8p2LT+ezy07Dz09N7Td3CmR7cSVNzNyxf3TnO3gbgCylVAbQCFwDXHfMwLK01judDy8FdjIVtB3+cRv4+cMnfg9+flQ3d5ERF2bZyZ0QQgghxGSKjwzmr19cxi9f3UF1UxffuGAueSPuvZcaE8oVZyRzzZI0Vv+1gusfW8fnCtN5f/cBapq7WZoRQ3CAjR++VMUb25q556pF9A7YeWXLfl7e0kRUSABPfn6J/NtNiGO4bYKntR5USq0GXgNswONa6yql1I+Bcq31S8BqpdR5wADQgeHwTHfI73vPcXPzTz4KUSkAVDd1sTDF2vduyc6ekrtU+BzpYiZdzKSLmXTxHbKuzKSL2fG6BPr78f3Lxl8mOymSf3+1iO//aytPvl9HWkwof/hMHhfmJALw1/X7uPs/21l57zsM2B1Hni1IjqS0to1fvLKDH348Z3K+mUkk24sraWLmji5uPQdPa/0y8PIxz/1gxMe3u/PPN9pfSdiG30L2J2DhpwA43D/IvvYePnVmypQPx5vY7XZPD8ErSRcz6WImXcyki++QdWUmXcwmq0tYkD+/uvoMbi7KYE58+Kh35a4vOI2z58zk0XdrmT0zjIsWzCIxKpi7/l3FE+/VsSwzdngyeDw1zV0c7h9k8Yh3Et1BthdX0sTMHV2sd4OSQ430BCfAZf8LynEsd02z4wpPVr4HHjB8oQMxmnQxky5m0sVMuvgOWVdm0sVssrvkJEUZD7lMiw3lJ59YwI3LM0iMCgbgzovnszA5im+/UElDR8+4r9s/aOe+16q55Dfv8smH3ufBd3YxNOS+u3PJ9uJKmpi5o4vHr6I55eZfwob9QawKjRl+6uglfOcnRnpqVEIIIYQQYgKC/G08cN1iLv1tCV99ZhO3rsjkUO8Ah3oH8FOKhMhgZkUF0zcwxF3/rmJnSzdX5qUwYB/i3teqqaw/yP1X5xIh9+oT04z1JnhAckraqMc7mroIDbSREh3ioRF5h+Rkl9sUCqTLWKSLmXQxky6+Q9aVmXQx83SX02LD+PknF/LVZzZx218qxlwuKSqYJ25awjnz4tFac0bqDO5+eTtXPOC4jUN+esyYX3syPN3FG0kTM3d0seQELyVl9Ll21U1dZCVE+NTld93h2C7CQbqYSRcz6WImXXyHrCsz6WLmDV0+nptETlIkfQNDRIb4ExUSwJCG5s4+9h/q41DvAOfMmzn8Tp1Sis8XZZCTFMntz37AVX8o5dKFs7jz4vmkxoROypi8oYu3kSZm7uhivXPwYPgmukfVNHcxP8Ha59+BaxfhIF3MpIuZdDGTLr5D1pWZdDHzli6ZM8PJTookJTqUiOAAokICmJsQwcq5M7k8N8l4GGZBZixvf2slt38si7d2NPOx+9dwx4uVPL+hnuqmLgbsQ2zf38mfS+v42jOb+MrTFfx53V72HDjM8e4j7S1dvIk0MXNHF0u+gzdSa1c/bYePWP4CK0IIIYQQVhMa6M/Xz5/LtUvTuO/1al7Z2sTz5Q2A41p8R+dxiZHBKAX/2bJ/+HFqTAhx4UHMjAgiJTqEnKQocpIimREayBG7purDQ+xq6SY2LIjlc2JRytpHiompY8kJXnh4+PDHH11gRSZ4I7uIj0gXM+liJl3MpIvvkHVlJl3MpkuXxKhg7vtULvdcuYi6tsNUNhxkZ3M3c+LDWZIeM3ydhrq2Hkp2HWBjXTtNnX3sbOnm/d1tHOodGH6tuPBA2rqPoN8oGX5ufmIEt67M5LJFSQTY/OgbsNPc2Uegv59z8jj9J3/TZVuZbO7ooo73FrO3yc/P1+Xl5ZP2eo+9W8tP/7Odjd87j9jwoEl7XSGEEKdOKbVRa53v6XH4isneRwohTkz74SNUfXiIrY2d7DnQzayoELISwsmKj2Br4yEeXrubmuZuYsICGdKagz0fTQijQwM4fVYkC5KjuDw3iQXJUR78ToSvGG//aMlz8EpLS4c/rm7qIi48SCZ3jO4iPiJdzKSLmXQxky6+Q9aVmXQxky4OMWGBnJ01ky+tms09V+WyLLyNyxYlMS8xgivPTOHV21fw+I35rMiK47JFs/jWBXO596pF/PiKHC7MSeRw/yBPvlfHZb8r4YoH3+P58np6jgx6+tuaVLKtmLmjiyUP0ezv7x/+uLq5Sw7PdBrZRXxEuphJFzPpYiZdfIesKzPpYiZdzI7t4uenOHd+AufOTxjzaw71DvD3igaeLtvHHS9u5rv/2MKC5CiWZsSw5LQYTk+KJCnKdw/nlG3FzB1dLDnBO6rnyCA79ndxU1G6p4cihBBCCCEsLCokgJuWZ3DjWems39NOcU0r6/e083jJHh5eUwtAWKCNOfHhzI4PJzMujMyZ4aRGh2LzUwxpjX1I429ThAX6ExpoIyjAxtCQZmBoiEG7JiYskOAAm4e/U+FuljwHb3BwEH9/f9bUtHLD4+t56vNLWTl35iSN0Hcd7SJGky5m0sVMupidbBc5B29iJnMfKUaTLmbSxWwyu/QesbOl8RA7W7rY2dzNzpYualsPs/9Q34RfKyLIn0sXzeLKM1PIPy16St8NlG3FzB37R0tWrqmpITs7m/d3HSDApliSHu3pIXmFo13EaNLFTLqYSRcz6TKaUuoi4DeADXhMa/2LYz5/I3Av0Oh86gGt9WNTMTZZV2bSxUy6mE1ml5BAG0szYliaETPq+Z4jg+w5cJj69l5A46cUNj/FgF3Tc2SQniN2+gbs2PwU/jY//P0UG+raeanyQ57dUE96bCj3XJXr8rrjsQ9p6tt7OC02dMKTQ9lWzNzRxZITvJaWFrKzs3lv9wHy0qIJDbRkBhdHu4jRpIuZdDGTLmbS5SNKKRvwIHA+0ABsUEq9pLXedsyiz2mtV0/1+GRdmUkXM+liNhVdQgP9nffeO/Grbl67NI2fXDHIa1VN/O7tXVz76Dq+c/F8bi7KGHfC1jdg54WNDTz2bi1723rIjAvj2qVpXHlmCjFhgS7Lr61p5WcvbydzZhj3f+oMQgJtsq2MwR1dLDuzOdhzhKoPO/n6eXM9PRQhhBDWshTYpbWuBVBKPQtcARw7wRNCiEkXFuTPJ/NSOD87gW+/sJmf/mc7Ffs6WH1OFp19A3QcPkJ7zxE6eweHH7++rZn2w0fITZ3BtUvTeGNbM3e/vJ17X6tm+ZxYCmfHUpgZR1iQjZ+9vIM3tzeTFBXMq1ub+PDgOv54gxxpP5UsOcFbsGABpbvb0BqWz4n19HC8xoIFCzw9BK8kXcyki5l0MZMuoyQD9SMeNwAFhuWuVEqtAGqAr2ut6w3LTDpZV2bSxUy6mPlCl4jgAB76TB6PvlvLL1+t5uUtTS7L+PspokICyEubwRfPzmRpRgxKKW5bOZua5i6eXV9PcXUL71S3Dn9NWKCNOy+ez03L01lT3cpXn9nElQ+9z6//K8s4Dq013f2DRAQHuHyu4/AR1u5sJTUmlPmJEdPyqDt3bCvTr9IJ6O/v573d7YQF2liUMsPTw/EacvlaM+liJl3MpIuZdJmwfwPPaK37lVK3Ak8B55oWVErdAtwCkJSURHFxMQCZmZlERERQWVkJQGxsLDk5OaxduxYAf39/ioqKqKiooLOzE4D8/Hz27dvH1q1bAcjKyiIoKGj4cXx8PHPnzqWkpASAoKAgCgsLKS8vp7u7G4CCggIaGhpobHScPjhv3jxsNhvbtjneoExMTCQjI2P43k8hISEUFBRQVlZGb28vAIWFhezZs4emJsc/OLOzs7Hb7VRXVwOQnJxMSkoKZWVlAISHh5Ofn09paenwtlZUVERNTQ0tLS2A4x9R/f397Ny5E4DU1FQSEhI4elGayMhI8vLyKCkpYXDQcf+xFStWUFVVRVtbGwMDA+Tn59PV1UVtreOKhunp6cTExFBRUQFAdHQ0ubm5rFmzBq01SilWrlxJZWUlHR0dAOTl5dHe3k5dXd0prafm5mYbkk+RAAAQmElEQVTq6+u9Yj0d/XO9YT0B5Obmenw9DQwMUFhY6FXraayfp3OSk1nyxXxeX7eZiEBFfFQoRUsWs2NzBXqwH6UURUWLqampYc2aLcPrKWywnxURLazIh5CY2ew5HMD6bbtZkexPWsQhgvxthHbs4ttnBvLrih5u/MtWbjijntyoPvyUIjc3l10ftnH3q7uobLVz2enRfPOCudRVO/rs6A7msQ+6OdDt2FYUkB4XRmrYEGmhdrKi/bh8RT6HDnb49M9TZ2cnAQEBE/57bzyWvIpmcXExd5VDRlwYj9+4ZJJG5vuKi4tZtWqVp4fhdaSLmXQxky5mJ9tlOl5FUylVCPxIa32h8/F3ALTWPx9jeRvQrrU+7ok2k7WPlG3YlXQxky5m0mW02tZuvvDYu9QeGmJhchTfu/R0qpu7+OUrO7BrzcfmJ/BaVRMhgTa+eu4cNjcc4v827ycnKZLvXno6XX2OW5tt23+ID+oP0tzpmPRFBvtz6aIkrsxL5swpviroZHHH/tGS7+C19Q6x50Av1xekeXooQgghrGcDkKWUysBxlcxrgOtGLqCUmqW13u98eDmwfWqHKIQQkydzZjjfWxZM54y5/OKVHXz6kXUAnJ0Vx92fWEhabCi7Wrq469/b+NnLOwiwKb51wVxuXTmbAJsfABfmJAKOQzobOnop39vOmupW/rmpkWfW7+O0WMdhnAAKRUigjayEcLJnRZI9K5KZEUE+OQE8GZac4DURBfRSlBXn6aF4ldTUVE8PwStJFzPpYiZdzKTLR7TWg0qp1cBrOG6T8LjWukop9WOgXGv9EvA1pdTlwCDQDtw4VeOTdWUmXcyki5l0cXVaWhqzZydzQU4Cfyrdy6yoYC7PTRqedM2Jj+BPn19K6e424iODmRMfbnwdpRSpMaGkxoTyX4tT6O4f5LWtTfzzg0bqDvQAoNF09w3yj02Nw18XaPMjOiyA6NBAokICiAj2JyzIn/AgfwoyY7lkQSL+zsnksYaGNA+8s4vXqpp44Lo8MuLCJq2LO7YVSx6iufovG1hXd5AN3z3PMjP5E9Hd3U14uPmHycqki5l0MZMuZifbZToeoulOk7GPlG3YTLqYSRcz6eLKE00O9QywvamT7fs7aers4+DhAdp7jnCw5wjd/XYO9w9ysOcInX2DJM8I4eaiDD69JJWwoI/eAzvUO8A3nvuAt3a0EOjvR1xYIM/dWkhqTOikjNEd+0fLvYOntaZkZytnz58lk7tjlJeXy/HiBtLFTLqYSRcz6eI7ZF2ZSRcz6WImXVx5oklUaADLMmNZljn2VfOHhjRv7WjhkbW7+fH/beO+16vJS4smPz2arPgI7n1tBw0dvfz4ihzyT4vh2kfXcd1j63j+1kJmRYXQ1t3PCxsbqNjbQXJ0CBlxYZwWG8ah3gF2tXSzu6WbQ70D5CRHsjh1BovTokmIDB7+893RxXITvN2t3Rzs1yyfLbdHEEIIIYQQwsr8/BTnZydwfnYCFfs6+OemRsrrOvjNWzvRGmZGBPHsLcvIT48B4E+fX8r1j5Vx/aNlLEiO4tWtTRyxD5EeG8q7Ow/QO2D/6LUVpMWEEh7sz+MlexiwO46cvGVFJv9zyelu+54sN8F7b5fjErrL58j5d8eKjIz09BC8knQxky5m0sVMuvgOWVdm0sVMuphJF1e+0CQvLZq8tGgAOvsGqGrsZF5iBDFhgcPL5KbO4MmblvC5x9fT2t3PdQVpXF+QRlZCBFprmjv72dt2mMiQADLiwggOsAHQN2Bn2/5ONu07SPasj1q4o4vlzsHrG7CzpfEQS5yzcCGEEN5LzsGbmMk4B08IIcTxtXb1Ex7kT0igzSN//nj7R/OlYqax4AAb/Q3bPD0Mr3T0Bo5iNOliJl3MpIuZdPEdsq7MpIuZdDGTLq6mW5OZEUGTMrlzRxfLTfAABgcHPT0EryRdzKSLmXQxky5m0sV3yLoyky5m0sVMuriSJmbu6GLJCZ4QQgghhBBCTEeWOwcPYGhoCD8/mdseS7qYSRcz6WImXcxOtoucgzcxso90H+liJl3MpIsraWLmjv2jWysrpS5SSlUrpXYppe40fP4bSqltSqnNSqm3lFKnuXM8R1VVVU3FH+NzpIuZdDGTLmbSxUy6+A5ZV2bSxUy6mEkXV9LEzB1d3DbBU0rZgAeBi4Fs4FqlVPYxi20C8rXWi4AXgXvcNZ6R2trapuKP8TnSxUy6mEkXM+liJl18h6wrM+liJl3MpIsraWLmji7ufAdvKbBLa12rtT4CPAtcMXIBrfU7Wuse58N1QIobxyOEEEIIIYQQ05o7b3SeDNSPeNwAFIyz/M3AK6ZPKKVuAW4BSEpKori4GIDMzEwiIiKorKwEIDY2lpycHNauXQuAv78/RUVFVFRU0NnZCUB+fj5xcXHDr5GVlUVQUBBbt24FID4+nrlz5w5fsjQoKIjCwkLKy8vp7u4GoKCggIaGBhobGwGYN28eNpuNbdsct19ITEwkIyOD0tJSAEJCQigoKKCsrIze3l4ACgsL2bNnD01NTQBkZ2djt9uprq52xEtOJiUlhbKyMgDCw8PJz8+ntLSU/v5+AIqKiqipqaGlpQWABQsW0N/fz86dOwFITU0lISGBo+djREZGkpeXR0lJyfAVe1asWEFVVRVtbW3Y7XY6Ojro6uqitrYWgPT0dGJiYqioqAAgOjqa3Nxc1qxZg9YapRQrV66ksrKSjo4OAPLy8mhvb6euru6U1lNzczP19fUeX08zZswY3l68YT0B5Obmenw92e12uru7vWY9ecvPU3h4+PD24g3rCbzj58lut9Pf3z/h9SSmXm5urqeH4JWki5l0MZMurqSJmTu6uO0iK0qpq4CLtNZfcD7+LFCgtV5tWPYzwGpgpda6f7zXnYwTyPft20daWtopvcZ0JF3MpIuZdDGTLmYn20UusjIxso90H+liJl3MpIsraWLmjv2jOw/RbARSRzxOcT43ilLqPOC7wOXHm9xNlqO/TRejSRcz6WImXcyki5l08R2yrsyki5l0MZMurqSJmTu6uHOCtwHIUkplKKUCgWuAl0YuoJRaDDyMY3LX4saxCCGEEEIIIcS059b74CmlLgF+DdiAx7XWdyulfgyUa61fUkq9CSwE9ju/ZJ/W+vLjvGYrsPcUhxYHHDjF15iOpIuZdDGTLmbSxexku5ymtZ452YOZrmQf6VbSxUy6mEkXV9LEbNL3jz53o/PJoJQql3M6XEkXM+liJl3MpIuZdPEdsq7MpIuZdDGTLq6kiZk7usjt5IUQQgghhBBimpAJnhBCCCGEEEJME1ad4D3i6QF4KeliJl3MpIuZdDGTLr5D1pWZdDGTLmbSxZU0MZv0LpY8B08IIYQQQgghpiOrvoMnhBBCCCGEENOO5SZ4SqmLlFLVSqldSqk7PT0eT1FKpSql3lFKbVNKVSmlbnc+H6OUekMptdP5/2hPj3WqKaVsSqlNSqn/cz7OUEqVObeZ55z3dbQUpdQMpdSLSqkdSqntSqlC2VZAKfV158/PVqXUM0qpYCtuL0qpx5VSLUqprSOeM24fyuG3zj6blVJ5nhu5GEn2jw6yfxyf7CNdyT7STPaRDp7YR1pqgqeUsgEPAhcD2cC1Sqlsz47KYwaBb2qts4FlwFecLe4E3tJaZwFvOR9bze3A9hGPfwn8r9Z6DtAB3OyRUXnWb4BXtdbzgVwcfSy9rSilkoGvAfla6wU47vd5DdbcXp4ELjrmubG2j4uBLOd/twAPTdEYxThk/ziK7B/HJ/tIV7KPPIbsI0d5kineR1pqggcsBXZprWu11keAZ4ErPDwmj9Ba79daVzg/7sLxl1Eyjh5PORd7CviEZ0boGUqpFOBS4DHnYwWcC7zoXMSKTaKAFcAfAbTWR7TWB7H4tuLkD4QopfyBUGA/FtxetNZrgfZjnh5r+7gC+JN2WAfMUErNmpqRinHI/tFJ9o9jk32kK9lHjkv2kXhmH2m1CV4yUD/icYPzOUtTSqUDi4EyIEFrvd/5qSYgwUPD8pRfA3cAQ87HscBBrfWg87EVt5kMoBV4wnlYzmNKqTAsvq1orRuB+4B9OHZah4CNyPZy1Fjbh/w97J1kvRjI/tGF7CNdyT7SQPaRx+XWfaTVJnjiGEqpcOBvwP/TWneO/Jx2XGLVMpdZVUpdBrRorTd6eixexh/IAx7SWi8GDnPMoSZW21YAnMfLX4Fj554EhOF6CIbAmtuH8H2yfxxN9pFjkn2kgewjT5w7tg+rTfAagdQRj1Ocz1mSUioAx87raa31351PNx99K9j5/xZPjc8DlgOXK6XqcByedC6O4+pnOA8vAGtuMw1Ag9a6zPn4RRw7MytvKwDnAXu01q1a6wHg7zi2IatvL0eNtX3I38PeSdbLCLJ/NJJ9pJnsI81kHzk+t+4jrTbB2wBkOa/gE4jjZM+XPDwmj3AeN/9HYLvW+lcjPvUScIPz4xuAf0312DxFa/0drXWK1jodx7bxttb6euAd4CrnYpZqAqC1bgLqlVLznE99DNiGhbcVp33AMqVUqPPn6WgXS28vI4y1fbwEfM55pbBlwKERh6kIz5H9o5PsH81kH2km+8gxyT5yfG7dR1ruRudKqUtwHENuAx7XWt/t4SF5hFKqCHgX2MJHx9L/D47zDJ4H0oC9wNVa62NPDJ32lFKrgG9prS9TSmXi+G1lDLAJ+IzWut+T45tqSqkzcJxUHwjUAjfh+AWRpbcVpdRdwKdxXHVvE/AFHMfKW2p7UUo9A6wC4oBm4IfAPzFsH84d/QM4DtXpAW7SWpd7YtxiNNk/Osj+8fhkHzma7CPNZB/p4Il9pOUmeEIIIYQQQggxXVntEE0hhBBCCCGEmLZkgieEEEIIIYQQ04RM8IQQQgghhBBimpAJnhBCCCGEEEJMEzLBE0IIIYQQQohpQiZ4QkwhpZRdKfXBiP/unMTXTldKbZ2s1xNCCCGmkuwjhZgc/sdfRAgxiXq11md4ehBCCCGEF5J9pBCTQN7BE8ILKKXqlFL3KKW2KKXWK6XmOJ9PV0q9rZTarJR6SymV5nw+QSn1D6VUpfO/s5wvZVNKPaqUqlJKva6UCvHYNyWEEEJMAtlHCjExMsETYmqFHHP4yadHfO6Q1noh8ADwa+dzvwOe0lovAp4Gfut8/rfAGq11LpAHVDmfzwIe1FrnAAeBK938/QghhBCTRfaRQkwCpbX29BiEsAylVLfWOtzwfB1wrta6VikVADRprWOVUgeAWVrrAefz+7XWcUqpViBFa90/4jXSgTe01lnOx/8NBGitf+r+70wIIYQ4NbKPFGJyyDt4QngPPcbHE9E/4mM7cp6tEEKI6UH2kUKcIJngCeE9Pj3i/6XOj98HrnF+fD3wrvPjt4AvASilbEqpqKkapBBCCOEBso8U4gTJby6EmFohSqkPRjx+VWt99DLQ0UqpzTh+w3it87mvAk8opb4NtAI3OZ+/HXhEKXUzjt9CfgnY7/bRCyGEEO4j+0ghJoGcgyeEF3CeX5CvtT7g6bEIIYQQ3kT2kUJMjByiKYQQQgghhBDThLyDJ4QQQgghhBDThLyDJ4QQQgghhBDThEzwhBBCCCGEEGKakAmeEEIIIYQQQkwTMsETQgghhBBCiGlCJnhCCCGEEEIIMU3IBE8IIYQQQgghpon/D/Vi6k6oJLjzAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 71.8104 %\n",
            "\tMin validation loss: 1.02665\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4_jscKyYXfZu"
      },
      "source": [
        "### fold-7 <a name=\"fold-7\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "I31v6arnXcgd",
        "outputId": "fbde8fb3-46fb-4ab0-ffa4-4e979537e6d9"
      },
      "source": [
        "FOLD_K = 7\n",
        "REPEAT = 1\n",
        "\n",
        "history7 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history7.append(history)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 13.1265%\n",
            "\n",
            "Epoch 1/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 2.0731 - accuracy: 0.2221 - val_loss: 1.9316 - val_accuracy: 0.3174\n",
            "Epoch 2/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.8943 - accuracy: 0.2834 - val_loss: 1.9635 - val_accuracy: 0.2375\n",
            "Epoch 3/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.7626 - accuracy: 0.3366 - val_loss: 1.7368 - val_accuracy: 0.3699\n",
            "Epoch 4/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.6426 - accuracy: 0.3982 - val_loss: 1.6141 - val_accuracy: 0.4570\n",
            "Epoch 5/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.5272 - accuracy: 0.4467 - val_loss: 1.5468 - val_accuracy: 0.4558\n",
            "Epoch 6/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.4640 - accuracy: 0.4777 - val_loss: 1.5297 - val_accuracy: 0.4726\n",
            "Epoch 7/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.4066 - accuracy: 0.4911 - val_loss: 1.5769 - val_accuracy: 0.4547\n",
            "Epoch 8/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.3428 - accuracy: 0.5133 - val_loss: 1.3526 - val_accuracy: 0.5251\n",
            "Epoch 9/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.3017 - accuracy: 0.5345 - val_loss: 1.3062 - val_accuracy: 0.5370\n",
            "Epoch 10/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.2614 - accuracy: 0.5475 - val_loss: 1.3200 - val_accuracy: 0.4905\n",
            "Epoch 11/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.2476 - accuracy: 0.5560 - val_loss: 1.2109 - val_accuracy: 0.5561\n",
            "Epoch 12/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.2184 - accuracy: 0.5683 - val_loss: 1.2723 - val_accuracy: 0.5597\n",
            "Epoch 13/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 1.1769 - accuracy: 0.5820 - val_loss: 1.2086 - val_accuracy: 0.5513\n",
            "Epoch 14/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.1413 - accuracy: 0.6020 - val_loss: 1.1594 - val_accuracy: 0.5537\n",
            "Epoch 15/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.1181 - accuracy: 0.6056 - val_loss: 1.1101 - val_accuracy: 0.5883\n",
            "Epoch 16/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.1078 - accuracy: 0.6145 - val_loss: 1.1398 - val_accuracy: 0.5847\n",
            "Epoch 17/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.0745 - accuracy: 0.6292 - val_loss: 1.1698 - val_accuracy: 0.6086\n",
            "Epoch 18/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.0742 - accuracy: 0.6309 - val_loss: 1.1240 - val_accuracy: 0.5740\n",
            "Epoch 19/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.0339 - accuracy: 0.6490 - val_loss: 1.1168 - val_accuracy: 0.5967\n",
            "Epoch 20/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.0172 - accuracy: 0.6523 - val_loss: 1.0402 - val_accuracy: 0.6575\n",
            "Epoch 21/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.9844 - accuracy: 0.6619 - val_loss: 1.1217 - val_accuracy: 0.5955\n",
            "Epoch 22/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.9763 - accuracy: 0.6720 - val_loss: 1.2743 - val_accuracy: 0.5346\n",
            "Epoch 23/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.9658 - accuracy: 0.6758 - val_loss: 1.1360 - val_accuracy: 0.6086\n",
            "Epoch 24/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.9481 - accuracy: 0.6870 - val_loss: 1.0012 - val_accuracy: 0.6408\n",
            "Epoch 25/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.9414 - accuracy: 0.6904 - val_loss: 1.0055 - val_accuracy: 0.6277\n",
            "Epoch 26/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.9044 - accuracy: 0.7042 - val_loss: 1.0110 - val_accuracy: 0.6372\n",
            "Epoch 27/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.9014 - accuracy: 0.7046 - val_loss: 1.0543 - val_accuracy: 0.6456\n",
            "Epoch 28/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8986 - accuracy: 0.7071 - val_loss: 1.0659 - val_accuracy: 0.6396\n",
            "Epoch 29/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8883 - accuracy: 0.7105 - val_loss: 1.0898 - val_accuracy: 0.6420\n",
            "Epoch 30/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8700 - accuracy: 0.7171 - val_loss: 1.0136 - val_accuracy: 0.6504\n",
            "Epoch 31/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.8523 - accuracy: 0.7279 - val_loss: 1.0360 - val_accuracy: 0.6623\n",
            "Epoch 32/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.8520 - accuracy: 0.7276 - val_loss: 1.0094 - val_accuracy: 0.6635\n",
            "Epoch 33/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.8382 - accuracy: 0.7349 - val_loss: 1.0278 - val_accuracy: 0.6539\n",
            "Epoch 34/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.8303 - accuracy: 0.7354 - val_loss: 0.9618 - val_accuracy: 0.6730\n",
            "Epoch 35/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.8355 - accuracy: 0.7342 - val_loss: 0.9774 - val_accuracy: 0.6718\n",
            "Epoch 36/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8014 - accuracy: 0.7522 - val_loss: 1.0472 - val_accuracy: 0.6635\n",
            "Epoch 37/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.7807 - accuracy: 0.7540 - val_loss: 1.0566 - val_accuracy: 0.6539\n",
            "Epoch 38/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.7826 - accuracy: 0.7548 - val_loss: 1.0871 - val_accuracy: 0.6539\n",
            "Epoch 39/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.7715 - accuracy: 0.7567 - val_loss: 1.0739 - val_accuracy: 0.6516\n",
            "Epoch 40/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.7718 - accuracy: 0.7622 - val_loss: 0.9709 - val_accuracy: 0.6659\n",
            "Epoch 41/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.7526 - accuracy: 0.7674 - val_loss: 1.0316 - val_accuracy: 0.6527\n",
            "Epoch 42/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.7320 - accuracy: 0.7739 - val_loss: 1.0711 - val_accuracy: 0.6492\n",
            "Epoch 43/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.7464 - accuracy: 0.7708 - val_loss: 1.0801 - val_accuracy: 0.6492\n",
            "Epoch 44/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.7350 - accuracy: 0.7770 - val_loss: 1.1256 - val_accuracy: 0.6444\n",
            "Epoch 45/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.7325 - accuracy: 0.7739 - val_loss: 1.0240 - val_accuracy: 0.6635\n",
            "Epoch 46/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6974 - accuracy: 0.7887 - val_loss: 1.1275 - val_accuracy: 0.6730\n",
            "Epoch 47/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.7065 - accuracy: 0.7869 - val_loss: 1.2845 - val_accuracy: 0.6074\n",
            "Epoch 48/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6861 - accuracy: 0.7938 - val_loss: 1.0671 - val_accuracy: 0.6683\n",
            "Epoch 49/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6815 - accuracy: 0.7959 - val_loss: 1.0237 - val_accuracy: 0.6695\n",
            "Epoch 50/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.6958 - accuracy: 0.7882 - val_loss: 1.0171 - val_accuracy: 0.6802\n",
            "Epoch 51/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6833 - accuracy: 0.7930 - val_loss: 1.0834 - val_accuracy: 0.6563\n",
            "Epoch 52/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.6667 - accuracy: 0.7986 - val_loss: 1.1483 - val_accuracy: 0.6587\n",
            "Epoch 53/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.6604 - accuracy: 0.8006 - val_loss: 1.0932 - val_accuracy: 0.6766\n",
            "Epoch 54/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6523 - accuracy: 0.8066 - val_loss: 1.1680 - val_accuracy: 0.6611\n",
            "Epoch 55/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6559 - accuracy: 0.8043 - val_loss: 1.0738 - val_accuracy: 0.6587\n",
            "Epoch 56/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6400 - accuracy: 0.8072 - val_loss: 1.1587 - val_accuracy: 0.6599\n",
            "Epoch 57/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6515 - accuracy: 0.8086 - val_loss: 1.2689 - val_accuracy: 0.6826\n",
            "Epoch 58/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6397 - accuracy: 0.8066 - val_loss: 1.1499 - val_accuracy: 0.6635\n",
            "Epoch 59/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6352 - accuracy: 0.8101 - val_loss: 1.1025 - val_accuracy: 0.6647\n",
            "Epoch 60/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6189 - accuracy: 0.8152 - val_loss: 1.0127 - val_accuracy: 0.6862\n",
            "Epoch 61/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6218 - accuracy: 0.8154 - val_loss: 1.1404 - val_accuracy: 0.6683\n",
            "Epoch 62/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6188 - accuracy: 0.8192 - val_loss: 1.1327 - val_accuracy: 0.6599\n",
            "Epoch 63/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6022 - accuracy: 0.8275 - val_loss: 1.0968 - val_accuracy: 0.6897\n",
            "Epoch 64/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.5992 - accuracy: 0.8249 - val_loss: 1.0744 - val_accuracy: 0.6647\n",
            "Epoch 65/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.6094 - accuracy: 0.8207 - val_loss: 1.1832 - val_accuracy: 0.6802\n",
            "Epoch 66/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6014 - accuracy: 0.8238 - val_loss: 1.2312 - val_accuracy: 0.6838\n",
            "Epoch 67/100\n",
            "247/246 [==============================] - 10s 43ms/step - loss: 0.5995 - accuracy: 0.8233 - val_loss: 1.0883 - val_accuracy: 0.6850\n",
            "Epoch 68/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5854 - accuracy: 0.8313 - val_loss: 1.0453 - val_accuracy: 0.6683\n",
            "Epoch 69/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5913 - accuracy: 0.8304 - val_loss: 1.1401 - val_accuracy: 0.6623\n",
            "Epoch 70/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5804 - accuracy: 0.8309 - val_loss: 1.1817 - val_accuracy: 0.7005\n",
            "Epoch 71/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5736 - accuracy: 0.8313 - val_loss: 1.1994 - val_accuracy: 0.6790\n",
            "Epoch 72/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5610 - accuracy: 0.8365 - val_loss: 1.2305 - val_accuracy: 0.6635\n",
            "Epoch 73/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5754 - accuracy: 0.8309 - val_loss: 1.1318 - val_accuracy: 0.6909\n",
            "Epoch 74/100\n",
            "247/246 [==============================] - 11s 44ms/step - loss: 0.5655 - accuracy: 0.8342 - val_loss: 1.2074 - val_accuracy: 0.6611\n",
            "Epoch 75/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5678 - accuracy: 0.8347 - val_loss: 1.0784 - val_accuracy: 0.6563\n",
            "Epoch 76/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5499 - accuracy: 0.8418 - val_loss: 1.1680 - val_accuracy: 0.6790\n",
            "Epoch 77/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5710 - accuracy: 0.8314 - val_loss: 1.1702 - val_accuracy: 0.6718\n",
            "Epoch 78/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5434 - accuracy: 0.8425 - val_loss: 1.1582 - val_accuracy: 0.6766\n",
            "Epoch 79/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5436 - accuracy: 0.8430 - val_loss: 1.1463 - val_accuracy: 0.6766\n",
            "Epoch 80/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5370 - accuracy: 0.8433 - val_loss: 1.2777 - val_accuracy: 0.6659\n",
            "Epoch 81/100\n",
            "247/246 [==============================] - 11s 43ms/step - loss: 0.5449 - accuracy: 0.8427 - val_loss: 1.1828 - val_accuracy: 0.6611\n",
            "Epoch 82/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.5331 - accuracy: 0.8462 - val_loss: 1.2299 - val_accuracy: 0.6504\n",
            "Epoch 83/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.5398 - accuracy: 0.8434 - val_loss: 1.2484 - val_accuracy: 0.6468\n",
            "Epoch 84/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5306 - accuracy: 0.8467 - val_loss: 1.1667 - val_accuracy: 0.6790\n",
            "Epoch 85/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5106 - accuracy: 0.8529 - val_loss: 1.5871 - val_accuracy: 0.6432\n",
            "Epoch 86/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5303 - accuracy: 0.8513 - val_loss: 1.1762 - val_accuracy: 0.6814\n",
            "Epoch 87/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5057 - accuracy: 0.8547 - val_loss: 1.3877 - val_accuracy: 0.6539\n",
            "Epoch 88/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5122 - accuracy: 0.8525 - val_loss: 1.4947 - val_accuracy: 0.6432\n",
            "Epoch 89/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5402 - accuracy: 0.8442 - val_loss: 1.1903 - val_accuracy: 0.6671\n",
            "Epoch 90/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.4963 - accuracy: 0.8599 - val_loss: 1.3205 - val_accuracy: 0.6695\n",
            "Epoch 91/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.5091 - accuracy: 0.8591 - val_loss: 1.1741 - val_accuracy: 0.6647\n",
            "Epoch 92/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.5014 - accuracy: 0.8588 - val_loss: 1.1535 - val_accuracy: 0.6790\n",
            "Epoch 93/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.5155 - accuracy: 0.8519 - val_loss: 1.2438 - val_accuracy: 0.6468\n",
            "Epoch 94/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.5086 - accuracy: 0.8541 - val_loss: 1.2097 - val_accuracy: 0.6671\n",
            "Epoch 95/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.4837 - accuracy: 0.8599 - val_loss: 1.3535 - val_accuracy: 0.6372\n",
            "Epoch 96/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.4868 - accuracy: 0.8574 - val_loss: 1.2816 - val_accuracy: 0.6754\n",
            "Epoch 97/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.5021 - accuracy: 0.8590 - val_loss: 1.1612 - val_accuracy: 0.6826\n",
            "Epoch 98/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.4865 - accuracy: 0.8623 - val_loss: 1.3559 - val_accuracy: 0.6253\n",
            "Epoch 99/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.4873 - accuracy: 0.8632 - val_loss: 1.1859 - val_accuracy: 0.6897\n",
            "Epoch 100/100\n",
            "247/246 [==============================] - 10s 42ms/step - loss: 0.4823 - accuracy: 0.8670 - val_loss: 1.4868 - val_accuracy: 0.6480\n",
            "Training completed in time:  0:17:01.923874 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "id": "xTZnQ4vVXlrQ",
        "outputId": "5107b719-9970-48f3-a549-1aec396993f8"
      },
      "source": [
        "show_results(history7)"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3gAAAE9CAYAAABZZMC4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOydeVxc1fn/33dmYCBAEiAhJEACxJCEEMlCRJQs7hq1LnX3q9VWrdbqV6v91qW12vbX1Vq1rbZWrRobNdXGfdcsUkkMIWIAk2AgCxAgLAEmrDNzfn8cIMBcCEkYZoDn/Xrxgpl75tyHz3nunPvcc85zDKUUgiAIgiAIgiAIwvDH4msDBEEQBEEQBEEQhMFBAjxBEARBEARBEIQRggR4giAIgiAIgiAIIwQJ8ARBEARBEARBEEYIEuAJgiAIgiAIgiCMECTAEwRBEARBEARBGCHYfG3AkTJhwgQVHx9/THW0tbURGBg4OAaNIEQXc0QXc0QXc0QXc45Wl82bN1crpSZ6waQRifSR3kN0MUd0MUd08UQ0Mccb/eOwC/Di4+PJyck5pjqam5sJDg4eJItGDqKLOaKLOaKLOaKLOUeri2EYu71gzohF+kjvIbqYI7qYI7p4IpqY443+cVRO0SwtLfW1CX6J6GKO6GKO6GKO6GKO6DJ8kLYyR3QxR3QxR3TxRDQxxxu6jMoAr6yszNcm+CWiizmiizmiizmiizmiy/BB2soc0cUc0cUc0cUT0cQcb+gyKgM8QRAEQRAEQRCEkciwW4NnRnt7O6WlpbS0tAyo/IQJE/j666+9bNXwo1OXoKAgYmNjCQgI8LVJfsHMmTN9bYJfIrqYI7qYI7r4DukjBwfpI82Ra9sc0cUT0cQcb+gyIgK80tJSwsLCiI+PxzCMw5Zvb2+XL2YT2tvbsdls1NTUUFpaSkJCgq9N8gusVquvTfBLRBdzRBdzRBffIX3k4CB9pDlybZsjungimpjjDV1GxBTNlpYWIiMjB9RxdZYXPGlpacEwDCIjI0WjbhQWFvraBL9EdDFHdDFHdPEd0kcODtJHmiPXtjmiiyeiiTne0GVEBHjAgDsu4fCIloIgCCML+V4fPERLQRD8nRET4B0JNtvgzkw9cOAATzzxxBF/bvny5Rw4cGBQbTkWBluXkUJ0dLSvTfBLRBdzRBdzRJfhg/SR5kgfaY5c2+aILp6IJuZ4Q5dRGeDZ7fZBra+vzsvpdPb7uXfffZfx48cPqi3HwmDrMlKQdRbmiC7miC7miC7DB+kjzZE+0hy5ts0RXTwRTczxhi6jMsA7ePDgoNZ3zz33sHPnTubNm8eiRYtYvHgx3/rWt0hOTgbgwgsvZOHChcyZM4ennnqq63Px8fFUV1eza9cuZs+ezY033sicOXM488wzaW5uHlQbB8Jg6zJSyM7O9rUJfonoYo7oAt9UOVi5cU+P90SX4UF9UxvVBxyDWqf0kSMbubbNEV08EU3M8YYuozLAG2x++9vfMn36dL788kv+8Ic/kJuby2OPPcaOHTsAePbZZ9m8eTM5OTk8/vjj1NTUeNRRVFTErbfeSkFBAePHj+e1114b6n9DEAThqHG63Lyfv4+r/rGB0x9Zxy/eLuBAU5uvzRKOkMrGVhra1KDWKX2kIAjC0DLiJpQ/9FYBheUN/ZZxuVxHlJI0ecpYfn7+nAGXP+GEE3oMtz7++OOsXr0agL1791JUVERkZGSPzyQkJDBv3jwAFi5cyK5duwZ8vsFCFo6bExwc7GsT/BLRxZzRoEtjSzvv5VewobiGqoZWqhpb2HeghcZWJzHjg/nxWTO5fFEc48cEdn1mNOgyHDhcH9nS7sKtFGMCB357IH3k6EaubXNEF09EE3O8ocuIC/AGgrf34QgJCen6e+3atXz88cdkZ2czZswYli1bZppeufvcfqvV6pPpJ6GhoUN+zuFAenq6r03wS0QXc3ylS35ZPU+tL2ZssI3IEDuRoYGE2m2MCbQSHGgjZnwwx0X1f41XO1r5+7qdBAfaWJo0gdTY8disFpRSlNY1s7Wsnne27uPjwkpanW4mhtmJGR9MfGQIi+IjWJo0kdNmT8Jq8bwRFn8ZHlgMA7ca3BG83kgfObKQa9sc0cUT0cQcb+gy4gK8gTxFdDgcg/pFHRYWRmNjo+mx+vp6wsPDGTNmDNu2bWPDhg2Ddt7BZrB1GSls3LhRvpRMEF3M8YUuLrfix69+xa7qg9gDLBxoajctd+sp07nz9CRs1p6z85VSvJZbxq/eKcTR4sStFI9/UkRYkI3EiaHsrHLgaNUJMcLHBHD5ojgumh/DvLjxAx7VEH/xDw7XR+5vbGVffTPJU8ZiswzOKg7pI0c2cm2bI7p4IpqY4w1dRlyANxDUID+djIyM5OSTTyYlJYXg4GAmTZrUdezss8/mb3/7G7Nnz2bmzJmceOKJg3ruwWSwdRkp+OJJ8XBAdDFnsHRxuxVf7KrlrbxyggOspMVHsCg+nMhQz0x+r2zay9f7GvjzlfM5P3UKTpeb2qY2Dra6aGpz0tLuYtWmUv66ZiebdtXx5yvnM2lsEA0t7WzeXcczn5WQ9U01adPC+e235zIxNIisb6pZv2M/u2sPcvGCGGZFj2XW5DBSpowj0HbkN/7iL8ODQKsO2NudbmyBgxPgSR85spFr2xzRxRPRxBxv6DIqAzxvsHLlStP37XY77733numxzjUEEyZMID8/v+v9u+++e9DtEwTBP8kqqub+17fS1OZiVnQYyZPHEmC18EZeGXtrmwkJtNLuVjydVQLA3Jhx/OWq+UyL1NPc6pvaefjD7ZyQEMF5x08GwGa1EBUWBGGHzrNwWgQnTo/g/tX5nPPYZ0waG8S2igaUglC7jV9emMLVJ0zF0jG98tzjJ3NuR33C6CGgI3hvcykGc1WI9JGCIAhDx6gM8LrP/xcOIbqYk5GR4WsT/BLRxZyB6tLU5uQ3725jxYbdJE4MYcmMiWyraOCf/91Fu9vNydMncNcZMzlrTjQWC2wtreeLXbU8tb6Ybz+ZzfPfXcScKeN49JMd1DW18fPzkw87XfKi+bHMjRnPA2/kYzEM7jgtiUXx4cybOv6IkmocDeIvw4PAjum7bU63jy3xP6SPNEeubXNEF09EE3O8ocuoDPBaW1slk48Joos5JSUlzJo1y9dm+B2iizmduhRVNpJXWk96QgRxEWO6jndmoPzrmm/YU9vE9zIT+PFZMwkK0MmfnC43B9tcjAsO6FFvWnwEafERnJk8iWue+YIr/r6Be5fP5oXs3VyxaCpzpowbkH3HRYWy8sahnwYn/jI8sFoMLIZBu0sCvN5IH2mOXNvmiC6eiCbmeEOXURngOZ1OX5vgl4gu5lRUVMgXkgmjSZeWdhcNLe00tjhxtDhpbnfR5nTT5nTjVopxwQGMGxNAqN3Gqpy9fPlpDbl7DnR9flZ0GKfOimJPbRMfdWSgnD4xhJduPJETE3umg7dZLYwL7nvt03FRYbx2y0lc88xG7lu9lbAgG3efmeS1/32wGE3+cjgMw4gDXgAmAQp4Sin1WK8yBvAYsBxoAq5TSuUOgW3YDBnBM0P6SHPk2jZHdPFENDHHG7qMygBPEAShO59/U82ruaXcv3y2RxKTlRv38MAb+TjdA0+wcFxUIPcvn03G9Eiyd9bw0deV/G3dTsYF6wyUF86PYf4RZKDszZTxwfz75pO47z9bOWdutGniFcGvcQJ3KaVyDcMIAzYbhvGRUqqwW5lzgBkdP+nAkx2/vY7NAm0ygicIgjBsGZUBXlBQkK9N8EtEF3OSk5N9bYJfMhJ0cbsVT67byR8/3I5bQfH+g7x044kEB+rpkp/vrOaBN/JZFB/B8uMnE2a3EWq3ERxoxW6zEGizYDEMGprbOdDczoGmdqLs7ZyWmtgVvKXEjOPGJYk4Wp3YbRYCrIOTmTAiJJC/XbNwUOoaCkaCvwwWSql9wL6OvxsNw/gaiAG6B3gXAC8onbpxg2EY4w3DmNzxWa8SFGijvtmJUko29+6G9JHmyLVtjujiiWhijjd0GZUBnqQ6Nkd0McflcvnaBL/E33RRSvHptirWbK/iqhOmkTxlbL/lG1rauWtVHh8VVnLe8ZM5bXYUP1qVx/++vIUn/2ch5QeaufVfuUyLHMPfr13I2KCAfuvrZN++faY3xaH2Ufl124W/+Yu/YBhGPDAf2NjrUAywt9vr0o73vB7g2SwGLqVwuRU2qwR4nUgfaY5c2+aILp6IJuZ4Q5dRecfR2tpKYGCgz84fGhqKw+GgvLyc22+/nVdffdWjzLJly3j44YdJS0vrs55HH32Um266iTFjdAKH5cuXs3LlSsaPH39UdvlaF39l+/btTJ4s6eJ740+67Khs5JdvF/JZUTWGAS99sZfrTornzjOSCAm0UrivgdW5ZXxWVE19czuOVieOVic2i8HPz0/mupPiMQyDA03tPPRWIT97I5/c3XW43Iqnv7NowMEd+Jcu/oTo4olhGKHAa8AdSqmGY6jnJuAmgClTprB27VoAEhMTCQsLo6WlhcbGRqxWK8HBwTgcjq7PhoWFcfDgQdxuPSVzzJgxKFc7AAcaHYwdE4RhGLS0tABgs9kICgrqqsMwDEJDQ3vUERISQltbG+3tuh673e5Rh91u5+DBgz3qcDgcKKWYPHkyDQ0NlJSUcNddd7FixQqCgoJQStHa2grAueeey8MPP8zs2bMBsFgshISEdNUB8PTTT3Pttdd29WuXXXYZzz//fFeilICAAAICAmhqaupRR/dN2UNDQ2lubsblcuFyubBYLLhcLlpaWli7di3x8fFERESQm6uXR4aHh5Oamsq6deu6RkCXLl1KXl4edXV1ACxYsIDa2tqubSA62ykvLw/Q+wbOmTOH9evXd+mVmZlJbm4uDQ3aTdLS0qisrGTvXv0MYMaMGdjt9q7tJKKiokhKSiIrK6urDTIyMsjJyelqu/T0dEpLSykrKwNg5syZWK1WCgv1QHJ0dDQJCQlkZ2cDEBwcTHp6Ohs3buzatysjI4O8vDy2b98O6JEIl8vV9TomJobY2Fg2btzYpWdaWhrZ2dldbZmZmcmOHTuoqqoCICUlhdbWVoqKigCIi4tj0qRJ5OTkADB27FgWLFhAVlZW17rIJUuWUFBQQE1NDQCpqak0NjZSXFwM4JN2cjgcLFu2zG/aqaSkhIqKCp+2k1KK6upqv2on8P31tH//fkJDQ4+4nfrDGG5PpNLS0lSn83Ty9ddfd33JD4TGxkbCwsIOX9BLdHZk/TGQAC8+Pp6cnBwmTJgwKHZ11+VINR3JrF27lmXLlvnaDL/Dm7oUlNfz5NqdjAm0EhUWRNRYO1aL0ZXkxNHqpLnNRVO7i4bmdj4r2k+o3cYdpydxfuoU/vTxDl76Yg9RYXbGBQewo9JBgNUgY/oEosfaCbUHEBpk47RZUaTG9Xwg8qu3C3k6qwSLAc9dfwJLkiYeke3iL+YcrS6GYWxWSvX9RThMMQwjAHgb+EAp9YjJ8b8Da5VSL3W83g4sO9wUzcHoI6sPNFDucDMtcgzjgof+oZ/0kcML+c4zR3TxRDQxxxv946gcwQsIGPjT+IFwzz33EBcXx6233grAgw8+iM1mY82aNdTV1dHe3s6vfvUrLrjggh6f27VrF+eddx75+fk0Nzdz/fXXk5eXx6xZs3rsan/LLbewadMmmpubueSSS3jooYd4/PHHKS8v55RTTmHChAmsWbOmR2f2yCOP8OyzzwJwww03cMcdd7Br1y7OOeccMjMz+fzzz4mJieGNN97o8TRT8CQmJsbXJvglfelyoKmNxhZnj60BjoS126u49V+5WC0GQQFWqh2tdM9vYjEgxG5jTKCVMYE2ggOsfOekeG4/dQbhIfpm9NcXzeWShbH89t1tKBS/ujCFc+dO7jreH/ctn02AzULSpNAjDu5A/KUvRJdDdGTIfAb42iy46+BN4IeGYbyMTq5SPxTr7wCC7QHgaKXNOTgPgKWPHNnItW2O6OKJaGKOV3RRSg2rn4ULF6reFBYWerzXHy6X64jKH47c3Fy1ZMmSrtezZ89We/bsUfX19Uoppfbv36+mT5+u3G63UkqpkJAQpZRSJSUlas6cOUoppf74xz+q66+/XimlVF5enrJarWrTpk1KKaVqamqUUko5nU61dOlSlZeXp5RSatq0aWr//v1d5+18nZOTo1JSUpTD4VCNjY0qOTlZ5ebmqpKSEmW1WtWWLVuUUkpdeumlasWKFaa6HKmmI5mmpiZfm+CXmOmSs6tGHf/gB2raT95WS37/qfrp6q3q48IK1e4c2DX38he7VeK976izH12v9h1oVkop5XS5VWVDs6qsb1YHW9u7riN/RfzFnKPVBchRftD3DOYPkIneHuEr4MuOn+XAzcDNHWUM4K/ATmArkDaQugejj3Q6nSq/9IAqqxscX5Y+cmQj33nmiC6eiCbmeKN/HHkjeO/dAxVb+y3idjmxWI/gX4+eC+f8ts/D8+fPp6qqivLycvbv3094eDjR0dHceeedrF+/HovFQllZGZWVlURHR5vWsX79em6//XYAjj/+eI4//viuY6tWreKpp57C6XSyb98+CgsLexzvTVZWFhdddBEhISEAXHzxxXz22Wd861vfIiEhgXnz5gGwcOHCrjnLAAcPHvTp1FV/ZePGjTKlwITeunxUWMkPV+YyZXwwt516HJ/vrOHVzaWs2LCb2PBgbshM4LJFcQTZrGzeU8c7X+1jQ3ENdpuF0CAbFsPgs6JqliRN5ImrF3QlJbFaDKLChk/2OvEXc0SXQyilstABXH9lFHDroJ98AH2kcjlJcIMFA6wKLIfpL6WPHNXItW2O6OKJaGKON3QZeQGej7j00kt59dVXqaio4PLLL+df//oX+/fvZ/PmzQQEBBAfH9+1wPxIKCkp4eGHH2bTpk2Eh4dz3XXXHVU9ndjth/bLslqtPaa5CMLR8sqmPdz7n63MjRnHs9ctIjLUzg2LE2l1uli7fT//WF/Mg28V8ugnRdhtFiobWrHbLKQnRmIx6Fpbd/3J8XqK5CBtJSAIwtFhwcCm2sDphIAQMI7tmpQ+UhAEYegYeQFeP08RO2k9eBBbx5O7weLyyy/nxhtvpLq6mnXr1rFq1SqioqIICAhgzZo17N69u9/PL1myhJUrV3LqqaeSn5/PV199BUBDQwMhISGMGzeOyspK3nvvva4oPywsjMbGRo8F5IsXL+a6667jnnvuQSnF6tWrWbFixWH/B4tFbqrNCA0N9bUJfol9TAhv5ZWzIns3X+yqZUnSRJ68egEh3bYDsNusnDUnmrPmRLN5dy3P/ncXbrfinLmTOXVW1IjcOkD8xRzRxU8YYB/Z2nKQiFadDY6I6RDU/7Yjh0P6yJGLXNvmiC6eiCbmeEOXkXd3NQBCBjm4A5gzZw6NjY3ExMQwefJkrr76as4//3zmzp1LWloas2bN6vfzt9xyC9dffz2zZ89m9uzZLFyoNzBOTU1l/vz5zJo1i7i4OE4++eSuz9x0002cffbZTJkyhTVr1nS9v2DBAq677jpOOOEEQC8gnz9/fo+pJmZ4Q5eRQH9Z2kYLe2qaWJWzl8aWdtpcblra3WR9U8/+xiqmRozh/uWzue7k+H5H3hZOi2DhtIghtNo3iL+YI7oMH0KCAglu2EmbshFoOMHdfsx1Sh85cpFr2xzRxRPRxBxv6DIqt0lwOBzyFMGE7rpICuhDZGdnk5GR4WszfEJ9Uzt//rSI57N3oRSEBtkItFoItFmICnRy2znzWJo0EYtFNkPuZDT7S38crS4jdZsEb3HMfaRSOKt2YHU18417CjMsZRA2GcLM18aNJqSPNEe+88wRXTwRTczxRv84KkfwhltQO1SILuZ0bvI5Eulsc521/RBVDS38e3MpT60vpqGlncsWxvGjM5OYNPZQspO1a9eybFbUkNo7HBjJ/nIsiC7DhIPV2FxNtIXG0NwQiBsLFrfT11b5BdJHmiPXtjmiiyeiiTne0MWrAZ5hGGcDjwFW4Gml1G97HZ8KPA+M7yhzj1LqXW/aJAgjHaUURVUOYsYH91gPp5Ria1k967bvp6jKQXG1g5L9B7FZLcyfOp75ceHERQTz7tYK1myvwuVWLEmayL3nzGL25GNbfyMIwjBAueHgfpzWMVhCJkBDA26LDYvr2KdoCoIgCEOH1wI8wzCs6D18zgBKgU2GYbyplCrsVuynwCql1JOGYSQD7wLx3rKpE5meaY7oYk5mZqavTRgwZQeaufc/W1m/Yz82i8Hc2HGcmBhJS7uLDwsqKTvQjGFAzPhgEieGkjYtgpZ2F7l76li3Yz9KwYRQOzcuTuSytFgSJ/btE8NJl6FEdDFHdBkGGBaYmITVrTCsFqwWAxc2bBLgAdJH9oVc2+aILp6IJuZ4QxdvjuCdAHyjlCoGMAzjZeACoHuAp4DOoYFxQPnRnkwp5THNrC9aWloIDg4+2lONWDp1kWkoPdmxYwfJycm+NqNf3G7Fvzbu5rfvbUMBPz5rJk1tTjYU1/KP9cVYLAZLZkzkzjOSOG1WFOEhgR511De3s7vmILMnjx3QNgXDQRdfILqYI7r4lgH3kRYbLa3NBNsCCLRaaFdW7G6ZVgXSR/aFXNvmiC6eiCbmeEMXbwZ4McDebq9LgfReZR4EPjQM4zYgBDj9aE4UFBRETU0NkZGRA+rAnE5ZT2CG0+lEKUVNTQ1BQcNnY2lvU1VV5ZdfSIXlDWzaVcuWPXXk7K6jtK6ZxTMm8JuL5xIbPqarXHObC4DgQGu/9Y0LDuD42PEDPr+/6uJrRBdzRBffcbR9ZIDVQlu7VWfRVAoG+BB1pCJ9pDlybZsjungimpjjDV18nWTlSuA5pdQfDcPIAFYYhpGilHJ3L2QYxk3ATQBTpkxh7dq1ACQmJhIWFsaePXsICQmhrKyMgICAHosVg4KCaGtrw+3WVQYGBtLa2trVydlsNgzDoL1dT0GxWq3YbLauOgzDwG6309ra2vXUzm6343Q6cbn0jXNAQACAX9ShlOrqnK1WK1arlba2NkDv4RMYGNhjE1i73U57eztutxulFHa7HcMw2L9/P7t37yY+Pp6IiAhyc3MBCA8PJzU1lXXr1nU9EV66dCl5eXnU1dUBOgV1bW1tV8rpznbKy8sDIDIykjlz5rB+/fquNsjMzCQ3N5eGhgZAp4ytrKxk7179jGDGjBnY7Xby8/MBiIqKIikpiaysrK7/IyMjg5ycHBwOBwDp6emUlpZSVqb3cpo5cyZWq5XCQj2IHB0dTUJCAtnZ2QAEBweTnp7Oxo0buza3zcjIoKWlpcvnkpOTcblcbN++HYCYmBhiY2PZuHEjoKfwpKWlkZ2d3dV2mZmZ7Nixg6qqKgBSUlJobW2lqKgIgLi4OCZNmkROTg4H2xWR48dy4qKFZGVldbXlkiVLKCgooKamhppmN2+XB7OmqBaA8XaD46eEcuvimUS37uWbvC+oGYJ2cjgcOBwOv2mnkpISKioqhqSdAMaOHcuCBQs82qm5ubnLX1JTU2lsbKS4uBhgVF9PDoeD1tbWI24n4diJjY2ltLSU/fv3D6h8S0sLQUFBHGhup6K1kXE44EDhMW92Ptzp1CUoKIjY2FhfmyMIgtAnXtsmoSNge1ApdVbH63sBlFK/6VamADhbKbW343UxcKJSqqqves1SQB8p1dXVHhufCqJLX3hTF7db8X5BBZ/vrGZTSR3bKxuZEBrIU9emsWBqeI+y7S43z2SV8NjHRSgUt506gwvmTSFmfPCApycPJuIv5ogu5hytLrJNwpExmH3kP/9bQu47T/PnwL/ADzZCVP971Y105No2R3QxR3TxRDQxxxv9ozdH8DYBMwzDSADKgCuAq3qV2QOcBjxnGMZsIAgY2CPGY0DStJojupjjLV2qHa3c+cqXfFZUTajdxoJp4ZwzN5rVW8q44qkN/PHSVM5PnYLbrXg3fx+PfLiD4uqDnD57Ej8/P5m4iDGHP4kXEX8xR3QxR3QZPnS21bTIMbyvOh40OSpGfYAnPmyO6GKO6OKJaGLOsNomQSnlNAzjh8AH6C0QnlVKFRiG8QsgRyn1JnAX8A/DMO5EJ1y5Tg3B6uWioiJiYmK8fZphh+hizrHq4nYrPiysICLETvKUsYTabWworuH2l7ZwoLmd/3dRCpenxWHrSGxybUY831+Rw20vbWHz7jo27aqloLyBpEmhPPOdNE6bPWmw/rVjQvzFHNHFHNFl+NDZVkmTwqiiY11uY4VvjfIDxIfNEV3MEV08EU3M8YYuXl2D17Gn3bu93nug29+FwMnetEEQfM0rOXu59z9bAZ2jYFrEGPbUNhEfGcJz159A8pSee8xFhATy4g3p3PvaVp77fBex4cE8clkqF8yLwWoZ3UkOBEEYOmLGB+MIiNQvJMATBEEYNvg6yYpPiIuL87UJfonoYs6x6FLtaOW3723jhIQIblk6nfyyevLL61maNJEfnz2LULv5JWi3WfnjZalckzGNOVPGEWjzv+QG4i/miC7miC7Dh862MgyDmElRNFcHEywBnvhwH4gu5ogunogm5nhDl1EZ4E2a5B9T3PwN0cWcY9HlN+9uo6nNya8vSuG4qDBOmRU14M8ahsH8XolW/AnxF3NEF3NEl+FD97aaOSmMyv3hxDskwBMfNkd0MUd08UQ0MccbuvjfsMAQcKwZxkYqoos5R6vLhuIaXsst5cbFiRwXFTbIVvke8RdzRBdzRJfhQ/e2SooOo8I9jvYD5T60yD8QHzZHdDFHdPFENDHHG7qMyhE8QRhMlFK8mVfOnz/9hqRJoXwrdQonHzeBn76eT2x4MLedOsPXJgqCIBwVMyeFUaXG42woI8DXxgiCIAgDYlQGeGPHjj18oVGI6GJOpy7fVDn427qdJE4M4dy5k5kWGUKNo5Wfvp7Pe/kVzIoOY2NxLe9urSDAatDuUjx7XRrBgVYf/wfeQfzFHNHFHNFl+NC9rZImhfKGCsd2cAsopTNFjVLEh80RXcwRXTwRTczxhi5e2+jcWwzGJq6CcCQopVj5xR5++XYhAC3tbgBSYsZSUd9CQ7OTO89I4qYliSilyC6u4a28ciJC7NxzzujeN0oQjv7Bs8AAACAASURBVBXZ6PzIGOw+UinFn35xGz9SK+CePRA0btDqFgRBEI6e/vrHUbkGLysry9cm+CWiiyc1jla+/eiH3L86n0XxEaz78Slk/eQU7l8+mwCrhcSJobx528ncsmw6VouBzWph8YyJ/P6S1BEf3Im/mCO6mCO6DB+6t5VhGFjHTdEvGit9ZJF/ID5sjuhijujiiWhijjd0GZVTNJ1Op69N8EtEl0O0Od28kL2Lxz4pornNyc/OS+b6k+KxdOxDd+OSRG5ckuhbI32M+Is5oos5osvwoXdbhUXGwAFQjeUYE5N8ZJXvER82R3QxR3TxRDQxxxu6jMoATxD6QinFx19X8et3v6ak+iBLkyZyVpSDqzITfG2aIAgjCMMwngXOA6qUUikmx8cBLwJT0X31w0qpfw6tlZrI6KmwEw5UlRI+up9rCYIgDAtG5Ro8t9uNxTIqZ6f2y2jXZU9NEw++VcCn26qYPjGEn56XzCkzo0a9Ln0hupgjuphztLqM1DV4hmEsARzAC30EePcB45RSPzEMYyKwHYhWSrX1V683+sicbbtIezmVnfN+wvQL7zumuoczcm2bI7qYI7p4IpqY443+cVSqXFBQ4GsT/JLRqktLu4u/fFrEGX9ax8biGn567mzev2MJp8zUm5KPVl0Oh+hijuhijujSE6XUeqC2vyJAmGEYBhDaUXZI5jf1bqvpsZNpUnYO1pQOxen9FvFhc0QXc0QXT0QTc7yhy6icollTU+NrE/ySkaxLm9NN+YFmag62UXuwjarGFgrKG/iq9ADb9jXidCvOnTuZn543m8njgnt8diTrciyILuaILuaILkfMX4A3gXIgDLhcKeUeihP3bqvwUDt7jHBc9RVDcXq/RXzYHNHFHNHFE9HEHG/oMioDPGHko5Ti7a/28d9vqskvr2d7RSPtrp7TkcOCbKTGjuemJYksSZrIiYmRPrJWEATBg7OAL4FTgenAR4ZhfKaUauhd0DCMm4CbAKZMmcLatWsBSExMJCwsjLy8PAAiIyOZM2cO69evB8Bms5GZmUlubi4NDbratLQ0Wltbu+qYMWMGdrudBst4LI3lFBYWkpSU1JX1zW63k5GRQU5ODg6HA4D09HRKS0spKysDYObMmVitVgoL9VYz0dHRJCQkkJ2dDUBwcDDp6els3LiR5uZmADIyMigpKaGiQgeVycnJuFwutm/fDkBMTAyxsbFs3LgRgNDQUNLS0sjOzqa1tRWAzMxMduzYQVVVFQApKSm0trZSVFQEQFxcHJMmTaJzSuvYsWNZsGABWVlZXUkPlixZQkFBATU1NTgcDurq6mhsbKS4uBiA+Ph4IiIiyM3NBSA8PJzU1FTWrVuHUgrDMFi6dCl5eXnU1dUBsGDBAmpra9m1a9cxtVNlZSV79+7t0U75+fkAREVFDVk7tbS0dPmLP7QTQGpqqs/byeFw4HA4/Kad/OF6UkqxdetWv2on8P315HA4WLt27RG3U3+MyjV4dXV1hIeHD5JFI4eRpMua7VVc/89NhI8JICVmHMlTxnLcxFAmhNmJDAkkMtTO5LFBXVkx+2Mk6TKYiC7miC7mHK0uI3UNHoBhGPHA232swXsH+K1S6rOO158C9yilvuivTm/1kQWPXUxIbQFTf75tQN+bIxG5ts0RXcwRXTwRTczxRv84KkfwGhsbxcFMGCm6KKX400c7iA0PZs3dywiwHttS05Giy2Ajupgjupgjuhwxe4DTgM8Mw5gEzASKh+LEZm1lGz+ZibVZ7K1rYlpkyFCY4XeID5sjupgjungimpjjDV1GZZKVzqFgoScjRZdPt1XxVWk9t58645iDOxg5ugw2oos5oos5oktPDMN4CcgGZhqGUWoYxvcMw7jZMIybO4r8EjjJMIytwCfAT5RS1UNhm1lbhUbGEmK0srN09K7DEx82R3QxR3TxRDQxxxu6jMoRPGHkopTikY92MC1yDBctiPG1OYIgCKYopa48zPFy4MwhMuewREZPBaC8dBekTvetMYIgCEK/jMoRvPj4eF+b4JeMBF0+LKykoLxh0EbvYGTo4g1EF3NEF3NEl+GDWVsFRegHZjUVu4fYGv9BfNgc0cUc0cUT0cQcb+gyKgO8iIgIX5vglwx3XdxuvfYuYUIIF8ybMmj1DnddvIXoYo7oYo7oMnwwbavQaAAc1aN3LzzxYXNEF3NEF09EE3O8ocuoDPA6068KPfF3XZRSPP1ZMef/OYtTHl5L2q8+JvmB98n83adc8VQ2N63IYVtFI/972gxsgzR6B/6vi68QXcwRXcwRXYYPpm0VpgM8GitobnMNrUF+gviwOaKLOaKLJ6KJOd7QRdbgCcOCg61O/u/Vr3hn6z7mTx1PSsw4Qu02xgRaqXG0sreuma1l9aRNC+f81MEbvRMEQRCAoHG4LHYmcoDtlY3Mixvva4sEQRCEPhiVAZ6kaDXHX3RRSlHtaMNmMbAHWCg/0MItL25m534H9y2fxY2LEzGModuHyV908TdEF3NEF3NEl+GDaVsZBu6wyUypraGwvGFUBnjiw+aILuaILp6IJuZ4Q5dRudG54L+43YrbXt7CO1/t6/F++JgA/nzlAjJnTPCRZYIg+IKRvNG5N/BmH6lWXsbO7QU8N/9lfnXhXK+cQxAEQRgY/fWPo3IN3rp163xtgl/iD7r8/oPtvPPVPq47KZ4HzkvmnnNm8X9nz+St2zJ9Ftz5gy7+iOhijuhijugyfOirrYxJc0kwyikqG5Lt+PwO8WFzRBdzRBdPRBNzvKHLqJyiOdxGLYcKX+uyKmcvf1u3k6vTp/Lz85OHdBpmf/haF39FdDFHdDFHdBk+9NlW0SlYceOq/BqXexlWi398Rw8V4sPmiC7miC6eiCbmeEOXUTmC5y+Bg7/hS12yd9Zw33+2snjGBB781hy/aiN/ssWfEF3MEV3MEV2GD3221SQ9LTPBVcLumoNDaJF/ID5sjuhijujiiWhijjd0kTV4gs/5qLCSu1Z9SdTYIF675STGBQf42iRBEPwEWYN3ZHi1j3S7cP86hudalhJ12Z8473jJWCwIguArZA1eL/Ly8nxtgl8y1Lq0tLv42ev53PhCDnERY3ju+kV+GdyJv5gjupgjupgjugwf+mwrixUmJZNs2UNhecPQGuUHiA+bI7qYI7p4IpqY4w1dRuUavLq6Ol+b4JcMlS5ut2JDSQ0PvVnI9spGblycwN1nzcRusw7J+Y8U8RdzRBdzRBdzRJfhQ39tZYmey5zyV/lbef0QWuQfiA+bI7qYI7p4IpqY4w1dRmWAJ/iG3TUH+U9uGa/lllJa18yEUDvPf/cEliZN9LVpgiAIwkCYlEKYeo7q8hIg3dfWCIIgCCZ4NcAzDONs4DHACjytlPptr+N/Ak7peDkGiFJKeX331AULFnj7FMOSwdDF6XJzsM1Fc5uLpjYnu2uaWLdjP+uL9lO8/yCGAZnHTeDHZ83kzORoggP9c9SuO+Iv5ogu5ogu5oguw4d+2ypaJ1qJaipif2MrE8PsQ2SV7xEfNkd0MUd08UQ0MccbungtwDMMwwr8FTgDKAU2GYbxplKqsLOMUurObuVvA+Z7y57u1NbWMnbs2KE41bDiWHXZV9/MxU98zr76lh7v220W0hMjueqEqZwzdzIx44OP1dQhRfzFHNHFHNHFHNFl+NBvW02aA8BsYw9f72tgYtjomYEhPmyO6GKO6OKJaGKON3TxZpKVE4BvlFLFSqk24GXggn7KXwm85EV7uti1a9dQnGbYcSy6KKW47z9bqWtq477ls/j1RXN59PJ5/OuGdPJ+fiYvfPcEblicOOyCOxB/6Quf6dJUC+t+D+0thy/rA8RfzBFdhg/9tpU9DNe4eGZbdlO4b3QlWhEfNkd0MUd08UQ0MccbunhzimYMsLfb61L6mLBvGMY0IAH4tI/jNwE3AUyZMoW1a9cCkJiYSFhYWFf2mcjISObMmcP69esBsNlsZGZmkpubS0OD7ojS0tJobW3tqmPGjBnY7Xby8/MBiIqKIikpiaysLADsdjsZGRnk5OTgcDgASE9Pp7S0lLKyMgBmzpyJ1WqlsFAPTkZHR5OQkEB2djYAwcHBpKens3HjRpqbmwHIyMigpKSEiooKAJKTk3G5XGzfvl2LFxNDbGwsGzduBCA0NJS0tDSys7NpbW0FIDMzkx07dlBVVQVASkoKra2tFBUVARAXF8ekSZPoTJk9duxYFixYQFZWFk6nE4AlS5ZQUFBATU0NDoeDuro6GhsbKS4uBiA+Pp6IiAhyc3MBCA8PJzU1lXXr1qGUwjAMli5dyqOvZ7Nmex1XzQrkinkTqa2tZdeuYtrroSrw6NqpsrKSvXv3+rydWlpauvzFH9oJIDU19ajaKS8vr2sx74IFCzraaRdw5NeTw+HA4XAMeTvFV75P/NdPUry/iT0TlnW1k79cT83NzV3+4g/tBP5xPTkcDlpbW4+4nQT/wzo5hbkNuXw0ygI8QRCE4YLX9sEzDOMS4Gyl1A0dr68B0pVSPzQp+xMgVil12+HqHYw9fvbs2cPUqVOPqY6RyNHqUtnQwhmPrCNpUhirvp+BxTKyNrIUfzHHZ7qsuhYK34BJKXBzFvjZxqniL+YcrS6yD96RMSR95Jrf4F73Oy4Ie4W37jrrmM41nJBr2xzRxRzRxRPRxBxv9I/enKJZBsR1ex3b8Z4ZVzBE0zMBwsLChupUw4qj0UUpxf2r82l1uvn9JcePuOAOxF/6wie6KAW7syFoHFTmw57sobfhMIi/mCO6DB8O21bRKVhQ2Gq20dLuGhqj/ADxYXNEF3NEF09EE3O8oYs3A7xNwAzDMBIMwwhEB3Fv9i5kGMYsIBwYsjs12WjRnCPVZed+Bw+9VcjHX1dy15lJJE4cmdOpxF/M6VOXvV9A+RbvnLS2GA5WwdJ7IGg8bPx7/+VfuwHW/8E7tvTBqPWXbe9CY2Wfh0etLsOQw7bVpBQAZnYkWhktiA+bI7qYI7p4IpqYM6w2OldKOQ3D+CHwAXqbhGeVUgWGYfwCyFFKdQZ7VwAvK2/NFRUGndVbSlmRvZvcPQewGHDhvCl8LzPR12YJ3qatCQLH9F9GKfj39WALhNtyB3/65O7P9e/jToPGcsh+AurLYFyMZ9kDe2DrvyE0GjLvAos3n2eNcloa4OWr4MRb4Ozf+NoawduMn4Y7IJTZzt0UlDcwf2q4ry0SBEEQuuHVffCUUu8C7/Z674Ferx/0pg1mREZGDvUphwUD0WXlxj3ct3orM6JCuW/5LC6cF0PU2KAhsM53iL8Au7LghQvhmtWQsBjoQ5eKr6ChVP9dmgNxi/qu0+2GfV/qUbm6XTogm34qzLmw78/s/hzGRMKEJEj7Hnz+F9j8Tzj1p55lC9/Qvx0VsG8LxCwc2P96jIxKf6krARSUbuqzyKjUZZhy2LayWDCiU5i7dy//Lq8fGqP8APFhc0QXc0QXT0QTc7yhy6h8pD1nzhxfm+CXHE6XL0pqeeCNfJYmTeT9O5Zw05LpIz64gxHuLznPwsanDl8u+6/gbodPfqFH6ehDl+3vAQbYguCrl83rqi+Dtb+Dx1LhH6fAa9+DT38JX70Cb/wQDlb3bceez2Fqhh4ZjEiApLNh83PgbPUsW/A6REwHw9ph19BwzP7iaof92wfHmMHkYDW8fSeUbvY8VquzhLIvz7wtGOHX0VFgGMazhmFUGYaR30+ZZYZhfGkYRoFhGOuGyraBtJURncIsYw/5paMnwBMfNkd0MUd08UQ0MccbuozKAK8znbjQk/50Ka1r4pYXNzM1YgyPXzkf6whMptIXPvOXmp2w8gr92xu0NsKHP4MP74eG8r7L1e3SAdKEJCj9Ar75BOhDl23vQFw6zDoX8l8DZ1vP4+/fC4+mwNpfQ2QiXPwP+MEGuK8cvv8ZtDf1vWauYZ+2ZWrGoffSb4KD+6Fgdc+yB/ZCWQ7Mv1qX3/Yu/eJqh+J1elTxGBmQv7Q3H5pu2ptNT8MTGfp/HSxaG3VymqOldDP8fal+ILBlhefxTh91tUHFVtMq5HvXg+eAs/s6aBjGeOAJ4FtKqTnApUNk18DaKnouY1QTbZU7aHMe+3UzHBAfNkd0MUd08UQ0MccbuozKAE84MpranNz4wmbanG7+8Z00xgUH+NqkkU9TLfzrUtjxHmx48vDl92/XoydHwtZ/Q5tD35R//pe+y216GgwLXP1vGD8V1vyqaxSvB/WleormzHPg+CuguQ6++ejQ8eJ1sOEJmHsZ3P4lXPsGHH8ZRM2GwBCYmAQLrtHnMwtq93QERNO6BXiJp8CEmbD+4Z7BZOf0zOQLtT1VBf0HTNl/gRe+pdeRNR/oeaxutx7B/O9jh36qv+m7roGw4Un45znm/2fxOlAuPQI5WKy+WZ+vYd+Rf3bzc/DPs7UPhCdA9Q7PMrUlENCxPrOfaZrCIZRS64HafopcBfxHKbWno3zVkBg2UKafBsCpfEFRVaOPjREEQRC6MyoDPJvNq0sPhy29damob+GRj3aw9A9r2VbRwONXzWe6v2XKbG8Gt3fTdA+Kv7jaB17W2QavXAP1e2HKAti6Ctpb+i7f3gIrLoLnzgPHAO8BlYKcf+pseMdfrtexHazxLNfWBLkrYPZ5EB4PS/5PZ8jc8b6nLp3TIGcu12vpQiZCXsfuJy4nvPcTHSCe/6ieXmnGsnvBGqinbPZmdzYEhEB06qH3DAPO+AXUFMEX3TJqFr4O0XMhcroO8AC2v29+TrdbBzHj4nRA+o9ToLJAZ4R898fw54XwwX3w0QOHfv55tl4zaMKA/KXoQ/27pNdTO7f70NYPhYMU4H39Fmx7G1DwzcdH9tktL8Jb/wvxmfD9dRB/ch8BXjFMngdjY/oM8OR794hJAsINw1hrGMZmwzCuHaoTD6itxsfRMmk+51o3UFA2OjJpig+bI7qYI7p4IpqY4w1dRqXSmZmZvjbBL+nUpaXdxb3/2cqbeeW4lWJZ0kRuXJzIScdN8LGFvTiwF/6arkegxk+F8GkQvxhOvmNQMyYes798/ZZO1/8/r+kb5f5QCt6+A3Zn6emLIRNhxYWw/R1I+bb5Z3JfgIYywIBPHoIL/np4m8pz9Wjb8oe1Zl+9Ahv/Bqfe37Pc1n9DywE44fv6deoV8NkfYc3/I/P7n/Usu/09veZtwgwdeM29VI/GNddB3iuw/2u4/EUICO7brrBoyPghrP89ZNwGsd0So+zJhrgTwNrra2vm2TDjTL2ub+5leq1g6SY4rSOfU+R0mDgLtr8LJ97sec7iNXp079vPwLhYWPUdePp0fczZqkcVM+/UbQF6tOqfy2Hl5fDdDyBobI/qDusvzQf0VhIAuz6DtOsPHdu/TesdNUcH0nW7dGB9tLTUwzt362D3YLUO8BZcM/DPb3tHj9pd/SpYrHq0dMuLeoR5TMShcrXFMON0CJ3YZ4An37tHjA1YCJwGBAPZhmFsUEp5RNiGYdwE3AQwZcoU1q5dC0BiYiJhYWFdKbgjIyOZM2dO13Qgm81GZmYmubm5NDToIC0tLY3Jkyd31TFjxgzsdjv5+XqpYFRUFElJSWRlZRETMo8Uyz/5pKSAHKMSh8MBQHp6OqWlpZSV6a1vZ86cidVqpbCwEIDo6GgSEhLIztYPM4KDg0lPT2fjxo00NzcDkJGRQUlJCRUVFQAkJyfjcrnYvl2vT42JiSE2NpaNGzcCEBoaSlpaGtnZ2bS26nWgmZmZ7Nixg6oq/eArJSWF1tZWioqKAIiLi2PSpEl0bgw/duxYFixYQFZWFk6nE4AlS5ZQUFBATY1+AFZXV0djYyPFxXrdaXx8PBEREeTm5gIQHh5Oamoq69atQymFYRgsXbqUvLw86urqAFiwYAG1tbXs2rXrmNqpsrKSvXv3HradAOx2OxkZGeTk5Ax6O02YMKHLX/ylnVJTU/2inRwOh9+0k79cT1u3bvW7dvL19eR0Olm7du0Rt1O/KKWG1c/ChQvVsbJ58+ZjrmMksnnzZuV0udX3X8hR037ytvrFWwVqV7XD12b1zTt3K/VQhFLv36fUK9cq9eTJSv18rFL/vl6p9tZBO80x+cuOj5R6KFLblfWYeZm9m5Ta/LxSHz+k1IuX6rKf/lofc7mUemSOUi9caP7Ztial/pCk1LPnKPXB/fqze3MOb9frtyr1q2ilmg/o1y9frdRv4pRqrj9Uxu1W6omTlXriJP13J1tWKvXzsWrnW3869F5zvf4/P7j/0HtlW7Q9a3+v1K/jlHr+gp719EVLg1K/S9T/U2f5plqlfj5OqbW/M/9M9Tf6/KtvUerzv+jzVn9z6PhHP9e+0lTn+dmXr1bqdwlKtbfo1w37dDu8ekPPOrqzc42ub8XFSjnbexw6rL/kr9b2PZqq1O+P66nJF//Qx3au6fCZR/uvSymlWg8q9c0nSn34gFKrf6BUed6hY2/dodSD45Uqy1Xq9R/oNu5lb5+4XEr9dpr+XCfb39d27c7udn6Hfm/9w0r993H9d0OFR3VHex2ht9Xxed/jjR8gHsjv49g9wEPdXj8DXHq4Ooe0j6zbo9TPx6oVf7j9mM85HJB7B3NEF3NEF0+OSpPyPKWeOUv3dSMUb/SPo3KKZmfELvSkvr6eB98s4P2CCn52XjI/Oy+ZaZEhvjbLHEeVHrlKvQLO+n9w2fNwc5aerpf/Grx0ObQ6jq7u8i09plQetb+UfAavXK3XmAWNO5RpsDtVX8PTp8Gbt0HWo3oEJ+OHsOwefdxigXlXw841esSyNznP6m0ATrlPT58MnQTv/bj/ZCEt9VqjlIu1XQCZP9Lv5zx7qNyebKjcCifc1HM/u+MvgwlJxOY9cmjd385P9MjZzOWHyk1O1SNna34F7QfhnN8NbF88exicci/s/q9eh1hfCns2AqpngpXuRE6HjFvhy3/pvfE6p2d2MnM5uJ2eUxQbK3QClnlXg82u3wuLhqtXwbf/0bOO7iQug3P/qOt77/96TBM+rL988zHYx8FJt+lN27tnzNydDWGTIWEpTJnf/zo8peD1W+F30/QU3ey/6Gmdf1+s9yL88iXdnif+QNd13Om6jctyetZTsh5eu9FzqvP+bXr0depJh96bkNRxrJvNnX4dkQixHdti9D7HQHQRevMGkGkYhs0wjDFAOvD1UJx4wG01Po69IXNY0LgWl7ufrWxrdppPAR9miA+bI7qYI7p4clSalH6h70fqTe6BRgje8JVRGeAJ5rxV3M6KDbv5/pJEvpfZxxqpoaK9BaqLoOgj+HKlnhLWnQ1P6KmZJ9/Z8/2T/1dPUSxeq5Nm9P7c4Sh4HZ5aptdeHQulm+GlK/T0umteh8jjOvYK60Vlgf597Rvw0yq44ysdsHYPhOZdCahD69k6aTsIWX/SwUB8pp4qePpDULYZ8lb2bdtXq3S2yoXfPfRezAK9bi77L3qt3Mor4NXvQtB4PdWyOxYrXLEStyUQ/nmuDmS3vwfBERB7wqFyhqHX9wGk3wwTZx5OtUOkfQ/O/p0O8v56op4WagmA2LS+P7Pkxzo4aijVyVW6E7NQT7Hc3iub5pYVOqHJwusGblsnC6/TQVrOM/DkSVD4pnnyme4opbOQTl+m9QY9TbPz2O5u20AkX6in0tbtNq+rPBe+fBFmfwuufg1+shvuLNA67PgAXr9ZT10+5T5dPvEUvWVEUbfEN0rpTKpbV+l9C7tjltRm/FS9BUb3dXjdA7zJqWCxedYleGAYxktANjDTMIxSwzC+ZxjGzYZh3AyglPoaeB/4CvgCeFop1eeWCr6ibtpyko1dlO40z55KS71e1/rJg0NqlyAII4TWxp6/hQExKgO8tLR+bhJHKS99sYf/FLVz8fwYfnL2LN8a8+mv4P9Fw1/S4F+XwOu3wPPf0qMJoH9/8bS+AZ5wnOfn5/+PXutVka9HNloGuE9Tc50ejbEF6aQjOz4AjsJf3G5441a9RunaNyAkUq9jMhvBq9kJGHprgd5ryzoJj4eEJXp0qvvI3Bf/0FsEnNJt3dzxl+sg6+MHYceHnqOYSumEItFzdVDXnaU/gaYavcaqM8HLxU9B4BhPmybMwHXdezAuBl68GL5+W+9J1/t/SLseFt99aERyoBiGXi93y+cQM18/wZsyv//1e/ZQOPs3Optj7/WKFqu2r+gjcOzX77ldsPkFHSD3NVJ3OM74JVz6PCg3rLoGnlrGCbH9ZJmtKoTGcj2aFh6vE7t0Jlo5sFsfm9YxYta54XtnRtDebHkRbMFw3iN6/Zs9FILH603f/zdPJ6y59HmdoRT0sdhFPUcxd/9XbzYPnsHv7mwIjda+24nFqh9W9BXgBQRr3zJZhyffuz1RSl2plJqslApQSsUqpZ5RSv1NKfW3bmX+oJRKVkqlKKUeHSrbjqStQuZfAoAj9zXzApue0d/BfT2oGEaID5sjupgjunhyVJp0BXgjd0TUG74yKgO8yspKX5vgH7Q1gdvNm3nl3Ld6K4tiQ/jdJcdj8eUed4Vv6H3QZp8PF/0drn8fLnsBqrfDi5foC/2Lp6GtERb/qO96Zp0Ll6+Aynz412V6tOtwfPSATkTxnbd1kos3fggHa47cX3a8pxOKnPqAnu4H+ua3vtRzX7iab/RNfn+BC8D8a3TCjd3/1dkdv1ql0/UfdzpMTT9UzmKBcx/WyUFWXqqn7z17ts7K+fcl8Lt4rcnC6z2nS049Ee7bB/eWwi3/hStXQtJZfZpU0WSB69/TozbtBw9lq+xOcDic9jM97fJoiEiAa9+ES5+D5b8/fPk5F8E9e82zdM67SvvBY6n6IULBaqjf0zPJyZFiGDoQuyUbLngCHFXYV13Z996FnaNnx52uPxu/GHZl6cC9c5+6zmmo4fE6M6VZNs22Jtj6KiRfcGiabXdCJ+qguncQP+N0HdB1Zlv9/C8wJhLiTuy5GbxSekrMtAxPP5mQ5DlFMyTqUBvHLoKyG4gF5AAAIABJREFUXI8pn/K9O3w4kraalpjEFjWDyF3veB5sa9JbjICeDj3MER82R3QxR3Tx5Kg06XxQPYJH8LzhK6MywOvMkjOqcbXDY8ez7a0/8aNXvmRRfAQ3zFYEWH3oEjU7dVAVk6YzGqZeoW8wky/QN/jlW3Tmwg1P6NGY6Ln915d0Fnz7aT368/JV/W81sCtLr+nLuBXiFumRq5YD8Nbt7N1jng7fFKX0dMLweB1sdBKRqEd5eqfWryka2OjR7PP1uq2XroQ/JsF/btT7kp3+oGfZyalw13a4ZrVez+ds1Wv9QibC3Evg3EdgwXfMzxMQNLB1cnRcR52jlJf/C2adN6DPHTGGobWcMn9g5fsaCZ12Ety6EZLO1A8RXvue1mTmucduo9WmN1W/7m3cbjesvOzQiHN3vvlYPzwYO0W/TlgMzbV6ZG/P5zpYi0o+VH7OhXrKbW+/2fa2fpo5/+ojs/O4juygOz/VU6B3vAeLbtTrMau3HwpMD+zRmVmnnexZx8SZ+ni7zs5GbYn2705iF+mAv6rncjH53h0+HElb2awWckOXEd1c5PlgY8uL0FQN0cfrtcLDHPFhc0QXc0QXT45Kk1EwRdMbvjIqAzwBPV3m4H62bV7DnCljeeY7aditPhy5a2+GVdfqKWCXPge2wJ7HZ52rg67dn+sb4sV3DazeORfpkZXitXqd1DNn6p9nz4Z3/0+ngXdU6b2+wuP1tDaA6BQ91W3b20ze98Hh11Z1UrJO35CffEfPQKNzRKn7Ojyl9A3RhBmHrzcgWI/ITE3X6+xuXAN37+g7yA0co9d4nfEQ3LQGbsvR2zSc+0dY9L2+g6CjITBE75M3iFtTeI0JM7R/fX+99o3THvD0tWMhcjr5Kffq6+uVa3qO2LY2wp4NehStk/jF+veuz/QIXtyJPXXsXEv4Za/1l1tWwPhpMO0Itx6ITtVB7Tcf65EVqx0W3aAfmMChaZp7eo0mdmdCEqB0gAjah3sEeB1TTWTD81FDzVTtPyrnuUNvOtv0LIO4E/WDipZ6PaInCIJwJHROzRzBAZ43GJX74M2YMYAb6hGKo9XJB/kVlHz+DncDMwOqeO76EwgLCvCdLkrBu3frqYNXvwrj48zLzb1Ej1rt3673Qxso867Un8t7CegI1JytesSu++bY16zuud4s44ew40Nm7ngSnlwHC67Va9y67//Vm8/+qBN9zLuq5/udN8Dd1+E5qvQXV6TJOkIzMn6gf/yEYX0dTU7VgZ4XiFp0ISREwurvw1u3w/I/6OmLJet1ptHjugV44+P0Grf81/Ro7vz/6VlZRILOALr+9zq4T1ymp+qWrNdrL480qLZYYPppeuTO2apHyUM79veblKKnaZ50m36Q0ns0sZPOTJrVO7TvNpb3DPDCE/S0z9KcHtNfh7W/jDKOtK3iEpJ4Oz+d87Ifh5ZaOOcPegp0Qymc9yc9igd6FK+7rwwzxIfNEV3MEV08OSpN2kb+FE1v+MqoDPDsdruvTfAJ7+fv445XvqSl3c3dYTrQmBlQiSVEj2D4RBe3Syc22fKizv4344z+y6dcfHTnSb1c/3TH2apHGYrX6RvSzqyGnViscPW/acx+lrDtr8H798AH9x9KWgH6BnzpT/RNeGmOvvE+69eHUu53EjIRAkL0dLZOar7Rv482wYePGa3X0eGw2+06cKothnW/09k1Uy7S6zsDQ/WIRncSFuuHDXAowUp3Lvo7PHsWvHIt3PCRDgYxIPXKozNwxhnw1cv674wfHnp/5jn6AUVTbcem8ieaB5CRx+kHJtU7DgWA3dc8GoaeptlrBE/8ZfhwpG2VMmUcF7Tfxqw58zluy9+gbIvO1Dtprva34jW6YGPlsA7wxIfNEV3MEV08OSpNRkGSFW/4yjCYUzX4dO5MP5poc7r55dtfEx8Zwqs3Z3Brx8w+S8uBrq0EBl2X0s16Glhfa9+crXod1Kan4aTbe2aDHApsdr29wKn364yNZgSOYbNK0dMcb86CzDv1KMv8/9F7p7U69Pq+p0+DD+7TWwWYrW8zDH1j030EryvAG55P+UbjdTQQunQ55T644RP9UCJ/Nex4X2fs7D0lNH6J/m0L1klVehM0lv/P3nmHtXWdf/xzmMYGbIYxyzZeeMaOR+xgOx5JmjirmU2a0aRpZkdGZ1abtGnTlY78mtkmzWr2bnbSJN7BE++FsfHANsYGzDDDjPP749VFElwhARIgOJ/n4RG69+jq8D1HQq/exVWvy+NeuRzWvQwj5nv2dHtj+Hwx0EadDQMzncdHnyt5outfFuNtqE14Jkie5oCh4km39nPzLynSp0lRIZfiRma/BA9tXavM5GhCQkJ5e8D1EoVRcUjC0U/7ibz3RTuKTVUcCsBsOw+zh+0xuthjdGlJuzTpBTl4gdgrvdKD1xt5Y81+Dhyr5vnrT2FaRjwscpTnR4uh0bcNIY++sONTePM6qK+RsvwXPu4eVllVAm9dL7lx33hQ+td1d5JPapnzdtbv5APx4j9JUYr590m5ejviM6Bou/N+8U7JgeqfHrApG7qY9Gnys+APEv5o18dv2GnOsZ7yAQcMgStfg+fPk9fUWQ+2f079EuDbr8Kg8e7HU06W8OIlf5H7Q2y8iRYDR4sRaBl4cc2qls74viMPtZWWEYYeQ2RYKBPT+7N4xxHuWvANuHWpvLdbOaRWNeEeUEnTYDB0Mr2gimYg6JUevKSkpK6eQqdSW9/AEwvzmDxkAHMzHfk2R3c6DS6HJ8knXRob4YM7Yc9yz2PWvyperaSxcNlzUkDl32dJ8/AvfgNPnw4Pj5AG2Rc+0e2Nu1Z1CQ2DqdfBbTnyAXzWnZ7Hxg+XXmdW+XirOEVIqH8n3En0tteRr9jqEhkDEy+3D0+LSZb8Tm/N1tOnwWXPSmPzjlb+HL2gpQcwJESKrdQck16QrVUtTRwl7xtHcyW8OWqA+/nI6BbGndkvwUN71uqbk1LZeqic3MMV8qXV5GucIb5RcfJlVpBX0jR72B6jiz1Gl5a0S5Ne4MELxF7plQZeZmam90E9iDfWFHCwrIYfn5mJUkpeJJWFjl5coU2lrX3SpXCDNAF/6/qm0E43vn4M3rtVQh+v+0DC036QLZUbV/1LqqqFhEnz65sXtr3Mexfgky7hfSSHqbWKjPHDoeGEePpAjGy7Ru1BQm97HflKu3T55qNSRMgbVn/H8D5tfw5fGH2u3Kaf0vpeThwte3n3Ip9zqsx+CR7as1bnT0olNETx3roDLU8qBTGDgt6DZ/awPUYXe4wuLWmzJo2N0vcYerSBF4i90isNvGXLlnX1FDoNy3s3dWgcp41KlINW7tfAMRA3tOm+T7rkfSG3VSXSWsC1fcDXj8Hn90nfuqvfdDY+joyR8vw/3gJ374UbPpe8t5RJfvorA4vf9osVxlaSDw31kqPiawXNbkhveh21haDWZdgcaVrurdjRwNFyW7bfZwMvqHXpZbRnrRKjI5k9MpH/rj9IY6NNW5mYlKDPwTN72B6jiz1Gl5a0WROrgib0aAMvEHulVxp4vYnXV+/nkKv3DpzNaBNGyk/z5rStsfMLydU5/T7Y9j5sfF2Or3nWYdxdBJc+27KKJEjYjmX09UZcWyUc2wuN9UFbYMXQQwnvA3duhKzbWh/n2rsxPjirwBr8z0WTUzlwrJq1+0pbnoweJFU0DQaDwVfcDDwfqmhu+xAObwncfIKIXllkpSeXrq2pa+CzLYVsPVTO9kMVrN5TwikZccwameAcVJwHOKo6xo+APctAa++6VJdCwSppMj7zdsj9TPLqKg5Jbt2os+GSp/3bRLsb4Lf9EpsKoRHiuYtNlWNB7MHrya+jjhD0uoRHeR8TFSeevuNFPnvwgl6XXkR71+qscclEhW/m3XUHOCWjWb/QmBRpSRPEmD1sj9HFHqNLS9qsieW165sANV4MvIZ6ePtGGH8RXPxU+ybYWWx5V3rKOoqdmTYJfiIry0P57x7AXW9v5I7X1vPssnyKKmpZMD6ZP182yem9A8n9GjBYvq1PGCH9iioOeddl9yIpoz7yTCkMcvFTcv+LX0slwMtfbD1vJ0jx234JCYW4DPHgNbVICF4Drye/jjpCr9HFCtP00cDrNbr0ANq7Vv0iw/jGuEF8vOkQJ+ob3U/GJENtGZyo8sMMuwazh+0xuthjdGlJmzWxDLzYVPld24R/WxzZBvXVUFXc/gl2BlqLIbrpzaZDgdgrvdLAW7NmTVdPISAsyT3Cf9cf5AfzRrD1wQV8csdp/O2KkxmW2M99YHGe07Cwbot3edcl7wvo0x/SHKXe4zKkCfOkK6XseqAKP3Qxft0v8cOhZI8Y2VFxUrI+SOmpr6OO0mt0SXQkhccPa32cg16jSw+gI2t10eRUjlXVsTj3iPsJq1VCEFfSNHvYHqOLPUaXlrRZE8vAi0kF3SBV2T1xYK3c2hUA7E7UVkiKTpQzyiEQe6VXGniVlZXeBwUZNXUN/Oq/mxmW2I/bzxhFeKiHpdVacu6s3C+rQXFxXuu6aA15X0qTZNcQzLHniyfPU++3HoBf90vcMKcHL4i9d9AzX0f+oNfoMuVamHsX9I33PpZepEsPoCNrddqogcT3i+C99c2qafaAXnhmD9tjdLHH6NKSNmvi6sFzvW+HZeBVd3MDz5qfy//OQOyVXmng9UQeX5jH3uIqHrpoAn3CW+mrVnlYSs5axkVsuvQnskIGPXF4i+TaeauuZ2id+OFQd1zeiILcwDP0clJPhvn3dvUsDN2M8NAQzp+YwhdbD1NeU+c8EZMit0FeSdNgMHQiTQZemvt9Ow7kyG139+BZ84vy7cvR9tIrDbwZM2Z09RT8Sl5RBU8t3sXFk9OYOTKx9cFNuV8Oz11IiPxesrt1Xaz2CCPO6PiEgwy/7hcrnK2uKugNvJ72OvIXRhd7jC7BQ0fX6pIp6dTWN/LBhoPOg9GD5DaIK2maPWyP0cUeo0tL2qyJVUWzyYPnodDKieNQtE0K2dWUQWND+ycZaGw8eIHYK73SwCsoKOjqKfiN2voG7n1nM1Hhodx77ljvD7Ar7hE/HIrzWtcl7wsYdBLEpnRswkGIX/eLa0GKIDfwetLryJ8YXewxugQPHV2rSen9GZMcw+ur9zsPRsVJtEgQe/DMHrbH6GKP0aUlbdbEMuisz56ePHiHNkqO3tCZgBYjr7tiefD6OmswBGKv9EoD78CBA94HBQE1dQ3c+p+1rNpTwq+/OZ6BMT6UWT26U/7J9h/sPJYwEkryObh/n/1jaitgXzaM7H3eO/Dzfuk/GJTjZRfkBl5PeR35G6OLPUaX4KGja6WU4opTBrOxoIwtB8usg5KHF8Q5eGYP22N0scfo0pI2a1JbIZ9ZLWPIk4Fn5d+NdKQRdecwTZsQzUDslV5p4PUEqk80cNOLa1iUe4TfX3wSl0xJd5w4BnuWe35g8S4JyQxxWfqEkdBYR2TtEfvH5C+Rij8jz/TfH9BbCYtwGtcJpkG0wWDomVw8OY2IsBDecPXixSQHdRVNg8HQydRWQmSM/EDrBl7/Ic72Pd250Ep1CaAgakBAnyagBp5SaoFSaodSKk8pdbeHMZcrpbYqpbYopV4J5HwsRo8e3RlPEzDKquq4/vlVLMs7yp8vnchVM4Y4T37xa3jhfKj0YKwV57U0LBz3M+M9bId92RDWBwb3znhyv++X+OFi5PnSULobE+yvo0BhdLHH6BI8+GOtBvSNYMH4ZN5dd4CaOkc+TJB78MwetsfoYo/RpSVt1qS2Qqq0R8Y679txYC2kTXF6xbq7B69Pf+mN7CAQeyVgBp5SKhR4HDgHGAdcqZQa12zMKOAeYJbWejxwZ6Dm40poaCtVJrspWmvW7z/GL97awIw/fMHqPaU8csXJfGuaS6jliSrY/LY0H89f3PIiDXVQmu9skWDhCBUMr9hr/+Sle2HAkB7ZxNwX/L5fTv8VnP+If6/ZBQTj66gzMLrYY3QJHvy1Vt+ePpjymno+3eww6mJSgtrAM3vYHqOLPUaXlrRZk9qKZh48myIrx4/Csb1i4PWNk2Pd3YPXrL1QIPZKID1404E8rfVurfUJ4DXgwmZjbgIe11qXAmitiwI4nya2bt3aGU/jNxoaNd/59youenw5H248xMWT03j/R7O48OQ094HbPpDNr0LsDbxj+yTUsnnuV7+BEBlLWf46+wmU7XfP2etl+H2/pE+FUcEf7hpsr6POwuhij9HFHaXUs0qpIqXUZi/jTlFK1SulLuusuflrrU4dlsDQhL68usqR3x09SP5HnTjul+t3NmYP22N0scfo0pI2a1JbId67sEipkGnnwbPaI6RNDR4PXrMWCYHYK4E08NIAl+B7ChzHXMkEMpVSy5VSK5RSCwI4n6Dlf1sLWZZ3lNvPGMXKe8/gD5dMZHxq/5YD178EcRmQeQ7sXtTyvF0FTZDk9/jh9K3ykOR5bJ948AwGg8HgL54HWv2f54iE+RPweWdMyN+EhCgunzaYlfkl7D5S6dILL3i9eAaDoRM5UQER0fJ7ZKwHA2+tODZSTpbQRxUadB68QBDmbYBS6gLgI611Y4CefxQwD0gHliilTtJaH2s2h5uBmwFSU1NZtGgRAMOHDycmJoYNGzYAkJCQwPjx41myZIlcPCyM2bNnk5OTQ3m5uHWnTZtGZGRk0zVGjRpFZGQkmzfLl6hJSUlkZmaybNkyACIjI8nKymLNmjVNneZnzJhBQUFBU9Wb0aNHExoa2mSBJycnM2zYMLKzswGIiopixowZrFy5kurqagCysrLIz8+nsFD+0Y0bN46GhgZ27NgBQFpaGunp6axYsYKHV9QwqF8od5wxilUrV1BbWwvA7Nmzyc3NpaioiD7Vhzk1fwnlU2+nsLyGzGMfsXf9YhJGTmXNmjUAjCz6inRgRV4xNQ4DcM6cOWzZsoWkhlhiqrZRWlpKRUUFu3fvBmBYWhJDq4rZXVLPvkWLiIuLY9KkSSxevBitNUop5s6dy4YNGygtLQVgypQplJSUsGfPng6t0+HDh9m/f3+Xr1NYWFjTfvG0TitXrgQgOjqaadOmkZ2dbbtOABMmTKC2tpadO3cCMHjwYAYNGtS0TrGxsUyZMoVly5ZRX1/vtk7FxcUATJo0yW2dMjIyiI+PJydHvsnqjHWqqamhsrKy26yTL6+nzlinkJCQpv3SHdYJusfrqaamhtra2javU09Fa71EKZXhZdhtwNvAKQGfkAvJycl+u9ZlU9P52/9yeWNNAXdnWr3wCoOyyJQ/delJGF3sMbq0pM2a1FY404oiY+wNvIM5MHCM5OqBtGTp1h68Ukhyy1gLyF5RWuvWByj1EpCF/JN5Vmu93acLK5UF/Fprfbbj/j0AWus/uIx5CliptX7Ocf9L4G6t9WpP1502bZq2PmC1l9raWiIjfWgp0A1YvaeEbz2VzYMXjufarAzPAxf+Hhb/GX68WXLxHj9F8rymXe8c86/5UFcNP1xh8/g/oBf/CXVfIYT3cR4v2g5PzIBL/w0ndVqEULcimPZLZ2J0scfoYk97dVFKrdVaTwvAlLoch4H3odZ6gs25NOAVYD7wrGPcW96u2R3/R37v+dVsP1TOsutTCHkqK2j/n5jXtj1GF3uMLi1psyYPj4Qx58MFj8BTp0FsGlz1mvO81vDwCBh9Dlz4uBx77BRIGguXv+jfyfuLh1Jh6ndhwe+bDgXi/6NXD57W+hqlVCxwJfC8UkoDzwGvaq09lLMBYDUwSik1DDgAfBu4qtmY9xzXfU4plYiEbO72NqeOkp2dzbx58wL9NH7hX0t2M6BvOJdNTfc8qLEB1r0MI06H/umy4WNSJQ/PMvAObZBvORb8yf4aCSNRaCnCkuTSMP2YI3eiF+fgBdN+6UyMLvYYXewxurSZR4C7tNaNSqlWB/o7ymX58uVNHzb84e2dM3QgX20v4oUlO7keOLx7EwMyLwgar7wVPVFZWcmsWbOMV77ZOi1cuJA+ffp0m3WC7hE9UVlZybx587rNOnWH15PWmsTERJ/XaU5NOSGRMSxevJhJ1Y1Qs58B0LROfaoPc2pVMcWRQ9nkeN87NaQfoeVFLHfc71avp8GpRNYdZ/fhYxxZubJpnY4cOUJ0dHSb16lVtNY+/QAJSJXLPcAnwE7gNi+PORfIBXYB9zmOPQh80/G7Av4GbAU2Ad/2No+pU6fqjrJw4cIOX6MzyCuq0Bl3f6j/+tl29xNHcrX+3wNaF6zVurFR67wvtX4gVutNbzvHvHOL1n/M0LqhQe6/f4fWv03SuqrE/skK1so1tr7vfnzV03K87KC//qygI1j2S2djdLHH6GJPe3UB1mgf/08F2w+QAWz2cC7f8f92D1AJFAEXebtmd/wfWVlTp8f88hN9z9sbtH5woNaf3efX63cW5rVtj9HFHqNLS9qkSf0J+fy56E9y/+UrtH5ytvuYTW/LmAM5zmOvfFvrJ2Z2eK4BoeygzHfVM26HA/H/0ZccvG8C1wMjgReB6VrrIqVUX4dh9mgrxuPHwMfNjt3v8rsGfuL46TSiooKj/9gzS3cTERrCtTMz3E9kPwZrn4dlf4fkkyAkTGKOx5znHDN8Hmx4FQ5vhvhhsOlNGH+JjLPDKrxiFWKxOLZPKhdFD/LPHxWEBMt+6WyMLvYYXewxurQNrfUw63el1PNIiOZ7nfHc/l6rfpFhnDV+EB9vLuSh2GRUkBZZMXvYHqOLPUaXlrRJEyvfzmqRYJeDV7pHbl3bf0XFw8H17Z5jQLGKvzQrshKIveLVwAMuBf6utV7ielBrXaWUusHvM+oEZszo/g27j1TU8nbOAS6bmk5idLO43N2LYfh8GHs+rH0BDq6DU38gZWQths11jF0kFYZOVLrn4zWnT6wYcUebG3j7JewzJJAFV7s3wbBfugKjiz1GF3uMLu4opV5FCowlKqUKgAeAcACt9VNdOLWArNWFJ6fy3/UHKYtPYECQGnhmD9tjdLHH6NKSNmlyQkIfWzXwKg5BZH9ngRWAqAHdt4qmVfylWZuEQOwVXz61/xpYZd1RSkVZlb+01l/6fUadgE+xq13Mo1/tpK6hkRtnD3M/UbpX8uQyF8ApN8KtS+G2HDjjAfdxsSmQmCl5eGufh6TxkN56Ibby8CR7D14vzr+D4NgvXYHRxR6jiz1GF3e01ldqrVO01uFa63St9b+11k/ZGXda6+9qHwqs+ItArNVpowYS1zec/JqYoG2TYPawPUYXe4wuLWmTJpYx19QmwcbAKz8on3dd6RsP9TVScLC74cGDF4i94ouB9ybg2iKhwXEsaLGSTrsr2buKeTF7L9+dmcHwgc3KhFsNzIfPcx5LGOFe+dJi+DzYtRAOrRfvnZdE/co+KfYGXi/vgdfd90tXYXSxx+hij9EleAjEWoWHhnDexBQ2lUehKw75/fqdgdnD9hhd7DG6tKRNmtiFaDbUQn2tc0zFIWd/TQvLO1Zd2v6JBgrLg9c3we1wIPaKLwZemNb6hHXH8XuE32diAOB4bT2/eHsDQxP68ouzx7QcsHsRRCfDwNHeLzZsLugGCIuCiZd7HV4dlQpVR50virpqOF4EA4a27Y8wGAwGg6EZF56cxsGGAagTlVBb2dXTMRgM3RnrPSIy1v3W9b2j/KC0TnDF8o4FIkyzdC8s/Rs0+tAa/NlzIKdZq4Zq+xDNQOCLgXfEUWgFAKXUhcDRwE0p8GRlZXX1FDzy50+3U1BazcOXTSIqItT9ZGOjI/9urldvHAAZsyEkHE66FPr09zp8yJTT5ZfiXXJbViC3A3p3iGZ33i9didHFHqOLPUaX4CFQazV1SBylfTPkzu6FAXmOQGL2sD1GF3uMLi1pkya10sKgKb/O8uRZxxvqofJwyxBNy3gKRLPz9S/Dl7+R2hetUVsB+76Wz+yuVJVAeN8WUXeB2Cu+GHi3AvcqpfYppfYDdwG3+H0mnUh+fn5XT8GW7F3FvOAIzZw+zMa6L9oqHjargIo3ogbADZ/B2b/3PhbYX+XYcFaYptUDr5eHaHbX/dLVGF3sMbrYY3QJHgK1ViEhivjJF7BXD+LEwj9Lz9Ygwuxhe4wu9hhdWtImTexCNF2PHy8C3dgyRDOQHryjuXK7/cPWx5VJL7ymKp8WVSW23rtA7BWvBp7WepfW+lRgHDBWaz1Ta53n7XHdGauBYHeivKau9dBMcMm/89HAA0ib6pP3DmB/ZQio0JYGXi8vstId90t3wOhij9HFHqNL8BDItbrq1OE81XAREUUbIS+46rSZPWyP0cUeo0tL2qSJXRVNcBp45Y5c3thU98cF0oN3VJq6ezfwHBFwzQ286hLo27JdWSD2ik+175VS5wE/AH6ilLpfKXW/t8cYfEdrzT3vbOLgsRr+drlNaKbF7kXSr65/emDmERIOcUOdG/jYPumx1/zbEYPBYDA0oZTqp5QKcfyeqZT6plIqvKvn1R0ZHN+XmOlXc0AnUvXFH4LOi2cwGDoJuyqarsfLHV6yzvLgNTaIAyQqXjx5R3I9jy13GHhVR90rf3rw4AUCrwaeUuop4ArgNkAB3wKCuurGuHHjunoKbryyah8fbTzEz84azdShHha+oQ72LPc9PLMdjBs3TppFNuXg7ZdvRkJ9aZfYc+lu+6W7YHSxx+hiTw/XZQnQRymVBnwOfAd4vktn1AECvVY/OHMsz6sL6Xt4DexZ5jxRWdStvXo9fA+3G6OLPUaXlrRJk9oKCO8HIQ6nR1ORFYfBZFXjbV5kJSxSHlfVrIpm3pfOz7ftoWy/tF+YfrPc3/FRK2MLnL+X7nX+Xl3SokUCBGav+OLBm6m1vhYo1Vr/BsgCMv0+k06koaGhq6fQxLZD5fzmg61cm1HKrdXPeP7nVrAG6o67t0fwMw0NDeIhLNklBV2O7TMVNOle+6U7YXSxx+hiTw/XRWmtq4BLgCe01t8CxnfxnNpNoNdqQN8IkufeSJFDybZ7AAAgAElEQVQewLHPHpJvxlc9DY9Og5cugaPdMwukh+/hdmN0scfo0pI2aVJb7vTaQcsiK+UHpZBgs5YDgBhRrh68xkZ4/Tuw6I9tn7SFFd02Yj6kTobtrRl4B5y/H3Mx8KpKbOcbiL3ii4FX47itUkqlAnVAUMfs7dixo6unAMDx6ho+e/53vB92Dw8W/hC18gl4+Vuw4bWWg/MXA0oqYwaIHTt2SE+9uiqoOAjH9vf6/DvoPvulu2F0scfoYk8P10UppbKAqwHrv76HWPvuT2es1dWzR/Na+EUMKMxGPzkTPv4ZxGfIyQNrAv787aGH7+F2Y3Sxx+jSkjZpUlvpwcBz8eDFpECIjSkTFeeeg3dsjzhJjrYSVumNI465J2bCmPOgYLUzD7A5ZfslIg6ceXiNjVBzzDZEMxB7xRcD7wOl1ADgYSAH2AO84veZ9DIqa+t5/anfcmftP0mPi4Jz/wI/3iIG3Lu3QPbjzsHVpZD7GaSebOva9SuJjg1ZtE1ePL28gqbBYDD4wJ3APcC7WustSqnhQPD1AehE+oSHknH2Dzmi+1NbfhQu/Tfc+JWUEPdWgtxgMPR8aiucLRIAwqOkEGBTDt7Bli0SLJp78Iq2yW1xXvvzfo/mivetbzyMOV+O7fjYfmz5AUg+ScJKLQOv5phU/Qz053gHrSZXOZLGv9RaHwPeVkp9CPTRWpd1yuwCRFpamvdBAaTk+Am++9wqrizZRm3UAKLvyHaevPpNeOcm+Oxe2JctcbwH1wMaznggoPNKS0uDhH5yZ/ciec5e3gMPun6/dFeMLvYYXezpybporRcDi6Hp/+ZRrfXtXTur9tNZa3X+1FFcvfwJjlQrPhyzgD6hoZAyCQ7kdMrzt5WevIc7gtHFHqNLS9qkSW2FuwdPKbnvauAln2T/2Kh4iUKzKNoqtycqoaLQs2HYGkd3ivcOYOAYiB8hYZqn3OA+rrFRQjTHnC+FCy0Dr7rUObdmBGKvtOrB01o3Ao+73K8NduMOID09MFUofeFQWTWX/zObHYUVnJ1ynMiBI90HhEXCZc/BKTfBzi8grA/MvQuu/wRm3RnQuaWnp4u7O7wf7PpKDhoPXpful+6M0cUeo4s9PVkXpdQrSqlYpVQ/YDOwVSn1866eV3vprLUKCVHcdsGp5JXBc8v3yMHUKVC4UQqLdTN68h7uCEYXe4wuLWmTJicqnYVVLCJjxcDTWqLMmhdYsfDkwYP2h2kezXVGuSklYZr5S6CmmVlUdRQaaiXFKS7DaeBVFTvn1oxA7BVfQjS/VEpdqpRSfn/2LmLlypVd8rzlNXV8+18rKCyr4YXvTSe+9iDED285MCQUzvsL3HcIvvcpzL8Hhs60jzP2IytXrpRNmzDC+W2HycHrsv3S3TG62GN0saeH6zJOa10OXAR8AgxDKmkGJZ25VjNHJHLm2CSeWJhHcWUtpE2RSnWuH8i6CT18D7cbv+pSWwnPLoBDG/13zS7C7JeWtEmT2nJniwSLyBg5XlMm9SI8eeKi4qH6mBRwAnk/SZ4ovxfvbPvEq0rEcEt0qTE55nxorIOd/3Mfa1XQ7J/uMPD2ilfPygm08eAFYq/4YjHcArwJ1CqlypVSFUqpcr/PpBfwm/e3sr+kiueuP4VTh0RLEmb8MM8P6CqbOsHhVVQhnr8dMRgMBoNFuKPv3UXA+1rrOsA0ePORu88ZS1VdA498sVOq00H3zsM7UQWHt3b1LHomhZskPWXP0q6eicEffP5LWP3v9j22eYgmOEM0rRYJnvo0R8UBWgzB+hPifRt5hkSotadKr1VB09XASz9Fnmf3IvexTQZemhh4DbVQWej0KNo0Og8EXg08rXWM1jpEax2htY513I/19rjuTHR0tPdBfuaTTYd4O6eAH80fySkZ8dKCAG3vwesimnSxDLyYFAiL6LoJdRO6Yr8EA0YXe4wu9vRwXf6JFCDrByxRSg0FgvaL0M5eq5FJ0Vw1fQivrNpHXv1A6NMfDjbLw9v+EfxtnHtlvE6mSZdV/4J/zYWaoF1iv+LX/VKyW24rPFQn7EwaG6WqeX1tux7ew9/zvNPYCKufhdXPNB3yWROtW1bRBKeBZzU5j021f7wVBllVIoVVGushabxEqLUnRNN6jBWiCRJVlzIJDm92H2vNrf9gGJAhv5fuadWDF4i94kuj8zl2P36fSScybdq0Tn2+w+U13PPuJial9+e2Mxybw3oT60YGXpMu1gY2+XdA5++XYMHoYo/RxZ6erIvW+h9a6zSt9bla2AvM7+p5tZeuWKs7zxxF3/BQfv/JDnTq5JaFVlY+JR+ctrzb6XOzaNKlaCs0nHB+q9/L8et+KXE0ovZUfr4z2btMqppvea9dD+/J73k+UbZPWhMUbW0ybnzWpL5Wwh89GnhWk3MPBp5lRFWXOlOOksaKB649IZpHcyE0omVv6EETxJvfUO88VlYAYVHi3YvLkGOle8WDp0LlC6xmBGKv+BKi+XOXn18BHwC/9vtMOpHs7Gzvg/yEznmRukdnoOuq+dsVJxMe6pDcMvDiWgnR7GSadEkYIbcm/w7o3P0STBhd7DG62NOTdVFK9VdK/U0ptcbx81fEmxeUdMVaJURH8qPTR/LV9iL+WzQIXbQV6hxteMsKIN8RsrfpzU6fm0WTLsWOEK8j27tsLt0Jv+6XYoeBV1Hov2u2lwNr5bawffmAPfk9zyeKXF4f+1cBbdDEqpTZ3hBNy4NXXSL5dyFhYtwljpLqmnXVPv4RDo7ulOi2kGbtTZMnSgimq9FYtl/y75RyVKJXTg9e33jb9KtA7BVfQjQvcPn5BjABKPX7TDqR2tr2udvbzIEcGj/8Cel1e/jjTBgx0MUFW5IPETHQL7Fz5uIDTbokjASU85uHXk6n7Zcgw+hij9HFnh6uy7NABXC546cceK5LZ9QBumqtbjptOL+/+CQWVw5GNdbz0n8/oKauATa+AWiY/B3Jzzq2r0vmV1tbK6FjxsBzw6/7pSlE86D/rtlerDzQdhp4Pfw9zzuW5ywkDPZ9DbSiSf0JOOISOnnCi4FXflB60oVF2l8vypHnVuUw8BJGSspRwkhAO79I8BXXCpquWG0aCjc5j5UdkPw7kPnFpomBV11iG54Jgdkr7SnLWACM9fdEehxVJeg3rqVC9wVgQfxh9/MluyE+o+sKqbRGn/5wzdsw49aunonBYDAEAyO01g9orXc7fn4DdJ/4+yAhJERx1Ywh/OqWawDIXbeUm15Yjd7wGgzJgjk/k4Gb3uq6SVaVOMuit7fcusEerV0MvML2N6T2F5aBd2hj188lGCnaJpFgqVNg3wrP4yqPwAsXwOPTndVzPXrwYiXss6wAYjyEZ0IzD95WCc8EZ5GU1sI0j+2HZY84K3DW14qB5lpgxSJxFIRGNjPwCsSDZ2G1SrA8eJ2ELzl4jyql/uH4eQxYCnTPLqQ+Mnv27MA+QWMjvPd9dEUh3635KTURcajCDe5jSvO7Vf4dNNNl5BnQL6HrJtONCPh+CVKMLvYYXezp4bpUK6Wa/kCl1CygjTFA3YeuXqv45AyIHsR3M0o5tms16ugOmHiFfFBKn95lBt7s2bOd3rs+A4wHz4Hf9ktlkfQ+GzBESuA37y/WmRwvFk9xXAbUHHNWRmwDXf066nKKtklD8CGnSk5tXXVLTQ5thH/Ng0PrAe1sOWAZeHZtEgCO7mi9WXlkf6kEX3ZAjKukcXLcSkFqrZLm6qfhiwdgxRNyvyQfdIO9gRcaLsajZeDVn4DKwxDrwcDz4MELxF7xxYO3Bljr+MkG7tJaX+P3mXQiubkB/tZt+SOQ+ymvx93C3qixhKefDIdcDLyGekm47Eb5d9AJugQpRhd7jC72GF3s6eG63Ao8rpTao5TaAzyGtBgKSrp8rZSC1MkMq93BXSnrqNXhLI10fACaeDkUbYHDWzp9Wrm5uU4DL/Ns+T/e1lyeHojf9otVYGWoY627Mg/P8t5NuVZuXT00PtJhXbb+V6rH+pO8L2H/av9e046GejHCksZKH+fGOjiw1l2TbR/As2cDGr73GQwcC7u+lHOt5eCBeNk8FVgBqXAZFecIDdVOD15EPzG+WvPg7Vkut1/9TgxBuwqariRPkP2htSO0WLf04FUWyjkPLRIC8Z7ri4H3FvCS1voFrfXLwAqlVF+/z6QTKSoqCtzFtYalf6M64xvcdzCLq2YMITR1snyTYZXaLS+Qzd7NPHgB1SWIMbrYY3Sxx+hiT0/WRWu9QWs9CZgITNRaTwZO7+JptZtusVapU1BHc5lV/RUrI2Zw27v5FJRWwbiLpBJdFxRbKSoqEgMvJAxGnQVoU0kTP+4XKzwzY5bcdmUenmXgnXw1oNqVh9chXapL4b0fwMLft/8azTm0AV65Aj75hf+u6YnSfKk0mzQOBs+QY/uynZpUlcA7t4jhdfMiSD0ZRpwOe7Olz2RtpYyLbNaVrcng062HaIJ4yyzniuXBAzHUPIVX11bI2p98jeTPvf8jOLJDziV4MvAmShP0ikL3HngWcY7Km9WlHj14gXjP9cXA+xKIcrkfBXzh95n0FI4fhRMVLG0Yj1IhXHPqUOmT0VjvTDgtyZfbbmbgGQwGg6H9aK3LtdZWc7SftDZWKfWsUqpIKbXZw/mrlVIblVKblFJfK6Um+X3C3Zm0KYBGVZcy+qybaGjQ/ODlHOqiEmDEfNj0tqRDdDYlu+Qb+UHj5b714c/QcYp3ifFsGQRd7cFLGAUxyVKYox0evA6x+hkJVy3Oc+aCdYQTx+GtG8S5ULhRjKhA4tqaoG+8eOf2ulSKXPlPyaX75mMQnSTHRp4uFSn3fg21jrfRSA8hmtB6iCbI8+pGaVngWjQwcZR45uzyKvetlHDMky6Ds/8gRZ2yH5NCKc3nYuFaaKXMpQeehetz9+281CdfDLw+WutK647j96D24E2YMCFwFz+2F4AP9kewYEIyKf2jxMAD5zcJTT3wuleIZkB1CWKMLvYYXewxutjTC3XxVkHreWBBK+fzgbla65OA3wL/8tO8vNIt1ip1itz2TWTQlPP446UT2VhQxuur98NJ35IeWwWrOnVKEyZMECMkYSTEjxBP4lFj4Pltv5Tskj5jVnhbeRd78FIny+/JJ0muWBtpty4nqmDFUxDWB+prpOx+R/nkLjEWs34kDoeDAS6lUbQNUM68taFZsH8VE8aNFS/ZyqdgzPkwyMWzNmSmFCzZ9VXrRVYsfPHgAQwc7d7eIGGUVOmsPNzyMXuWQkg4DJ4OJ18FI78hOZiewjPB+WXP4U3OtYp19eBlOH/3UGQlEO+5vhh4x5VSU6w7SqmpBHHyOAS4dG3pHgBya+P53qwMORaXIQmfloFXmi+b2Nvm7GR6fUlfDxhd7DG62GN0sacX6tJq2T2t9RKgpJXzX2utrZZEK4B0T2P9TbdYq34JUjnzlBshNJxzT0pmekY8j3yxk+PDzpZv5de91KlTqq2pdhp4YREShWMKrfhvv5TsliIY4VFSxKarPHgVjnwpy8BLmShfKFS3rUNYu3VZ/7KE/M29S+4f6WB+1pZ3Yd1/YPaP4bSfyrH9K1uO+/DHsPjPHXsui6Kt4sSIcPiDhsyEExU0HtoIa54Vo2l2syCHiL6Sr7frS/FeqhAIb+ZPcvPgefkMbRlTruGZ4DTW7MI09y6X6IGIfpILfMEjUlk+5WTPz9Onv3wxUbgJyg+IYRnhMu9+A51/Rzdrk3An8KZSaqlSahnwOvAjv8+kE9m5M3Ax87pUPHixKSOZMsSRTKmUvEE0efDyxegLaU+XisARSF2CGaOLPUYXe4wu9vREXZRSFUqpcpufCsCf3+DdAHzix+u1SrdZq+99CvPvAUApxd3njuFoZS3PrDoKk74t/fEqOy9fcN/mFVBf7azEN3C0CdHET/tFayjeLZ5RkAbWVjPrzsbKv0tz+DaaQvBso6k90i5dGupg+T8kTHXqd+VYR9pxHN4CH9wBadNg/r1i9CRmSiiiKzVlkPMibH6n/c/lStE2d8NqyKkAVGz4AL5+DIbPg/SpLR834nT50qRom/SKbt5KrC0hmlYvvEGeDLxm61NbKdU+M1wqWvZPh9tyRLvWSD7JEaLZrEUCyN9gefE8ePAC8Z4b5m2A1nq1UmoMMNpxaIfWus7vM+khFO3LJVTHcvnMMSjXjZkyCVY9LS/eku7XIsFgMBgMbUNrHeN9VMdQSs1HDDyPdbSVUjcDNwOkpqayaNEiAIYPH05MTAwbNsiXiwkJCYwfP54lS5YAEBYWxuzZs8nJyaG8XHJepk2bRm1tbdM1Ro0aRWRkJJs3y4fbpKQkMjMzWbZsGQCRkZFkZWWxZs0aKislm2PGjBkUFBRw4IDko4wePZrQ0FC2bpW8nOTkZIYNG0Z2tuTkREVFMWPGDFauXEl1tQQIZWVlkZ+fT2GheHHGjRtHSngD0waF8uTCXOaddw4TG55n3xv3kj/8GqKjo5k2bRrZ2dlN34bPnj2b3NzcpgIGEyZMoLa2tunD1ODBgxk0aBBr1qwBIDY2lilTprBs2TLq6+sBmDNnDlu2bKG4uJhIR5XHw/UxbFu0iGFVfRhSvJuK0qPkbBB94uLimDRpEosXL0ZrjVKKuXPnsmHDBkpLxQM0ZcoUSkpK2LNnT4fW6fDhw+zfv7/L16mmpqZpv4wbN46GhgZ27BDDNy0tjfT0dFauFIPC0zrt3vA1mXXH2VncQNzRo0RHxHPiwA5yFi1q8zoBTJo0iYqKCnbvlpSYjIwM4uPjycnJ8bpO/XPeZSghVPTLoGTPHg7srmQWULptMQyY4PM6VVZWUllZ2aZ1SitezqiyfezMvIkDqzYyMzyWkMPbyN+5s83rlPfuHxm26W/Uh/WFCx4nf+cuCgsLGR02hEH7VnL44AF25MprYXT9VlIa62k8msvSr/5Hv9i4dr+e1q7KZs7RPArjZ5ACTet0auRAhu59AxqqWR9zG8cWLWqxTqP6jyUNaMj9nLrwWHZs2OC2TmEN1cwGGkMjWbJiPSjl8fVUWljGcCD/eBRDGxud6xQawuzwvhze9jXbjo9oej2Vr3uXVN3AhrJYEg8caNPraWhNDBnFu6ivraYsMo3Nixa5rdOEhmgSgRNh0XzteK24vp4qKytZtGiR7ftea6+nVtFat/oD/BAY4HI/DviBt8c5xi4AdgB5wN02578LHAHWO35u9HbNqVOn6o6Sl5fX4Wt4Ytsf5+mND0zR1Sfq3U9seEPrB2K1PrRR698O0vqTuwM2h/YSSF2CGaOLPUYXe4wu9rRXF2CN9uH/TTD+ABnA5lbOTwR2AZm+XrO7/4/sKHlFFXr4PR/pX723SetXr9L6D0O0rqnolOcu+viP8n/8WIEc2PC63D+8rVOev7vil/2yZ7loufN/cv/d72v9lzEdv257eOkyrR8/1f3Yw5lav3NLmy7TZl0aG+V5H5uhdUODHPv32fLTFupPaP3xXaLnv8/WuvyQ+/m1L8q5oh3OY2/fLMceiNW6cHPbnq85hzbJdTa95X78rRvl+NNnyt9qR2OjaP1ArOjQnIYGOfd/J3ufx6a3tX5woNYVh1uee3K21v+5xP3Y/36t9W/ita6t9H7t5mz7yKnfRz9ref6Tu+Wc3Vx0YP4/+hIjeJPW+piLQVgK3OTtQUqpUOBx4BxgHHClUmqczdDXtdYnO36e8WE+HWbQoEEBuW5ReQ19qwoIicugT3io+0mr0EruZxLi0Q09eIHSJdgxuthjdLHH6GKP0aVtKKWGAO8A39Fad2pjuu68ViMGRvPtUwbzysp9HBx/s+TydFIuXv/6I5JLE+MIDRvoCGzq5Xl4ftkvxY4eeK4hmpWH/VNBsi1o7V5gxaIdhVbarMu+FZK7Nut2ZwpPYmbbQzTfuh5WPgkzvg/XfSCVQF2xqpTuXyG3jQ2w83NnKOrhrW17vuYUbZPbgWPdj1vtL077acvQSwulJEwTWhZYAdElIsa3GhbjLoI7NzqrdLqSOKpliOaeZVLcKaKf92s3x9IO3AusWIw8E4bN9VhFMxDvub4YeKHKJdbQYbhF+PC46UCe1nq31voE8BpwYfum6V8sN7+/eX3VHlIoJm3YmJYnE0ZCRDRsfU/ud7MKmhA4XYIdo4s9Rhd7jC72GF3cUUq9CmQDo5VSBUqpG5RStyqlbnUMuR9IAJ5QSq1XSnWagN19re44cxQRYSHcn9MXPfhUWPG4NFb2hYZ6+/LoPlCRv04MEOvDd8IoQHUsR6oH4Jf9UuJokWCVl49JlnL1x490/NptoaxAnrO5gZcyUSqm1tX4fKk262JVhR11tvNYYiZUFcPxYt+usXuRNBCf/0s4548QGt5yTOIoyU+zCq0UrIHqEsi6TdagqBUDT2spNlNhU4HSomirXCdhpPvxSVexYeJvIPNs+8dZNBl4HtoS9I2HAUNavwbI67S5cWuRmAnH9kG5I8+ztlIqi2Z4jIRvnf7pUmzF+r05I8+A6953r+bpQiDec30x8D4FXldKnaGUOgN4Fd+SvdMA19quBY5jzbnU0evnLaXUYJvzQUF9QyNfrVpHuGpgQOrIlgNCQpxJmABx3c/AMxgMBkPnoLW+UmudorUO11qna63/rbV+Smv9lOP8jVrrOJcIl2ldPefuQlJMH+48cxRfbCti3eDvyAc168vT1ig/BP83Cb76bbueN6r6ICS4RN9E9JUPmr3cg+cXih39BUMdpSGsComBKLSy5ln46iH7c1aBldQp7seTT5L2AoFc6wM5sp/6uXh5rDYDvnyJ0NgI/3sAYtNh5m2exyklXrz9DoMy9xMxyDLPluezM/C++A08chI8lAx/yoC/ZsJrV0ORjR5F2+TLj7BmvqCwCErjT/bsvbMYMV9u7Tx4AFf8B874VevX8MaY88VT9+zZsvf2r5T1ba+Bp5Q0PAd7A68L8FpkBbgLSd62vlXcCHgwidvMB8CrWutapdQtwAvA6c0H+TuBXCnl9wTyZfnlRFYWQATsqwxlUG1ti4TXQjWIZEATwomoQeRv3+5zImVnJJBXVVVRWlra7sTknppArrXucAK5P9cJOpZA7q91qqqqanMCeWcWZOiqdWpoaGjaL91hnaB7vJ6qqqqora1t8zoZOp/Y2Fjvg7qY780axnvrDvL9VeF8HT+S0K//ARMu9fzhsaEe3r4Rygsg+3GYfrPnb/dtH19HVM3hll6JgWM6XsY+yPHLfinJd4ZngnNtyg+19KZ1BK1hyV/FKzbn5y2NkIPrxNixeptZWB/eCzdCaisl811osy4Hc1r+rQNdDLyhWa0/fut7cGg9XPQkhPdpfezg6ZD7KVSVSOrQkCyIGiCNyfevdh9bfwJWPCmev3EXQnSyePFW/hN2fAyTroR598AAh4+maCuk2VTIxEdN+iXClGs9XqMp5akjJE8Qj9pLl8GzC0SPkDBn+Gq7rnmS9NFrh4EXkPdcT8l5rj/AZOBhYC+wEPiRD4/JAj5zuX8PcE8r40OBMm/X9UcCeSC45pkV+sEH75EkyuJd9oPWvSznH5nYuZMzGAyGIIUeXGQlED/d9X9kINiwv1QPu/tD/e4zD8n/1l2LPA/+8ncy5qvfa/3rOK0/vbdtT3Y0Tx6/7mX345/dJ4UcGurtH2fwTmOj1r9LcS8+V3ZQ9F71jH+f69BGZzGMfatann/uPK2fOq3l8YYGrR9KtS+g4Q8qj8iclv692fPWa/3bJO/7tf6E1o9M0vrxLN/2Yv4yeb4V/5Tb5Y/K8cUPy/3qMufYvdlybOv7zeZ8VOb14ECtH0zU+pN7tC7ZI2MX/cn7HLoDRTu0/us4R/GXMzp2rYMbtP7wp84COZ1Aa/8fPYZoKqUylVIPKKW2A48C+xwG4Xyt9WM+2I6rgVFKqWFKqQjg28D7zZ7DtYnFN4FtPly3w1jfUvuLPUePs3TnUc5KrZXGjP09RJpa3zp0wwIr4H9degpGF3uMLvYYXewxugQPwbJWE9MHcN3MDO7KG0tdVCJ8/Q/7gbu+giUPw8nXSF+9iZfD6n9DZRvyu4rz5NbOg9dQC6V72vU39AQ6vF8qCqHuuPtno34D5fNUe5udb3wTlvyl5fEdnzp/3/e1+7kTVRKqN2xOy8eFhMCgCc6wRh9oky4H18ttWrPQ0JBQ2XPeQjTXPg+l+XDmrz3mebmROlk8Vkselvujz5Fbq3edayhq/lK5HTrL/Rr9EuDsh+D2HHlNrXwSHnNEkic1K7DioNu9twzMhBs+k79t8jUdu1bKRDjvL+3qcR0IXVqbxXYkXPJ8rfVsrfWjgM/ljLTW9UhD9M8Qw+0NrfUWpdSDSqlvOobdrpTaopTaANyOtE0IOFY4lT/QWvPHT7YTERrCxH6lUj3HLqkVIHG0FFqxYqq7Gf7UpSdhdLHH6GKP0cUeo0vwEExr9dOzRpMQG8tLjQsg74uWFQArCuHtm8QQO9fxYfa0n4pRlu3Ld9UOLAPPNYwQ5LrQqwutdHi/lEgIu5uBFxoG/ZKg4mDbr9fYCF/+BhY+BMf2u5/L/URC/xJGwt5s93P7sqHhhDThtmP8xRIC2fxxzWmoh5X/pG/xJpmLLxzMARSk2IR/Jo5qfX/VVsLiP8HQ2TDqG749X0RfCTs9XiRaJDj2tdUU3DUPb89SMW49NOmmfzpc+Dj8YIXk8UXGSmN1G7rle0v/dLj+Y2dj+S4gELq0ZuBdAhwCFiqlnnYUWPGSGemO1vpjrXWm1nqE1vohx7H7tdbvO36/R2s9Xms9yeEZDLpM5f+uP8inWwr56VmZRB0vgAFDPQ8ODYPvfQpz7+q8CRoMBoPB0EOJjgzjD5dO5NHy06hRkTQsd/HiNdTDWzdAXRV863n5UAvygXn8JbDqad+rExbvoi4suuWH3MRRclvUKQFIPRNHA/kmI8MiNqV9HryCVVC2H3Qj5LzoPF5xGA6shcxzJOdsX7a7AbZ7Eaaq57cAACAASURBVIRGyDk7plwrZe6X/a31589fBJ/8ginr7oZHJsCn9zrbQHjiQI7spT42uViJo6F0r+cKnrmfSuXP+fd4L2DiypBT5TZzgfNY/yHiiLC+KKmvFa9lxmnerzdwNFzxEtyzX9bO0KV4NPC01u9prb8NjEHy7u4EkpRSTyqlzuqsCQaCOXNs3O/toLCshvv/u5mpQ+O48bTh8gKMa8XAA0nC9PQtSBfjL116GkYXe4wu9hhd7DG6BA/BtlZzMwdy/+Wn8VrdPPTGN6kpdnhtFv0e9i6D8/8OSc3aF835mRh+K57w7UmK8wgbNKblB+g+/aUn15EdHf9DghSf9ktdNRw/an+ueBeEhEv1R1diUpxl7NvC5rchrI94tHJehIY6Ob7zM7kdvQCGzpQeiq6hiLsXSZENT33QIvrCqd+XnnGt9cQ7vAWAxvMfkdSc1U/Dc+dA9TH78Vo7CqxMsT+fOArQTkO4OXu/FqNs8Kme52SHFXI5+lznsZAQ8UpbHrwDOdK7ub3VJZsRbO8tnUUgdPEaKKq1Pq61fkVrfQGQDqxDKmsGLVu2bOnwNbTW3PX2Rk40NPKXb00itKEGKgulzG+Q4g9deiJGF3uMLvYYXewxugQPwbhWF01OY+A3fozSDSx88bfUbf8Mlv4VJn8HJn275QOSxkpFwOzH4J1bYNuHkoPlieJdHAu1b1JM0phe3SrBp/3y7i3w9On2PQiLtkmYoNUiwSImpe1tEhrqYcu7Eio48zb5XLbjYzm341MxIgdNcHrp9i6X2+PFUiFz+NzWr3/KTdJoe9nfPY85vBViUtkSOQ2ufBVu+Fw8bF8+aD++/KA0dW+ef2dhpfV4+hJh3wqpAtlcP2+MOQ9uWuhsQG4xaJwYqVpLeCZKDGI/EIzvLZ1BIHRpUyag1rpUa/0vrfUZfp9JJ2KVmu8Ir6/ez+LcI9xzzliGJfZzxnm3FqLZzfGHLj0Ro4s9Rhd7jC72GF2Ch2Bdq/PmZrE/5SxmHXufqtdv4FhsJifO+pPnByz4o7RWyP0UXr8aHh4BW99vOa78EJQXUBLmoa3CwLGSI+VrvlUPw+t+KdwEW/8Lx/ZK6KQrWkPBavuS+DEp0oC7Dc3F2bNUjKkJl0k+Wv/BUlCnrgZ2LxTDTyn5Mj4mRcI0AfIXy+2wea1fP2oAnHKDtCTwFHZZtAUGjXPqkjoZZtwq/ffsirQczHGM82DgJYwEFBzd2fJcdal424a0wwBTyt6oTBovulcWiZ7JreTftZFgfW8JNIHQpe2lXgzUNzTyx0+3c+rweL5zqsOgO7ZXbr2FaBoMBoPBYAgIGRfcTayqJlzXc/GRW5jz9xU8vzzfasfkTmwKXPQE/DwPrv0vRA+yD9nc9RUAJfEe+rENHC3hntbnAIM7i/8MylHZsbmBU5ovxkS6TVEOK4+rsg15eJvfEg/bqG9INcmp14nxtvY5WSOrWqRS4sXbmy1G5u5FUhzEl557WT+UXL3lj7Q811AvfRGtapQW8++V5u0f3OEMGbU4kCMVLZNPsn++iL7SY86u0Mq+lYB25tP5A6sC5sF1vuffGbodvdLAmzSpY00S1+4tJbr6ADdMSyQkxBGPb5VIDmIPXkd16akYXewxuthjdLHH6BI8BPVapU2BM39D1Hde4/7vfpMhCX359QdbeXfdAc+PCQ2XyokTLpUy+VUl7ud3fQX9khg580L7x1sfiHtpHl6r++XwFtj2Psy6A8KioGCN+3nrfvopLR/r2uzcF+prYesHMPZ8CI+SY5OvFePpfw9AeD93Y2XoTKnSeWyvGHgZp/kW5hidJCX1178q4ZWulOySCq2DxrvrEhkD5/5FvG1fP+r+mIM5YhC21pw8MROO2uyvfV9L/qKdgdxerCbv6/4D9TV+y7+DIH9vCSCB0KVXGngVFRUdevyKzdv5NOIe5m2733nw2F4IjZRvAIOUjurSUzG62GN0scfoYo/RJXgI+rWafSdqxDzmj07i1ZtOZdrQOB54fwuFZV5C/TIXSOVFh8cOkLDL3QthxOlUVFbaP27gaLk90jsraba6Xxb/WTxqM28T47ugmQevYLUYXnZ902JSHU/go4GX9yXUlkl4ZtM1BsGY88XoGjHf3Yiy8vA2vCaf4YbP8+15AGZ8HxrrnPl9Fo4CKySNa6nLmHNlLov/5ByntXjKvHkOEzPhaF7LMOC92fJYy6D1B/0SpQ/hjo/xZ/4d9ID3lgARCF16pYG3e/fuDj0+Y/NjRKtqwnd+LK51EA/egCHtanDYXeioLj0Vo4s9Rhd7jC72GF2Ch560VqEhioe/NYm6hkbuenujfaimRdoUKYOf+5nz2KH1UFUMI8/wrItVSbPIT4VW6mth4R8k1K8rOFElVcF9xKMuRdsk927GLZLDlX6KVJ90zakrWCO62zXntjx4vhp4m9+CqPiWhVKm3yS3Y853P540TtYu2xGWO3yeb88D0tIhNh32NGtQfXiLhKMOHG2vy7kPQ1QcvHCBFGMp2Q01ZZ4LrFgkZko1y/IC57G6ajEOh3po69ARksbJlx3JJ8l8/URPem/xJ4HQJXitkS7i0K5NnFv7GTuTz5NNv+gPcqJ0b1BX0DQYDAaDoScyLLEfdy8Yw+LcI7y+er/ngSGhMPIb0jC9sUGOWd684fNbf5KkMf7x4DXUw9s3wuI/wtK/dPx6baWxAV65HP5vEnz445bhqm1hycPSciDrh3J/8HTxeh3aIPfrqqVypV2BFZDPWGF9fDPwThyHHZ/A+Isk5NaVjNlw82KYeIX78ZAQaS1QWyYFV6yehr6glFx3zzL3yqBFW6UoSlik/eNiU+G7H0kO3wvnw/qX5binAisWKY4Qvm0fOo8dWCt6tqfAijesME2Tfxe09EoDLyMjo92PPfH5r6klnPBzHoKZt0s/lP2rxL0f5AVWOqJLT8boYo/RxR6jiz1Gl+ChJ67VtVkZZA1P4HcfbeOzLYUszztK9q5i9pc0a42QeZYU/bByw3Z9BckTIXpg67oMHCset45U0mxshPdvk5y1/oNh5/+chmZnseRhqZw46ixY+wL8Y7I0hG/l73LTpa5aQh6fXSD96Kbf5KzAaOXZWWGahzZCY719/h2IERWT7FsOXu6nUkRlwqX251NPto+wsrxfw+e1rUk4iIF3/Ih78ZPDUkETWnkdJYxwGnlL/ypGrF2IqitpU2DYHBlf6wgV3uuoADp4etvm7QvWfPyYfwc9873FHwRCl15p4MXHt7Pc6/5VDD38BW9EXMzQIRkw/Wbomwif3iMu9iAusAId0KWHY3Sxx+hij9HFHqNL8NAT1yokRPHnyyaigFv+s5arn1nJlU+v4PS/LmJHoUv+y4gzJMRu52dQWyFFV0acDnjRJWmMhNC1t5Km1vDp3bDhFZh3L3zjQXdDszPYs0zywyZeAVe9DrcukxC9j3/m9DTZ0KTLprfgr2Ok511lkfwN8+5xDoxOks9JViXNA1aBlVYKhMSkQoUPVTQ3vwPRyc68Ol8Z5mgwPaId3b8s4yd/idzWVsj6J4n3q9X9Yhl5MSky5+ZeRztOvx+qjsLKJ+X+vq8llNJPLQzcGHehrN1I/3ZF64nvLf4gELr0SgMvJyfHt4FF2+WFW7oXGupp+OxXFOkBFI6/AaUUREbD7Dudb1JB7sHzWZdehtHFHqOLPUYXe4wuwUNPXavB8X356mfzePv7M3njlixe/N50osJD+f3HLqGVUQOk5Hzu55C/VDxMjg+5reoycIzctqfhedF2eP0aWPVPyPoRzP2FGJWWodkZHC+W0NC4YXDeX8WbNWgcXPeBGCGWEWNDTk6OhHJ+9BNJVbnuQ7htraNyZrNQxcHTpbCK1f+u/2Bnrp0d/dOkemR9recxNeXi7Rx/kX0uX2ukTYUbv/Ts+WuNuAz3PDwrB9PhwfP6OkoYITpd/oJvzzf4FBh9Lix/FI4fFUO5rQatr/TpD/Pu9hxq2k566ntLRwmELr3SwPOZ/1wsibD/NxF+l0RowQr+Xn8pp43PcI6ZdgP0S5Lfg9yDZzAYDAZDT2ZgTCRTh8YxfVg8czIHctvpo1ice4QluUecg0adBYc3wbqXpMLj4Bk+XNiqpNkGA69kN7xzMzxxKuxeDKf/Es76nRhXUQOkemFuJxh45QfhnRulmMy3npOS/hZKiYetYHXr11j0B/FgXfwUDDvNc7hj+nTJqSs/IN5Jb+X9T75KwiBzXvQ8ZscnUiVz/MWtX8sT6dPaVyCveR5ekbOCps9E9BNjylfm3we15eIlPVHp1wqXhp5FrzTw4uJ8qAhUXyv9USZdCRf8A2bdweKka/gw9AymD3NxpUb0hdPvgz4D5NuYIMYnXXohRhd7jC72GF3sMboED71pra6dOZQh8X35/cfbaGh0FMvIXCC3Oz6SD/AOL0aruvTpD7FpvlfSrDgM/5wLW9+HWbfDHRtgzs/dDaNRZ8HhzVBW4Pk67UVrMUzeuBb+PgF2LYRz/uQs5uFK+inSkPz4UdtLJYeWwep/w9TrveeSWQbdtg+hbD+keTHwhs+HobMkN/BElf2YLe+I9ukByEXzRsZsCZs8skOqYob3a/qyPyCvo+QJcNJlUggI/NvgvBPoTe8tbSEQuvRKA8+nhoJW1aahM2Hqdegz7ueeskvIGjmIyLBmIQBTvws/3+X+rVcQYhpQ2mN0scfoYo/RxR6jS/DQm9YqMiyUuxaMYXthBW+ucVTYHDha2h6BWw6SV10Gjva9kubSv0rlx1sWS65av4SWYzLPlttAePGW/R2eP088h1k/gNtzYNr37Mc2FUexzwccs/c/EBEN8+/1/rzJJ0nD85VPuV/bE0rB6b+CysOw+pmW56tLpf/d+Iu7pk2VlYe3Z6lU0Ewa2zSPgL2O5t0j4bv9h0D/9MA8R4DoTe8tbcE0OvcTixcv9j7ISup1NNrcXljBwbIazhibZD8+NMxPs+s6fNKlF2J0scfoYo/RxR6jS/DQ29bq3JOSmTo0jr/+L5fK2noxKkY5jCtHgRXwQRdfK2ke2w9rn4PJVztDO+1IzJQ8r52fux8/vEWqVbYXrSXkcegs+Mk2CQuNH+55fMrJYlDYhWnu/ALy/id5g/0SvT93aLg05i7Nh5BwSJno/TFDs2DkmWKU1pS7n9v+kbQKGH+J9+sEgqY8vKVuFTQhgK+jhBFw9kMw52eBuX4A6W3vLb4SCF16pYHXaqNTi/KDchubAsDyPAlNmJvpwcDrAfikSy/E6GKP0cUeo4s9RpfgobetlVKK+84by5GKWn72xgYqaupg9o/hm49KTzMHXnXxtZLm4j/J7ZxfeJuYGJq7FznDE1f/G56cCa9cAQ11rT/eE4e3iIE18XJJM/FGRF8JDWxu4DU2wOf3URWVIlXFfWWww2uXPAHCo3x7zPz7pKroiifdj29+R0IivTUKDxRWHt7OL2R+gyY0nQro6+jU78PU6wJ3/QDR295bfCUQuvRKA0/50uvECtGMEQNv66FykmIiSe7fJ4Az61p80qUXYnSxx+hij9HFHqNL8NAb12rKkDjuPXcMn28t5IJHl7G5MhqmXOuWE+dVF18qaRbvgvWvSDjkgMHeJ5Z5NtTXiIdo9TNSqTJpPOQvlvYFrh8M876UNgWb3mr9mtveBxUCo8/z/vwW6afAgRz3vny7FsKR7ewZdjWERbThWtOd1/SVtCkw5nzIfgwKN8vfXVUixu/4i9vew86fDDsN6o7L7y4FVnrj68gbRhN7AqFLrzTw5s6d631Q+UFpPhkliY9bD5YzLjU2wDPrWnzSpRdidLHH6GKP0cUeo4s7SqlnlVJFSqnNHs4rpdQ/lFJ5SqmNSqlOc1H01rW6ec4IXrs5i5q6Ri554mteWuHuifOqixVuWdRKHt7C30vRltN+6tukMmZL4Y7Pfwkf/VQKwNy8UDyMa5+HFU+IsfP1Y/DyZfLl9KqnW7/mtg9gyEyIHujbHECMsRMVUkzEYsMrEBXHuEu8eCKbMyRLeghbOYa+Mv8+8Vo+NQsemQhvXge6of3VM/2FazPwQeObfu2tr6PWMJrYEwhdeqWBt2HDBu+DKg5JbxalqKlrIK+oknEpPdvA80mXXojRxR6jiz1GF3uMLi14HljQyvlzgFGOn5uBJ1sZ61d681pNHxbPx3ecxsyRCfzyvc18tf1w0zmvuliVNF2NIIsTx6Vi5ua3Ycat0vTbF8IiYcR8OJoLmefA5S/KsdPvh7EXwGf3wUuXwOf3iYdr9k9g/wrPlTeP5kkxkLEX+Pb8Fpa3zer7W1Mm+W8TLmPDljb2/uuXAL/YJXl1bWHQOLh9HVzwfxLeuX+1eDPtKn92JgOGSj+/6GS3puO9+XXkCaOJPYHQJfgrg7SD0tJS74MqCpsKrOQVVVLfqBmf2oZeJUGIT7r0Qowu9hhd7DG62GN0cUdrvUQpldHKkAuBF7UkZ6xQSg1QSqVorQ8Fem69fa3i+0Xwz+9M5cLHlvOLtzbx+Y/jiO8X4ZsuA0dLP7QjuXBoAxxaD/tWyG1jvRgBM29r24Tm/kJC/+b83BkKGRICF/8Ljp0Du76SyopzfiG5dcv+BlvetX+ebe/L7djz2zaH+OES0VSwWkJXt7wnoaOTrqQ0rxP3S2yKVC6f+l2oqwF014Zngjz/rDukD6ALvf11ZIfRxJ5A6NIrDTyfKD/YlLS79aBUberpIZoGg8Fg6DakAftd7hc4jgXcwDNI+4S/X3Ey33xsGb96bzOPXTXZtwcOHCsG1+MOj1dohFSNnHk7ZMyCwadCZHTbJpMyyd5LFdEXrnsfSvc4zyeMkN83v+PZwEub2vby+kqJF89qlbDhVanymTYF8rqoMmJ4N6qJMP2mrp6BweBGrzTwpkzxksqgtSNE01lgpW9EKEPjfag2FcR41aWXYnSxx+hij9HFHqNL4FBK3YyEcZKamsqiRYsAGD58ODExMU3hPwkJCYwfP54lS5YAEBYWxuzZs8nJyaG8XL7InDZtGklJSU3XGDVqFJGRkWzeLKmCSUlJZGZmsmzZMgAiIyPJyspizZo1VFZWAjBjxgwKCgo4cOAAAKNHjyY0NJStW7cCkJyczLBhw8jOzgYgKiqKGTNmsHLlSqqrpfx/VlYW+fn5FBZKy6Jx48bR0NDAjh0S/piWlkZ6ejor/7+9O4+Lq7r/P/46DGvYwk4CJEBCSEgiCcEQDCZRk7q29mutdanaVptatT9bu9rVLtZaq22sVttGW6tVq7V1qalLo1lQRAlKEkgCEUiABEiAhH3Ccn5/zAQhcyAbwwzcz/Px8BHmznXm8D5ncnLm3nNOYSEAISEhZGdnU1BQgN1uByAvL4/y8nIaGxsBmDdvHna7nYqKCgCSkpKIi4ujqMgxaAkLCyMrK4v8/Hx6e3sB+PrKNO59rZykp5tZHO9DS0sLbW1tVFZWApCcnExkZCTFxcUAxIWexZylfuxs0rSFpNAVPI1l55xHSUkJLbUtUFtEVlYWzc3NVFdXn1Y9NTQ0UFPj+A4grb9uoJ6SghYwo/Jxeg/sJr+09uN6mpMI+z7go9TrqNmw4aTraaaOJ7HxDba9/nfm7y2gMuVaEo4cISIiYqC9eKqeli1bRmlpKU1NTYBjX7GR6ikiIoLMzEw2btyI1hqlFMuXL3fUk/NqyunWU39/P+3t7UPryeKfpwULFrBt2zavqicwfJ7GuJ76+/vZsGHDSdfTSNR4W7I0OztbH208p6q6uprk5OThT+hshl+nwPm/hNxbuOKRAvq05vmvnnVa7+vtjpuLRUkuZpKLmeRidqq5KKW2aK2zR79Enue8RfM/Wut5huf+CGzQWj/tfLwLWHG8WzTHpI+0kL5+zWcfeYfdje08clkKZ50xy9NFOr5De+F38+G8Hw9dzKXgD/DaHfC1YseVvpO1e71jvt+0XMdtp98ohfAEaS/DkFxcSSZm7ugfLbnIytFR/rAGNjmPp79fU7a/dcIvsAInkItFSS5mkouZ5GImuZy0l4DrnKtpLgEOj8X8O5C6Gszmo7jvigX09muueaqCax8t5PkttY4N0b3V5GmO2ym3/3vo8R0vOxYlOZXBHThu7QTYWwCpKyA8AZD2MhzJxZVkYuaOXCw5wDuuNucm56FTqW3pot3eK/PvhBBCjBql1NNAAZCulKpVSt2glLpJKXWT85R1QCWwG/gzcLOHimp5KdHB/Pe2s7lkhh/VTR1887kScu76H3e9Ukb94W5PF89s7mXQsM2x2EuvHV7/Eex9B+Z++tRfM2gyRDu3glhw9eiUUwjhFpacg5eamjryCa3OL0nDplBWdxjAElfwjpuLRUkuZpKLmeRiJrkMpbW+6jjPa+CWMSrOEFJXrqZHBfPt82eTlJRE8d4WnijYw2NvV/P4O3u4LCuB21amMSU8yNPF/NjcT8Nr34fN90H9NsfKnou+ePIreB5req7jLqfZH6/CKe3FTHJxJZmYuSMXSw7wQkNDRz6hzTnAC51C6b5qbD6K9Pjj/D8TwHFzsSjJxUxyMZNczCSX8UPqyiw0NBSlFIumR7JoeiS3r0rnz5srebaohtfLGnjwqoWcNTPa08V0CJvqmCu39RkIjoWrnz35jcVNVv7UsSKo/8eLzkl7MZNcXEkmZu7IxZK3aB53Q8HWfTApCnwDKNvXyoyYYAL9bGNTOA+SDSjNJBczycVMcjGTXMYPqSuzY3OZFjWJn396HutuO5uoYH8+/2ghf9z4EV6zeN25P4DFq+HmgtEZ3IHjNs1j5vBJezGTXFxJJmbuyMWSA7zjaqsfskWCFW7PFEIIIcTJmxETwr9vWcoF8+K5+787+dJf3+fV7fs9vxBLch5cdC8Ee8lVRSHEmHHrAE8pdYFSapdSardS6nsjnPcZpZRWSo3JUthRUVEjn9C2D0Kn0NxxhP2Huy2zwMpxc7EoycVMcjGTXMwkl/FD6spspFxCAnx56OosfnDRHIr2tHDTk8Us/NnrfH5tIaX7Do9hKceetBczycWVZGLmjlzctg+eUsoGlAOrgFrgfeAqrXXZMeeFAq8A/sCtWusRN/AZjT1++vv78fEZYWx7bxqkX8DbGT/hmrWFPHlDDnlpE/8bsOPmYlGSi5nkYia5mJ1qLhN5Hzx3GJM+0qJONJeevn627GnhzZ2NPL+llsmT/Pjvbcvw952YmUp7MZNcXEkmZu7oH92Z8mJgt9a6Umt9BHgGuNRw3s+Be4AxW2v46G72Rn090HEAQqdSts+xw/2cKdaYFDpiLhYmuZhJLmaSi5nkMn5IXZmdaC5+Nh+WpEbx/Yvm8JvPZvLRgQ7W5le6uXSeI+3FTHJxJZmYuSMXdw7wEoCaQY9rnccGKKWygCSt9StuLMfJaW8AtGOLhP2txIcFEhUS4OlSCSGEEGKcOWd2LOfPjeOB9RXUtnR6ujhCCIvw2DYJSikf4H7gCydw7mpgNcDUqVPZsGED4Ng3IjQ0dGD1maioKObOnTswEvb19SUvL4/i4mJaWx1X47Kzs+np6Rl4jbS0NAICAti+fTsA022NpABbqw/yfsU+pgQ7IioqKqK9vR2AnJwcamtrqaurAyA9PR2bzUZZmePu0/j4eFJSUigoKAAgKCiInJwcCgsL6erqAiA3N5eqqirq6+sByMjIoK+vj127dgGQkJBAYmIihYWFAISEhJCdnU1BQQF2ux2AvLw8ysvLaWxsBGDevHnY7XYqKioASEpKIi4ujqO364SFhZGVlUV+fj69vY7J38uWLaO0tJSmpiba29tpaWmhra2NykrHt43JyclERkZSXFwMQEREBJmZmWzcuBGtNUopli9fTklJCS0tLQBkZWXR3NxMdXX1adVTQ0MDNTU1xnqKjY1l1qxZ5OfnAxAQEEBubq5b6unIkSMD7cUb6gkgMzPT4/XU3t5Oe3u719STt3ye7Hb7QHvxhnoC7/g8tbe3Y7fbT7qexNjz9bXkDkrHdaq5/PiTc1l530Z++nIZf75u4t1tLO3FTHJxJZmYuSMXd87BywXu1Fqf73x8B4DW+m7n43DgI6Dd+b/EA83Ap0aahzca8wtGVPYiPHsdvV/exJw/1PGlvBTuuHCO+95PCCHEsGQO3slxex8pTsnDGz7inld38uj12Zw3J87TxRFCTACemoP3PpCmlEpRSvkDVwIvHX1Sa31Yax2ttU7WWicD73Kcwd1oOfrtuVGrY5Pz/TqCnj7NjGjrfIM8Yi4WJrmYSS5mkouZ5DJ+SF2ZnU4uN+SlMDM2hG/840O+8Jf3+PGL21m7uZLXSusp3XeYw109o1jSsSXtxUxycSWZmLkjF7ddK9Va9yqlbgVeA2zAY1rrUqXUz4AirfVLI7+C+xy9bcmobR/Y/Nnd7ph3lxITPEal8rwRc7EwycVMcjGTXMwkl/FD6srsdHLx9/XhwasX8vv1u6lu6mBLdQttx+yTNyU8kOtyk7k6ZxrhQX6nW9wxI+3FTHJxJZmYuSMXt94Mq7VeB6w75tiPhzl3hTvLcsLa6iE0nqqDjsnQKdHWGeAJIYQQwj1mx4fx0DVZAGitaensobalk9qWLupauthYfoB7Xt3Jg29WcOXiady8YoYs8iaEOCVum4PnLqMxv6C9vX34yft/vQT6evhR1H288GEdW3/yCZRSp/V+48WIuViY5GImuZhJLmanmovMwTs5bu8jLWwsctled5g/b67kP1v3ExcawJ+uy2ZeQrhb3/N0SXsxk1xcSSZm7ugfLbnbYENDw/BPtu13XsHrIDU62DKDOzhOLhYmuZhJLmaSi5nkMn5IXZmNRS7zEsJZc+VCXrxlKRq4/JF3eLlkn9vf93RIezGTXFxJJmbuyMWSA7yjS4W70NqxyErYVKoOdlju9sxhc7E4ycVMcjGTXMwkl/FD6spsLHOZlxDOS7fmMT8hnK89/QH3vLqTvn7vvONK2ouZ5OJKMjFzRy6WHOANy94GT8QVrwAAIABJREFUPR30BMex73AXKRZaQVMIIYQQ3iMmNIC/37iEqxZP4+ENH/GFv7xHS8cRTxdLCDEOWHKAl5aWZn6izbFFwgGi0NpaK2jCCLlYnORiJrmYSS5mksv4IXVl5olc/H19uPuy+fzqsvkUVjbzyQfz2V53eMzLMRJpL2aSiyvJxMwduVhyS/mAgGFWpWp13Ode0+uY0JxqsVs0h83F4iQXM8nFTHIxk1zGD6krM0/mcuXiacyeEsbNT27hMw+/w/yEcGw+Cl+bInHyJG5bmcbUyUEeKZu0FzPJxZVkYuaOXCx5BW/79u3mJ9rqAajoCgUg2WIDvGFzsTjJxUxyMZNczCSX8UPqyszTuSxImszLX8vjU5lTCfSzAdDd08+LJXWsvH8jf9r0ET19/WNeLk/n4q0kF1eSiZk7crHkFbxhdTYBsLM1gJjQPkICJB4hhBBCeIeokADu/WzmkGM1zZ389OVSfrluJ89vqePmc2awKiOOSf4f/xtGa01Dq53Y0AB8fKyzOrgQVmXJEUxsbKz5ia4WUDZ2NmvLraAJI+RicZKLmeRiJrmYSS7jh9SVmbfmkhQ5ibXXn8nrpfX8/JUybnvmQ4L8bKzKiCNjahgf7j1E0Z5mDrYf4ablM/jehbNH9f29NRdPk1xcSSZm7sjFkgO8WbNmmZ/oaoGgyVQ1dbIqI25sC+UFhs3F4iQXM8nFTHIxk1zGD6krM2/P5RNz41k5J473q5t5sWQf67bt56WSfUyLnMSytBhaOo/wp00fcf7cOBZOixi19/X2XDxFcnElmZi5IxdLzsHLz883P9F9iL6AcJo6jljyCt6wuVic5GImuZhJLmaSy1BKqQuUUruUUruVUt8zPD9NKfWWUuoDpdRWpdRFY1U2qSuz8ZCLj48iJzWKX/7ffN77/kqKf7SKTd85h/s/t4A1Vy0kLiyQ7/xzK/bevlF7z/GQiydILq4kEzN35GLJAd6wulro9nWsoGnFAZ4QQgj3U0rZgIeAC4EM4CqlVMYxp/0QeFZrvRC4EvjD2JZSjHf+vj5EBvsPPA4L9OPuy+ZT0djOA+srPFgyIYS7WXKAN+xypF2HaFOOzc1TLbYHHsjytcORXMwkFzPJxUxyGWIxsFtrXam1PgI8A1x6zDkaCHP+HA7sG6vCSV2ZTYRcVqTHcvmiRB7ZWMm22tHZT28i5OIOkosrycTMHbkorfWov6g7ZWdn66KiIve8+JoF7LDN4uK669nx8wsI8LW5532EEEKcEKXUFq11tqfLMZqUUpcDF2itb3Q+vhbI0VrfOuicKcDrQAQQDKzUWm853mu7tY8UE8Lhzh5W/XYjYUF+PLN6CdEh8o9uIcajkfpHSy6yUlRURHa2IY+uFhoDJ5EYMcmSg7thc7E4ycVMcjGTXMwkl5N2FfBXrfV9Sqlc4Aml1DyttctGZ0qp1cBqgKlTp7JhwwYAUlNTCQ0NpaSkBICoqCjmzp3Lpk2bAPD19SUvL4/i4mJaW1sByM7OZvAAMS0tjYCAgIF9mmJjY5k1a9bAnJGAgAByc3MpKiqivb0dgJycHGpra6mrqwMgPT0dm81GWVkZAPHx8aSkpFBQUABAUFAQOTk5FBYW0tXVBUBubi5VVVXU1zv2p83IyKCvr49du3YBkJCQQGJiIoWFhQCEhISQnZ1NQUEBdrsdgLy8PMrLy2lsbARg3rx52O12KioctycmJSURFxc38PuGhYWRlZVFfn4+vb29ACxbtozS0lKampro7OwkNzeXtrY2KisrAUhOTiYyMpLi4mIAIiIiyMzMZOPGjWitUUqxfPlySkpKaGlpASArK4vm5maqq6tPq54aGhqoqak5pXr64mzFmg86ufSBjXx9gS8xk3xOuZ7efvttfHx8vKaeADIzMz1eT52dnSxbtuy06gkm1ucpODiYwMBAr6onOP3P0+nWU1NTE5MmTTrpehqJJa/gbdiwgRUrVgw92N8PP4vkqcAreS32Szz+pcWn9R7jkTEXIbkMQ3Ixk1zMTjWXCXoFLxe4U2t9vvPxHQBa67sHnVOK4ypfjfNxJbBEa9040mu7rY8UEy6XLXua+eJf3meSvy9/u2Exs+JCT+l1Jlouo0VycSWZmLmjf7TkFTwj+2FAs6fTTxZYEUII4U7vA2lKqRSgDsciKlcfc85e4Dzgr0qpOUAgcGBMSykmtEXTI3n2plyue/Q9PvtIAWckhnOw/QhN7XZCA31ZvSyVy7IS8bNZcrkGIcY1S17B6+rqIigoaOjB5kp4YCG3H7mJBZ/8KtflJp/We4xHxlyE5DIMycVMcjE71Vwm4hU8AOe2B78DbMBjWuu7lFI/A4q01i85V9X8MxCCY8GV72itXz/e67qtjxQTNpe9TZ189/mtdPX0ER3iT1RwAGX7W9lWd5jEiCBuWj6DWXGh9PVr+vo1SZFBTI/6+IvwiZrL6ZJcXEkmZu7oHy15Ba+2tpa0tLShB7sOAXCYYJKjrHkFz5iLkFyGIbmYSS5mkstQWut1wLpjjv140M9lwNKxLhdIXQ1nouYyLWoST69eMuSY1pq3djWyZv1ufvjC9iHP+dkUd192BpcvSgQmbi6nS3JxJZmYuSMXS153PzoBcogux2TNQzrEsrdoGnMRksswJBczycVMchk/pK7MrJSLUopzZ8fxws1n8cItS3nyhhye+nIOz6xewuKUSL71XAm/eW0X/f3aUrmcDMnFlWRi5o5cLHkFz8g5wLP7hZEwWS4fCyGEEMLalFIsSJo85Nii6RH88N/befCt3VQ1dXBp3Pia6iOEFVjyCl56errrwW7HLZpRMXH4+KgxLpF3MOYiJJdhSC5mkouZ5DJ+SF2ZSS4OfjYffvWZ+Xzvwtm8snU/fyj14XBXz5BztNZ8WHOI1u6eYV5l4pP24koyMXNHLpYc4Nlshj3unFfwpsbFj3FpvIcxFyG5DENyMZNczCSX8UPqykxy+ZhSipuWz+CBqxayvb6DK//0Lo1t3QDsaergmrWFfPqht8n71Zs8+GYF7fZeD5d47El7cSWZmLkjF0sO8I5uOjhYZ2sTHTqAGVOiPFAi72DKRUguw5FczCQXM8ll/JC6MpNcXH0qcypfz/JnT1MHlz9cwG/fKOcTv93EttrD3HHhbBanRPGb18vJu+dNfvGfMp4oqOaNsgZ21bd5uuhuJ+3FlWRi5o5cZA6eU1vLAfoIJj3+1Db6FEIIIYSwmnnRvvz9xkV88a/vs2Z9BZ/IiOPnn55HXFggX1kOJTWHWLO+gscLqunp+3i+3vcunM1Ny2d4ruBCTGCWHODFx7vehmlvPUiHDiE9zroDPFMuQnIZjuRiJrmYSS7jh9SVmeRiFh8fz+xpEbx8ax57mztZOjN6yPOZSZN57Atn0t+vOdhuZ9/hbn6/voL73yjnExlxpMaEeKjk7iXtxZVkYuaOXKx3i2bpC6S/egV0Ng853Nt5iA6fEGJCAzxUMM9LSUnxdBG8kuRiJrmYSS5mksv4IXVlJrmYHc0lKXKSy+BuMB8fRWxYIAuSJnP3ZfMJ8PXhjn9to79/Yq7CKe3FlWRi5o5crDfAUz6oQ3vgcO2Qwzb7IfoCJqOUNVfQBCgoKPB0EbyS5GImuZhJLmaSy/ghdWUmuZidSi6xYYH84KI5FFY182xRzZDn2ibIypvSXlxJJmbuyMV6A7zwBMefrR9vKqi1JrD3MLbgSA8VSgghhBDCOj53ZhJLUiO5a90Oals6eWXrfq5Z+y7z73ydH76wjb4JemVPiLFgvQFeWKLjz0FX8OpbuwnTHQSFWXcFTYCgINng3URyMZNczCQXM8ll/JC6MpNczE41F6UUd192Bkd6+1l+7wZueaqY6oOdXDA3niff3cutTxXT3dM37P+/rfYwb+1q9NpbPKW9uJJMzNyRi9LafR8MpdQFwBrABqzVWv/qmOdvAm4B+oB2YLXWesS1QrOzs3VRUdGpF6q/H34RC7m3wKqfArCpdC/LnptPzcJvkXTpj079tYUQQowqpdQWrXW2p8sxXpx2HynEGHu2qIaN5Qe4fFEiy9JisPko1m6u5Bev7GBJaiR/ui6bsEC/gfNbu3v49as7efLdvQBkJobzg4szWJwid2EJaxmpf3TbFTyllA14CLgQyACuUkplHHPaU1rr+VrrBcCvgfvdVZ4BPj50B0QNuUVz7759AERGW3t1n8LCQk8XwStJLmaSi5nkYia5jB9SV2aSi9np5nJFdhIPXZ3FOemx2Hwc6yDceHYqv/vcAoqqWzjn3g3c+HgRD6yv4Il397Dyvo08VbiXG/JS+PXlZ9DQaueKPxbwlSeK2H+4azR+pVEh7cWVZGLmjlzcuU3CYmC31roSQCn1DHApMHCFTmvdOuj8YGBMrrN3+0cTePjjAd6+/fWOAoRb+xbNri7v+YvRm0guZpKLmeRiJrmMH1JXZpKLmbty+fTCBOLDA/nH+zVsrT3E+p0NaA1zp4bx6PVnMj8xHIBPnjGVtZsr+cOGj7hozWZ+89lMzpsT55YynQxpL64kEzN35OLOAV4CMHhppFog59iTlFK3ALcD/sC5bizPgO7AaGitHHjcdNAxwCMoYizeXgghhBBCHMeS1CiWpDq+fG+397KnqYP0uFB8bR/fgBbkb+Nr56Vx8RlTuPWpD7jh8SJuyEvhuxfMxt/XektNCAFesNG51voh4CGl1NXAD4Hrjz1HKbUaWA0wdepUNmzYAEBqaiqhoaGUlJQAEBUVxdy5c9m0aRMAvr6+5OXlUVxcTGur42JhdnY2flHJ9O/YzKa31jMjLZ22lgNgg6Kyj5hkj2PWrFnk5+cDEBAQQG5uLkVFRbS3twOQk5NDbW0tdXWOq4Dp6enYbDbKyhwXJ+Pj40lJSRlY9jQoKIicnBwKCwsHRum5ublUVVVRX+8YXGZkZNDX18euXbsASEhIIDExceCybUhICNnZ2RQUFGC32wHIy8ujvLycxsZGAObNm4fdbqeiogKApKQk4uLiODofIywsjKysLPLz8+nt7QVg2bJllJaW0tTUhNaalpYW2traqKx0DICTk5OJjIykuLgYgIiICDIzM9m4cSNaa5RSLF++nJKSElpaWgDIysqiubmZ6urq06qnhoYGamoc3xGkpaUREBDA9u3bAYiNjR2zeoqJiRloc95QTwCZmZkeryetNe3t7V5TT97yeZo8efJAe/GGegLv+DxprbHb7SddT2Ls5ebmeroIXklyMRurXEICfJk7NXzY51NjQvjXzWfxq//u5NH8Kl4rree82bEsT49hSWoUk/zH9p+80l5cSSZm7sjFbYusKKVygTu11uc7H98BoLW+e5jzfYAWrfXwn15GZwJ5/ct3Eb/l13D7Tvb0hPH7+3/Kb/z+CLeVQETyab32eLZz505mz57t6WJ4HcnFTHIxk1zMTjUXWWTl5IxGHylt2ExyMfPGXNbvaODJd/dQUNlEd08//jYfzpkdw/8tTGBFeiyBfja01uw/3E1DazfzE8KHXBUcDd6Yi6dJJmbu6B/d+XXG+0CaUioFqAOuBK4+pmBpWusK58OLgQrGwAG7P/EArXXsOqwIx/ENtdVv0ayvr5cPnoHkYia5mEkuZpLL+CF1ZSa5mHljLufNieO8OXF09/RRVN3C/3Y08J+t+3mttIHQQF+So4KpPNBOxxHHNgyLpkfwu88tICly0sBr7G5s4z9b93Pe7LiB+X4nwxtz8TTJxMwdubhtgKe17lVK3Qq8hmObhMe01qVKqZ8BRVrrl4BblVIrgR6gBcPtme7QHRjj+OFwLeUN4YSrDrTyQfmHjsXbCyGEEEIINwv0s5GXFk1eWjQ/vHgOb3/UxIsf1nGgzc5ns5OYGRuCBn796k4uXLOZn106l0XTI1izvoIXPqijX8Pv/lfBxfOncPsnZjEjRm4bF+ODW29I1lqvA9Ydc+zHg36+zZ3vP5zkM5ZCEY4reA3TOSegG+U/GXysPRk3I+PYXSwESC7DkVzMJBczyWX8kLoyk1zMxksuvjYfls+KYfmsGJfnzkmP4fZ/lHD7syUoBf42H248O5VrcqbxfHEdazdX8mppPefOjmVB0mTmJYQzPyGcyGD/Yd9vvOQyliQTM3fk4vFFVjyh1zcY/IIdV/Dq2/h8oB0CJnu6WB7X19fn6SJ4JcnFTHIxk1zMJJfxQ+rKTHIxmwi5JEZM4unVS1i7uZIDbXZWL0slNiwQgNtXzeK63Ok8vOEj/rejgTfKGgBQClafnco3P5FuXK1zIuQy2iQTM3fkYslLVrvKyyE8AX24lqqDHcT4dll+/h0wsJKdGEpyMZNczCQXM8ll/JC6MpNczCZKLjYfxVeWz+CHl2QMDO6Oig4J4EeXZLDx2+dQ8pNP8NSNOVyxKIk/bqrks4+8w56mDpfXmyi5jCbJxMwduVjyCh4AYQkcaa7hSF+/Y5GVoHhPl0gIIYRFKKUuANbgmKO+Vmv9K8M5VwB3Ahoo0Vpffew5QoixFR7kx1kzozlrZjQr0mP47vNbuWjNZi45Yyr1rd3UtHRyoNXOJFsf6ZXvkRQRxIr0WFZleH7zdWEdlhzgJSQkwOEE2F8KQHB/GwTKqj4JCQmeLoJXklzMJBczycVMcvmYUsoGPASsAmqB95VSL2mtywadkwbcASzVWrcopWLHqnxSV2aSi5mVc7lw/hTOSJrMt58r4dXSepIig0iPC+XsmdFU1TfR0nGED/e28PfCvVx5ZhI/+eRcgvxtw77e0f1QJyort5WRuCMXSw7wEhMToS4J/64D+NGLf0+r3KKJMxfhQnIxk1zMJBczyWWIxcBurXUlgFLqGeBSoGzQOV8GHtJatwBorRvHqnBSV2aSi5nVc0mYHMRTX17icryrq4ugoCB6+vr57Rvl/GHDRxTvbeHBq7OYFTd01fbivS38eVMlb5Q1cGZyJJ9eOJUL5k0hPMhvrH6NMWH1tjIcd+RiyTl4hYWFEJaAQjMz4BCq+5AM8HDmIlxILmaSi5nkYia5DJEA1Ax6XOs8NtgsYJZS6m2l1LvOWzrHhNSVmeRiJrmYHc3Fz+bDdy6Yzd++tJjmjiNc8kA+F63ZzE1PbOGuV8q4/OF3uOwP7/D27oNclpVAfWs3331+G2fe9T8uf/gdbnpiC9//9zYefLOCpna7h3+r0yNtxcwduVjyCh4A4Y6+dGnYAVSbhiBZRVMIIYTX8AXSgBVAIrBJKTVfa33o2BOVUquB1QBTp05lw4YNAKSmphIaGkpJSQkAUVFRzJ07l02bNjnewNeXvLw8iouLaW1tBSA7Oxu73T7wGmlpaQQEBLB9+3YAYmNjmTVrFvn5+QAEBASQm5tLUVER7e3tAOTk5FBbW0tdXR0A6enp2Gw2ysocFyjj4+NJSUmhoKAAgKCgIHJycigsLKSrqwuA3NxcqqqqqK+vBxzLiPf19Q0sRpCQkEBiYuLAP4xCQkLIzs6moKAAu93xj+C8vDzKy8tpbHRc/Jw3bx52u52KigoAkpKSiIuLo6ioCICwsDCysrLIz8+nt7cXgGXLllFaWkpTUxPt7e20tLTQ1tZGZWUlAMnJyURGRlJcXAxAREQEmZmZbNy4ceB2u+XLl1NSUkJLSwsAWVlZNDc3U11dfVr11NDQQE1Njcfrqbu7e6C9eEM9AWRmZnq8ntrb22lvbx9ST49ekcY/SprYvvcAJdVtrN8B8eGBXDPbn7MTfQkP7uSezyznmTfe5c2PWqlpO0xLZxBNH3VzuLuPh98q5+vLp/Op+THs2LHD8btEx2ILj6Nx91Z8lPLqz5PWmm3btnlVPYHnP0/t7e1s2LDhpOtpJEprfdyTvEl2drY+2nhOVVFREdnTw+ChM3kl+ktcfPAx+PTDsMDa89eLiorIzs72dDG8juRiJrmYSS5mp5qLUmqL1npCBaqUygXu1Fqf73x8B4DW+u5B5zwCFGqt/+J8vB74ntb6/ZFee9T6SGnDLiQXM8nF7ERy6e/XKMUJzbsrb2jj+//aRtGeFnJSIjk7LZp3PmqiaE8LR3r7mR0fyq3nzuTCeVOw+XjnPD5pK2bu6B8tOcAD6O5oJfDeJMqjzmNW03q48mmYfdEolFAIIcRomaADPF+gHDgPqAPeB67WWpcOOucC4Cqt9fVKqWjgA2CB1rpppNcerT5SCOF9+vs1/yiq4e51O2jt7mV2fChLZ0aTFBHE397dQ+WBDmbEBPOFpSmsnBPLlPAgTxdZuNFI/aMl5+AVFBSwt13RokOY2r3bcVDm4A1cNhZDSS5mkouZ5GImuXxMa90L3Aq8BuwAntValyqlfqaU+pTztNeAJqVUGfAW8O3jDe5Gi9SVmeRiJrmYuSMXHx/FVYun8c4d51H8o1W8+vVl/OiSDL6wNIU3vrGcB69eiJ/Nhx+9sJ3cu9/k4gc287v/ldPcccTltd7a2cgtTxXzzy212HtPbqPtnr7+Uyq/tBUzd+RiyTl4drudygMd9Ooo5nTsdRyUAd7AfdZiKMnFTHIxk1zMJJehtNbrgHXHHPvxoJ81cLvzvzEldWUmuZhJLmbuzCUkwBcChh6z+SguOWMqF8+fwu7Gdv63o5H1OxpYs76CR/Or+Nq5M7n+rGTaunv56ctlvFyyj0n+Nl7Zup9f/XcH1+RM54zEcBrb7DS0dtPXr/nS0hQigv0H3kNrzZr1FfxxYyWPXLuI5bNiTqrc0lbM3JGLJQd4AJUH2/HVkWSwx3FAFlkRQgghhBDjmFKKtLhQ0uJC+eqKGZQ3tPHLdTv45bqd/K1gD+32Xjrtfdy+ahZfWZ7K+1UtPPZ2FWvWVxzzOvCv4jr+eO0i5iWE09+v+cUrO3js7SqC/W187aliXrw1j5ToYA/9pmIklpyD19vbyx3/LiVnx11c3v+a4+APGsAvcBRKOH719vbi62vZMf+wJBczycVMcjE71Vwm4hw8dxqtPlLasCvJxUxyMfO2XDZXHOD+N8oJCfDlJ5+cy8zYkCHP723q5GCHnbiwQGJCAtixv5WbntxCc8cR7vq/+RRWNvHcllq+cFYyX1qawqUP5RMZ7M8LtywlNPDE9uvztky8hTv6R0vOwSsvL6fqYAdHgqc4DvhNsvzgDhy5CFeSi5nkYia5mEku44fUlZnkYia5mHlbLmenxfDvm5fyxA05LoM7gGlRk8iaFkHC5CD8fX3ITJrMy1/LY0HSZL71XAnPbanl/52Xxk8+mcG0qEn84ZpFVDd18vVnPqSv/8QuFnlbJt7CHblYcoDX2NhI1cEObJOTHAcC5fZMYGBvEzGU5GImuZhJLmaSy/ghdWUmuZhJLmYTIZfokACevDGHb6ycxd2Xzef2VbMGtnTInRHFTz6Zwfqdjaz67UYuXLOZC363iUsfepu7/7uD96qa6e3r51DnEf5euIcrHingM09V8Yv/lLks+tJ1pI/djW2M5l2FvX39VB3sGLXXcyd3tBVLXift6NE0dRwhKHoa1CALrAghhBBCCHEMP5sPt61MMz537ZLpdB3p472qZpRS+Cg43NXDo5ur+OPGSsICfenq6aOnTzMjJpiMKBuPvV3FM+/XcENeCjGhAby5s5G3dx/E3ttPWmwI1+RM4/+yEgkPOrHbPk201nzzuRJe/HAf/7r5LLKmWe/f+ZYc4AXHpwClRExJdewsJAusADBv3jxPF8ErSS5mkouZ5GImuYwfUldmkouZ5GJmhVyUUnxl+Qy+snzGkONt3T3kVxxkY/kBwoL8+FTmVOZODaOpqYmWvgDuf6N8YFGXpMggrlo8jZToYP71QR13vlzGPa/u4uYVM7j13JlDNoHvOtLHL9ftYO7UMK5cPG3Ycq3dXMWLH+7D5qP47RvlPHFDjnsCGCXuaCuWHOBVNrYBMCUpFVByBc9Jlq81k1zMJBczycVMchk/pK7MJBczycXMyrmEBvpx4fwpXDh/ypDjdrudtIRoHv78InY7/y0+IyZkYBB3/VnJbKs9zMMbd3PfG+Xsbe7kl5fNx8/mQ1O7nRv/VsQHew8B4O/rw2VZiS7vvbniAHf/dwcXzY9nQdJkfrluJ4WVTeSkRrn5tz517mgrlpyD92HlfnwUTIuZDBHJEJbg6SJ5hYqKiuOfZEGSi5nkYia5mEku44fUlZnkYia5mEkurgZnMjM2lJmxoUOu0AHMTwznoauzuO28NJ7bUsuNjxdRuu8wlz38DmX7WnngqoWcNSOK7/xzK2/tHDp3bW9TJ197+gPSYkO59/JMrl2STExoAPe9UX7C8/tqWzp5rqiGvU2dp/8LnyB3tBVLXsGr7+gnKXIS/r4+cP3LEBjm6SIJIYQQQghheUopvrFqFlPCA/nBC9vZ+MABIib58dSXl7BoegTnpMdw1Z/f5at/38Kj159Jd08fG3Yd4NXSerSGP123iOAAxxDnlhUzuPPlMt7e3UReWjQAe5o6eLeyiaUzo0mMmAQ4FmX5y9vV3P9GOV09fQCkx4WyMiOWa5ckEx8+vlbbt+QAr7nH9+ONGY+upClISpIsTCQXM8nFTHIxk1zGD6krM8nFTHIxk1xcnWwmVy6eRlxYIH8v3MMPLs4Y+Ld7aKAff/3iYi5/+B2uWVsIwCR/G0tnRvPVFTOYHvXx5utX5UzjT5sque+NXWQmhfPgW7v5S341R/r6AciaNplVGfG8VLKPHftbWTknlq+umMkHe1v4344GHtlYycsl+3n+q2cRExowbFmb2u1s39fKsrRol6uSo53LibDcAE9rTV1rD2fPDj7+yRYTFxfn6SJ4JcnFTHIxk1zMJJfxQ+rKTHIxk1zMJBdXp5LJObNjOWd2rMvxo1s4PFtUy+LkSM5MiSDA1+ZyXoCvjVvPTeP7/97GWb96k7buXi5flMi1S6bz9kcHeblkP/e8upP4sEAe+fwizp8bh1KKRdMjuPHsVD7Y28LVfy7kC395j2dWLzFu6v7q9np+8O9tNHUc4Zz0GO7bWckaAAANBElEQVS5/AxiQ0/8ip872orl5uA1ttnp6uknNVoGeMcqKirydBG8kuRiJrmYSS5mksv4IXVlJrmYSS5mkour0c4kMWISt6+aRV5atHFwd9RnsxPJmBLGnPgwXrp1Kb/5bCaZSZO5ecVM/nvb2eR/9xze+tYKLpgX73L1beG0CB7+fBa76tv4yhNbsPf2DTx3uLOHb/zjQ256cgvx4YF8c9Us3vmoifN/u4lXt+8HoKevn9buHjrsvcOWzx1txXJX8CoPODY9TIkO8XBJhBBCCCGEEO7kZ/Nh3W1nD/v80Xl4w1mRHsuvLz+D258t4dq17xEW5Et1Uyd7mzvp79d8fWUat5wzEz+bDxfOn+Ic9BXj66Po7f94cZeZsSEsmhZB1vTJ5KZGMy1q5Pc9HdYb4B1sByAlRq7gHSssTBabMZFczCQXM8nFTHIZP6SuzCQXM8nFTHJxNZ4zuSwrkUOdPTz01m5iQgNIjQ7mnPQYLl2QwLyE8IHzZsaG8K+bz+Kpwr00tnUT5Gcj0M9G55E+Pqw5xGtl9fyjqIavLEvljovmAO7JRZ3osqHeIjs7W5/Opcy27h7KG9pZmDQZH5+TmwQphBBibCmltmitsz1djvHidPtIIYQQ7qO1pupgB/6+Pse9cng8I/WPlpuDFxroR1dNqQzuDPLz8z1dBK8kuZhJLmaSi5nkMn5IXZlJLmaSi5nk4koycWwBkRoTMmRw545cLDfAA+jtHX6io5VJLmaSi5nkYia5mEku44fUlZnkYia5mEkuriQTM3fkYskBnhBCCCGEEEJMRG6dg6eUugBYA9iAtVrrXx3z/O3AjUAvcAD4ktZ6z0ivORrzC/r7+/HxkbHtsSQXM8nFTHIxk1zMTjUXmYN3cqSPdB/JxUxyMZNcXEkmZu7oH92WslLKBjwEXAhkAFcppTKOOe0DIFtrfQbwT+DX7irPYKWlpWPxNuOO5GImuZhJLmaSi5nkMn5IXZlJLmaSi5nk4koyMXNHLu4cRi8GdmutK7XWR4BngEsHn6C1fktr3el8+C6Q6MbyDGhqahqLtxl3JBczycVMcjGTXMwkl/FD6spMcjGTXMwkF1eSiZk7cnHnAC8BqBn0uNZ5bDg3AP91Y3mEEEIIIYQQYkLzio3OlVKfB7KB5cM8vxpYDTB16lQ2bNgAQGpqKqGhoZSUlAAQFRXF3Llz2bRpEwC+vr7k5eVRXFxMa2srANnZ2URHRw+8RlpaGgEBAWzfvh2A2NhYZs2aNbBkaUBAALm5uRQVFdHe7tgkPScnh9raWurq6gBIT0/HZrNRVlYGQHx8PCkpKRQUFAAQFBRETk4OhYWFdHV1AZCbm0tVVRX19fUAZGRk0NfXx65duwBISEggMTGRwsJCAEJCQsjOzqagoAC73Q5AXl4e5eXlNDY2AjBv3jzsdjsVFRUAJCUlERcXx9H5GGFhYWRlZZGfnz+wYs+yZcsoLS2lqamJvr4+WlpaaGtro7KyEoDk5GQiIyMpLi4GICIigszMTDZu3IjWGqUUy5cvp6SkhJaWFgCysrJobm6murr6tOqpoaGBmpoaj9fT5MmTB9qLN9QTQGZmpsfrqa+vj/b2dq+pJ2/5PIWEhAy0F2+oJ/COz1NfXx92u/2k60mMvczMTE8XwStJLmaSi5nk4koyMXNHLm5bZEUplQvcqbU+3/n4DgCt9d3HnLcS+D2wXGvdeLzXHY0J5Hv37mXatGmn9RoTkeRiJrmYSS5mkovZqeYii6ycHOkj3UdyMZNczCQXV5KJmTv6R3feovk+kKaUSlFK+QNXAi8dU7CFwB+BT53I4G60HP02XQwluZhJLmaSi5nkYia5jB9SV2aSi5nkYia5uJJMzNyRi9sGeFrrXuBW4DVgB/Cs1rpUKfUzpdSnnKfdC4QAzymlPlRKvTTMywkhhBBCCCGEOA637oPnDkqpA8CIe+WdgGjg4CgUZ6KRXMwkFzPJxUxyMTvVXKZrrWNGuzATlfSRbiW5mEkuZpKLK8nEbNT7x3E3wBsNSqkimdPhSnIxk1zMJBczycVMchk/pK7MJBczycVMcnElmZi5IxfZTl4IIYQQQgghJggZ4AkhhBBCCCHEBGHVAd6fPF0ALyW5mEkuZpKLmeRiJrmMH1JXZpKLmeRiJrm4kkzMRj0XS87BE0IIIYQQQoiJyKpX8IQQQgghhBBiwrHcAE8pdYFSapdSardS6nueLo+nKKWSlFJvKaXKlFKlSqnbnMcjlVJvKKUqnH9GeLqsY00pZVNKfaCU+o/zcYpSqtDZZv6hlPL3dBnHmlJqslLqn0qpnUqpHUqpXGkroJT6hvPzs10p9bRSKtCK7UUp9ZhSqlEptX3QMWP7UA4POPPZqpTK8lzJxWDSPzpI/zgy6SNdSR9pJn2kgyf6SEsN8JRSNuAh4EIgA7hKKZXh2VJ5TC/wTa11BrAEuMWZxfeA9VrrNGC987HV3AbsGPT4HuC3WuuZQAtwg0dK5VlrgFe11rOBTBz5WLqtKKUSgP8HZGut5wE24Eqs2V7+ClxwzLHh2seFQJrzv9XAw2NURjEC6R+HkP5xZNJHupI+8hjSRw7xV8a4j7TUAA9YDOzWWldqrY8AzwCXerhMHqG13q+1Lnb+3IbjL6MEHHk87jztceDTnimhZyilEoGLgbXOxwo4F/in8xQrZhIOLAMeBdBaH9FaH8LibcXJFwhSSvkCk4D9WLC9aK03Ac3HHB6ufVwK/E07vAtMVkpNGZuSihFI/+gk/ePwpI90JX3kiKSPxDN9pNUGeAlAzaDHtc5jlqaUSgYWAoVAnNZ6v/OpeiDOQ8XylN8B3wH6nY+jgENa617nYyu2mRTgAPAX5205a5VSwVi8rWit64DfAHtxdFqHgS1IezlquPYhfw97J6kXA+kfXUgf6Ur6SAPpI4/LrX2k1QZ44hhKqRDgeeDrWuvWwc9pxxKrlllmVSl1CdCotd7i6bJ4GV8gC3hYa70Q6OCYW02s1lYAnPfLX4qjc58KBON6C4bAmu1DjH/SPw4lfeSwpI80kD7yxLmjfVhtgFcHJA16nOg8ZklKKT8cndfftdb/ch5uOHop2Plno6fK5wFLgU8ppapx3J50Lo776ic7by8Aa7aZWqBWa13ofPxPHJ2ZldsKwEqgSmt9QGvdA/wLRxuyens5arj2IX8Peyepl0GkfzSSPtJM+kgz6SNH5tY+0moDvPeBNOcKPv44Jnu+5OEyeYTzvvlHgR1a6/sHPfUScL3z5+uBF8e6bJ6itb5Da52otU7G0Tbe1FpfA7wFXO48zVKZAGit64EapVS689B5QBkWbitOe4ElSqlJzs/T0Vws3V4GGa59vARc51wpbAlweNBtKsJzpH90kv7RTPpIM+kjhyV95Mjc2kdabqNzpdRFOO4htwGPaa3v8nCRPEIplQdsBrbx8b3038cxz+BZYBqwB7hCa33sxNAJTym1AviW1voSpVQqjm8rI4EPgM9rre2eLN9YU0otwDGp3h+oBL6I4wsiS7cVpdRPgc/hWHXvA+BGHPfKW6q9KKWeBlYA0UAD8BPgBQztw9nRP4jjVp1O4Ita6yJPlFsMJf2jg/SPxyd95FDSR5pJH+ngiT7ScgM8IYQQQgghhJiorHaLphBCCCGEEEJMWDLAE0IIIYQQQogJQgZ4QgghhBBCCDFByABPCCGEEEIIISYIGeAJIYQQQgghxAQhAzwhxpBSqk8p9eGg/743iq+drJTaPlqvJ4QQQowl6SOFGB2+xz9FCDGKurTWCzxdCCGEEMILSR8pxCiQK3hCeAGlVLVS6tdKqW1KqfeUUjOdx5OVUm8qpbYqpdYrpaY5j8cppf6tlCpx/neW86VsSqk/K6VKlVKvK6WCPPZLCSGEEKNA+kghTo4M8IQYW0HH3H7yuUHPHdZazwceBH7nPPZ74HGt9RnA34EHnMcfADZqrTOBLKDUeTwNeEhrPRc4BHzGzb+PEEIIMVqkjxRiFCittafLIIRlKKXatdYhhuPVwLla60qllB9Qr7WOUkodBKZorXucx/drraOVUgeARK21fdBrJANvaK3TnI+/C/hprX/h/t9MCCGEOD3SRwoxOuQKnhDeQw/z88mwD/q5D5lnK4QQYmKQPlKIEyQDPCG8x+cG/Vng/Pkd4Ernz9cAm50/rwe+CqCUsimlwseqkEIIIYQHSB8pxAmSby6EGFtBSqkPBz1+VWt9dBnoCKXUVhzfMF7lPPY14C9KqW8DB4AvOo/fBvxJKXUDjm8hvwrsd3vphRBCCPeRPlKIUSBz8ITwAs75Bdla64OeLosQQgjhTaSPFOLkyC2aQgghhBBCCDFByBU8IYQQQgghhJgg5AqeEEIIIYQQQkwQMsATQgghhBBCiAlCBnhCCCGEEEIIMUHIAE8IIYQQQgghJggZ4AkhhBBCCCHEBCEDPCGEEEIIIYSYIP4/PF3HcBTMPagAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 70.0477 %\n",
            "\tMin validation loss: 0.96175\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gXe9zmvDXoL2"
      },
      "source": [
        "### fold-8 <a name=\"fold-8\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8ktJhhvRXnzn",
        "outputId": "1885798c-96de-47fe-ad62-931518e27245"
      },
      "source": [
        "FOLD_K = 8\n",
        "REPEAT = 1\n",
        "\n",
        "history8 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history8.append(history)"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 9.4293%\n",
            "\n",
            "Epoch 1/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 2.1113 - accuracy: 0.2194 - val_loss: 1.9106 - val_accuracy: 0.2283\n",
            "Epoch 2/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 1.8874 - accuracy: 0.2941 - val_loss: 1.6618 - val_accuracy: 0.3809\n",
            "Epoch 3/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.7201 - accuracy: 0.3735 - val_loss: 1.6236 - val_accuracy: 0.4107\n",
            "Epoch 4/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.6126 - accuracy: 0.4027 - val_loss: 1.4225 - val_accuracy: 0.4913\n",
            "Epoch 5/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.4911 - accuracy: 0.4599 - val_loss: 1.2990 - val_accuracy: 0.4938\n",
            "Epoch 6/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.4066 - accuracy: 0.4984 - val_loss: 1.4598 - val_accuracy: 0.5037\n",
            "Epoch 7/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.3553 - accuracy: 0.5218 - val_loss: 1.2884 - val_accuracy: 0.5744\n",
            "Epoch 8/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.3206 - accuracy: 0.5262 - val_loss: 1.2003 - val_accuracy: 0.5906\n",
            "Epoch 9/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.2512 - accuracy: 0.5619 - val_loss: 1.2778 - val_accuracy: 0.5323\n",
            "Epoch 10/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.2226 - accuracy: 0.5747 - val_loss: 1.2553 - val_accuracy: 0.5385\n",
            "Epoch 11/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.1908 - accuracy: 0.5858 - val_loss: 1.3356 - val_accuracy: 0.5136\n",
            "Epoch 12/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.1691 - accuracy: 0.5872 - val_loss: 1.2009 - val_accuracy: 0.5434\n",
            "Epoch 13/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.1299 - accuracy: 0.6132 - val_loss: 1.1913 - val_accuracy: 0.5583\n",
            "Epoch 14/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.1134 - accuracy: 0.6136 - val_loss: 1.2378 - val_accuracy: 0.5794\n",
            "Epoch 15/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.0806 - accuracy: 0.6276 - val_loss: 1.2277 - val_accuracy: 0.6005\n",
            "Epoch 16/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.0398 - accuracy: 0.6472 - val_loss: 1.1208 - val_accuracy: 0.5546\n",
            "Epoch 17/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.0364 - accuracy: 0.6520 - val_loss: 1.0986 - val_accuracy: 0.5608\n",
            "Epoch 18/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.9975 - accuracy: 0.6652 - val_loss: 1.2090 - val_accuracy: 0.5434\n",
            "Epoch 19/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.9903 - accuracy: 0.6664 - val_loss: 1.0959 - val_accuracy: 0.5955\n",
            "Epoch 20/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.9434 - accuracy: 0.6863 - val_loss: 1.1491 - val_accuracy: 0.5868\n",
            "Epoch 21/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.9300 - accuracy: 0.6853 - val_loss: 1.2868 - val_accuracy: 0.5707\n",
            "Epoch 22/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.9143 - accuracy: 0.6964 - val_loss: 1.1406 - val_accuracy: 0.6228\n",
            "Epoch 23/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.9254 - accuracy: 0.6972 - val_loss: 1.1633 - val_accuracy: 0.6005\n",
            "Epoch 24/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8756 - accuracy: 0.7131 - val_loss: 1.0419 - val_accuracy: 0.6216\n",
            "Epoch 25/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8820 - accuracy: 0.7149 - val_loss: 1.1487 - val_accuracy: 0.6328\n",
            "Epoch 26/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8404 - accuracy: 0.7345 - val_loss: 1.1061 - val_accuracy: 0.6253\n",
            "Epoch 27/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.8421 - accuracy: 0.7300 - val_loss: 1.0521 - val_accuracy: 0.6712\n",
            "Epoch 28/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.8318 - accuracy: 0.7361 - val_loss: 1.1162 - val_accuracy: 0.6191\n",
            "Epoch 29/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.8017 - accuracy: 0.7515 - val_loss: 1.1249 - val_accuracy: 0.6501\n",
            "Epoch 30/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.8183 - accuracy: 0.7487 - val_loss: 1.0862 - val_accuracy: 0.6427\n",
            "Epoch 31/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7861 - accuracy: 0.7530 - val_loss: 1.1048 - val_accuracy: 0.6439\n",
            "Epoch 32/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7892 - accuracy: 0.7545 - val_loss: 1.1406 - val_accuracy: 0.6042\n",
            "Epoch 33/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7635 - accuracy: 0.7620 - val_loss: 1.0961 - val_accuracy: 0.6886\n",
            "Epoch 34/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7729 - accuracy: 0.7595 - val_loss: 1.1157 - val_accuracy: 0.6638\n",
            "Epoch 35/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7529 - accuracy: 0.7667 - val_loss: 1.0665 - val_accuracy: 0.6960\n",
            "Epoch 36/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7328 - accuracy: 0.7748 - val_loss: 1.0427 - val_accuracy: 0.6898\n",
            "Epoch 37/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7523 - accuracy: 0.7686 - val_loss: 1.1950 - val_accuracy: 0.6290\n",
            "Epoch 38/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7409 - accuracy: 0.7748 - val_loss: 1.0398 - val_accuracy: 0.6935\n",
            "Epoch 39/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7289 - accuracy: 0.7773 - val_loss: 1.2425 - val_accuracy: 0.6079\n",
            "Epoch 40/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7043 - accuracy: 0.7906 - val_loss: 1.2485 - val_accuracy: 0.6179\n",
            "Epoch 41/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7154 - accuracy: 0.7844 - val_loss: 1.2104 - val_accuracy: 0.6141\n",
            "Epoch 42/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6830 - accuracy: 0.7916 - val_loss: 1.2166 - val_accuracy: 0.6166\n",
            "Epoch 43/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6763 - accuracy: 0.8014 - val_loss: 1.2816 - val_accuracy: 0.6141\n",
            "Epoch 44/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6705 - accuracy: 0.7998 - val_loss: 1.1400 - val_accuracy: 0.6514\n",
            "Epoch 45/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6703 - accuracy: 0.7990 - val_loss: 1.1460 - val_accuracy: 0.6762\n",
            "Epoch 46/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6557 - accuracy: 0.8038 - val_loss: 1.1390 - val_accuracy: 0.6886\n",
            "Epoch 47/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6527 - accuracy: 0.8067 - val_loss: 1.1946 - val_accuracy: 0.6737\n",
            "Epoch 48/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6510 - accuracy: 0.8068 - val_loss: 1.2277 - val_accuracy: 0.6650\n",
            "Epoch 49/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6447 - accuracy: 0.8118 - val_loss: 1.3630 - val_accuracy: 0.6427\n",
            "Epoch 50/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6430 - accuracy: 0.8039 - val_loss: 1.3913 - val_accuracy: 0.6104\n",
            "Epoch 51/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6398 - accuracy: 0.8097 - val_loss: 1.1778 - val_accuracy: 0.6849\n",
            "Epoch 52/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6233 - accuracy: 0.8142 - val_loss: 1.2431 - val_accuracy: 0.6588\n",
            "Epoch 53/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.6188 - accuracy: 0.8176 - val_loss: 1.2213 - val_accuracy: 0.6998\n",
            "Epoch 54/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6283 - accuracy: 0.8116 - val_loss: 1.2762 - val_accuracy: 0.6886\n",
            "Epoch 55/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6016 - accuracy: 0.8242 - val_loss: 1.2309 - val_accuracy: 0.6191\n",
            "Epoch 56/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6029 - accuracy: 0.8301 - val_loss: 1.2320 - val_accuracy: 0.6787\n",
            "Epoch 57/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6132 - accuracy: 0.8190 - val_loss: 1.3915 - val_accuracy: 0.6737\n",
            "Epoch 58/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5967 - accuracy: 0.8237 - val_loss: 1.3240 - val_accuracy: 0.6452\n",
            "Epoch 59/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5750 - accuracy: 0.8327 - val_loss: 1.3111 - val_accuracy: 0.6526\n",
            "Epoch 60/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5879 - accuracy: 0.8313 - val_loss: 1.2669 - val_accuracy: 0.6849\n",
            "Epoch 61/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5778 - accuracy: 0.8308 - val_loss: 1.4239 - val_accuracy: 0.6303\n",
            "Epoch 62/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5634 - accuracy: 0.8323 - val_loss: 1.2104 - val_accuracy: 0.6861\n",
            "Epoch 63/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5756 - accuracy: 0.8325 - val_loss: 1.2582 - val_accuracy: 0.6998\n",
            "Epoch 64/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5556 - accuracy: 0.8388 - val_loss: 1.3058 - val_accuracy: 0.6725\n",
            "Epoch 65/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.5564 - accuracy: 0.8375 - val_loss: 1.2483 - val_accuracy: 0.7208\n",
            "Epoch 66/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5570 - accuracy: 0.8389 - val_loss: 1.2910 - val_accuracy: 0.6687\n",
            "Epoch 67/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5378 - accuracy: 0.8434 - val_loss: 1.4390 - val_accuracy: 0.6303\n",
            "Epoch 68/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5351 - accuracy: 0.8423 - val_loss: 1.3479 - val_accuracy: 0.6824\n",
            "Epoch 69/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5383 - accuracy: 0.8457 - val_loss: 1.2037 - val_accuracy: 0.7146\n",
            "Epoch 70/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5379 - accuracy: 0.8483 - val_loss: 1.4548 - val_accuracy: 0.6873\n",
            "Epoch 71/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5247 - accuracy: 0.8480 - val_loss: 1.1756 - val_accuracy: 0.7035\n",
            "Epoch 72/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5128 - accuracy: 0.8552 - val_loss: 1.4238 - val_accuracy: 0.6675\n",
            "Epoch 73/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5211 - accuracy: 0.8482 - val_loss: 1.2999 - val_accuracy: 0.6985\n",
            "Epoch 74/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5113 - accuracy: 0.8534 - val_loss: 1.3225 - val_accuracy: 0.6935\n",
            "Epoch 75/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5071 - accuracy: 0.8574 - val_loss: 1.3683 - val_accuracy: 0.6849\n",
            "Epoch 76/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4981 - accuracy: 0.8588 - val_loss: 1.3264 - val_accuracy: 0.7184\n",
            "Epoch 77/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.5181 - accuracy: 0.8511 - val_loss: 1.4874 - val_accuracy: 0.6489\n",
            "Epoch 78/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.5140 - accuracy: 0.8542 - val_loss: 1.5246 - val_accuracy: 0.7159\n",
            "Epoch 79/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4864 - accuracy: 0.8617 - val_loss: 1.4082 - val_accuracy: 0.6650\n",
            "Epoch 80/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4885 - accuracy: 0.8640 - val_loss: 1.4246 - val_accuracy: 0.6464\n",
            "Epoch 81/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4928 - accuracy: 0.8593 - val_loss: 1.4034 - val_accuracy: 0.7146\n",
            "Epoch 82/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.4874 - accuracy: 0.8586 - val_loss: 1.4919 - val_accuracy: 0.6725\n",
            "Epoch 83/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4836 - accuracy: 0.8598 - val_loss: 1.3893 - val_accuracy: 0.7035\n",
            "Epoch 84/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4668 - accuracy: 0.8625 - val_loss: 1.4463 - val_accuracy: 0.6613\n",
            "Epoch 85/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4640 - accuracy: 0.8707 - val_loss: 1.3544 - val_accuracy: 0.7097\n",
            "Epoch 86/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4783 - accuracy: 0.8666 - val_loss: 1.4496 - val_accuracy: 0.6700\n",
            "Epoch 87/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4718 - accuracy: 0.8646 - val_loss: 1.3945 - val_accuracy: 0.6911\n",
            "Epoch 88/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4589 - accuracy: 0.8741 - val_loss: 1.4290 - val_accuracy: 0.6985\n",
            "Epoch 89/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4470 - accuracy: 0.8748 - val_loss: 1.6091 - val_accuracy: 0.6166\n",
            "Epoch 90/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4721 - accuracy: 0.8660 - val_loss: 1.4619 - val_accuracy: 0.7022\n",
            "Epoch 91/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4503 - accuracy: 0.8722 - val_loss: 1.6722 - val_accuracy: 0.6414\n",
            "Epoch 92/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4461 - accuracy: 0.8738 - val_loss: 1.5055 - val_accuracy: 0.6588\n",
            "Epoch 93/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4417 - accuracy: 0.8805 - val_loss: 1.5006 - val_accuracy: 0.7072\n",
            "Epoch 94/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4381 - accuracy: 0.8745 - val_loss: 1.4045 - val_accuracy: 0.7419\n",
            "Epoch 95/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4334 - accuracy: 0.8800 - val_loss: 1.6072 - val_accuracy: 0.6538\n",
            "Epoch 96/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4473 - accuracy: 0.8733 - val_loss: 1.6961 - val_accuracy: 0.6650\n",
            "Epoch 97/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4449 - accuracy: 0.8756 - val_loss: 1.6186 - val_accuracy: 0.6538\n",
            "Epoch 98/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4201 - accuracy: 0.8830 - val_loss: 1.6151 - val_accuracy: 0.6861\n",
            "Epoch 99/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4403 - accuracy: 0.8756 - val_loss: 1.4751 - val_accuracy: 0.6613\n",
            "Epoch 100/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4156 - accuracy: 0.8818 - val_loss: 1.6345 - val_accuracy: 0.6737\n",
            "Training completed in time:  0:16:56.051185 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "id": "9DAiZu0KXvmv",
        "outputId": "0f50a78a-ceff-4c5a-9222-fde5291bb5a9"
      },
      "source": [
        "show_results(history8)"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 74.1935 %\n",
            "\tMin validation loss: 1.03979\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pNnepO-aX0aM"
      },
      "source": [
        "### fold-9 <a name=\"fold-9\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RzxaMxsWXyAQ",
        "outputId": "e077cbfc-b125-456d-dcbe-6b71e98af4d3"
      },
      "source": [
        "FOLD_K = 9\n",
        "REPEAT = 1\n",
        "\n",
        "history9 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history9.append(history)"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 5.6373%\n",
            "\n",
            "Epoch 1/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 2.0925 - accuracy: 0.2213 - val_loss: 1.8308 - val_accuracy: 0.2843\n",
            "Epoch 2/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.9048 - accuracy: 0.2740 - val_loss: 1.6679 - val_accuracy: 0.3529\n",
            "Epoch 3/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 1.7443 - accuracy: 0.3517 - val_loss: 1.6095 - val_accuracy: 0.3615\n",
            "Epoch 4/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.5884 - accuracy: 0.4279 - val_loss: 1.5297 - val_accuracy: 0.4436\n",
            "Epoch 5/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.4769 - accuracy: 0.4698 - val_loss: 1.4539 - val_accuracy: 0.4473\n",
            "Epoch 6/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.3970 - accuracy: 0.5035 - val_loss: 1.3238 - val_accuracy: 0.4926\n",
            "Epoch 7/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.3336 - accuracy: 0.5251 - val_loss: 1.3504 - val_accuracy: 0.4939\n",
            "Epoch 8/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.2978 - accuracy: 0.5430 - val_loss: 1.2320 - val_accuracy: 0.5196\n",
            "Epoch 9/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.2527 - accuracy: 0.5475 - val_loss: 1.2808 - val_accuracy: 0.5588\n",
            "Epoch 10/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.2123 - accuracy: 0.5718 - val_loss: 1.3063 - val_accuracy: 0.5392\n",
            "Epoch 11/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.1905 - accuracy: 0.5859 - val_loss: 1.1833 - val_accuracy: 0.5858\n",
            "Epoch 12/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.1451 - accuracy: 0.6026 - val_loss: 1.1715 - val_accuracy: 0.5870\n",
            "Epoch 13/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 1.1339 - accuracy: 0.6070 - val_loss: 1.1050 - val_accuracy: 0.6225\n",
            "Epoch 14/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.1000 - accuracy: 0.6156 - val_loss: 1.0768 - val_accuracy: 0.6409\n",
            "Epoch 15/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.1009 - accuracy: 0.6184 - val_loss: 1.0620 - val_accuracy: 0.6336\n",
            "Epoch 16/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.0586 - accuracy: 0.6321 - val_loss: 1.1608 - val_accuracy: 0.6238\n",
            "Epoch 17/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 1.0345 - accuracy: 0.6478 - val_loss: 1.1365 - val_accuracy: 0.6152\n",
            "Epoch 18/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 1.0289 - accuracy: 0.6467 - val_loss: 1.0092 - val_accuracy: 0.6789\n",
            "Epoch 19/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.9934 - accuracy: 0.6608 - val_loss: 1.0103 - val_accuracy: 0.7181\n",
            "Epoch 20/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.9778 - accuracy: 0.6630 - val_loss: 1.0293 - val_accuracy: 0.6789\n",
            "Epoch 21/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.9745 - accuracy: 0.6732 - val_loss: 1.0139 - val_accuracy: 0.7022\n",
            "Epoch 22/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.9576 - accuracy: 0.6761 - val_loss: 1.0823 - val_accuracy: 0.6409\n",
            "Epoch 23/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.9343 - accuracy: 0.6921 - val_loss: 1.0143 - val_accuracy: 0.6716\n",
            "Epoch 24/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.9161 - accuracy: 0.6939 - val_loss: 1.0051 - val_accuracy: 0.6850\n",
            "Epoch 25/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.8937 - accuracy: 0.7089 - val_loss: 1.0295 - val_accuracy: 0.6961\n",
            "Epoch 26/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.8917 - accuracy: 0.7070 - val_loss: 0.8841 - val_accuracy: 0.7267\n",
            "Epoch 27/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.8700 - accuracy: 0.7245 - val_loss: 0.9133 - val_accuracy: 0.7255\n",
            "Epoch 28/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.8532 - accuracy: 0.7240 - val_loss: 0.9609 - val_accuracy: 0.7169\n",
            "Epoch 29/100\n",
            "248/247 [==============================] - 11s 43ms/step - loss: 0.8397 - accuracy: 0.7247 - val_loss: 0.9875 - val_accuracy: 0.7255\n",
            "Epoch 30/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.8315 - accuracy: 0.7356 - val_loss: 1.0076 - val_accuracy: 0.7157\n",
            "Epoch 31/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8182 - accuracy: 0.7396 - val_loss: 0.9437 - val_accuracy: 0.7243\n",
            "Epoch 32/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.8046 - accuracy: 0.7361 - val_loss: 0.8651 - val_accuracy: 0.7304\n",
            "Epoch 33/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7942 - accuracy: 0.7515 - val_loss: 0.9322 - val_accuracy: 0.7194\n",
            "Epoch 34/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7980 - accuracy: 0.7476 - val_loss: 0.9044 - val_accuracy: 0.7230\n",
            "Epoch 35/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7729 - accuracy: 0.7600 - val_loss: 0.9867 - val_accuracy: 0.7230\n",
            "Epoch 36/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.7871 - accuracy: 0.7567 - val_loss: 0.8520 - val_accuracy: 0.7439\n",
            "Epoch 37/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7576 - accuracy: 0.7689 - val_loss: 0.8630 - val_accuracy: 0.7328\n",
            "Epoch 38/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7636 - accuracy: 0.7644 - val_loss: 0.8870 - val_accuracy: 0.7353\n",
            "Epoch 39/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7518 - accuracy: 0.7725 - val_loss: 0.9471 - val_accuracy: 0.7292\n",
            "Epoch 40/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7325 - accuracy: 0.7807 - val_loss: 0.8328 - val_accuracy: 0.7537\n",
            "Epoch 41/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7046 - accuracy: 0.7868 - val_loss: 1.0144 - val_accuracy: 0.7402\n",
            "Epoch 42/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7274 - accuracy: 0.7743 - val_loss: 0.9827 - val_accuracy: 0.7279\n",
            "Epoch 43/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.7087 - accuracy: 0.7847 - val_loss: 0.8838 - val_accuracy: 0.7292\n",
            "Epoch 44/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.6921 - accuracy: 0.7894 - val_loss: 0.9357 - val_accuracy: 0.7463\n",
            "Epoch 45/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6826 - accuracy: 0.7961 - val_loss: 1.0181 - val_accuracy: 0.7132\n",
            "Epoch 46/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.6740 - accuracy: 0.7972 - val_loss: 0.9240 - val_accuracy: 0.7304\n",
            "Epoch 47/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6692 - accuracy: 0.7972 - val_loss: 1.0930 - val_accuracy: 0.7194\n",
            "Epoch 48/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6633 - accuracy: 0.7969 - val_loss: 1.0118 - val_accuracy: 0.7341\n",
            "Epoch 49/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.6718 - accuracy: 0.8068 - val_loss: 0.9472 - val_accuracy: 0.7512\n",
            "Epoch 50/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.6599 - accuracy: 0.8039 - val_loss: 1.0476 - val_accuracy: 0.7500\n",
            "Epoch 51/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6436 - accuracy: 0.8109 - val_loss: 1.0476 - val_accuracy: 0.7316\n",
            "Epoch 52/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.6139 - accuracy: 0.8159 - val_loss: 1.0148 - val_accuracy: 0.7328\n",
            "Epoch 53/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6278 - accuracy: 0.8108 - val_loss: 1.0125 - val_accuracy: 0.7206\n",
            "Epoch 54/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6177 - accuracy: 0.8171 - val_loss: 1.0209 - val_accuracy: 0.7194\n",
            "Epoch 55/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6159 - accuracy: 0.8166 - val_loss: 1.0242 - val_accuracy: 0.7255\n",
            "Epoch 56/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.6137 - accuracy: 0.8253 - val_loss: 1.1463 - val_accuracy: 0.7181\n",
            "Epoch 57/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6157 - accuracy: 0.8192 - val_loss: 1.2246 - val_accuracy: 0.7230\n",
            "Epoch 58/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.6124 - accuracy: 0.8191 - val_loss: 1.0485 - val_accuracy: 0.7341\n",
            "Epoch 59/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5830 - accuracy: 0.8302 - val_loss: 0.9627 - val_accuracy: 0.7500\n",
            "Epoch 60/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5714 - accuracy: 0.8322 - val_loss: 1.0811 - val_accuracy: 0.7316\n",
            "Epoch 61/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5801 - accuracy: 0.8322 - val_loss: 1.0088 - val_accuracy: 0.7439\n",
            "Epoch 62/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5813 - accuracy: 0.8254 - val_loss: 0.9690 - val_accuracy: 0.7426\n",
            "Epoch 63/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5746 - accuracy: 0.8306 - val_loss: 0.9835 - val_accuracy: 0.7365\n",
            "Epoch 64/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5677 - accuracy: 0.8358 - val_loss: 1.0997 - val_accuracy: 0.7304\n",
            "Epoch 65/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5694 - accuracy: 0.8319 - val_loss: 1.0794 - val_accuracy: 0.7439\n",
            "Epoch 66/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5564 - accuracy: 0.8399 - val_loss: 1.1398 - val_accuracy: 0.7365\n",
            "Epoch 67/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5404 - accuracy: 0.8439 - val_loss: 1.1221 - val_accuracy: 0.7426\n",
            "Epoch 68/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5603 - accuracy: 0.8359 - val_loss: 1.0902 - val_accuracy: 0.7353\n",
            "Epoch 69/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5404 - accuracy: 0.8449 - val_loss: 1.1965 - val_accuracy: 0.7402\n",
            "Epoch 70/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5312 - accuracy: 0.8434 - val_loss: 1.1768 - val_accuracy: 0.7390\n",
            "Epoch 71/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5495 - accuracy: 0.8389 - val_loss: 1.1799 - val_accuracy: 0.7328\n",
            "Epoch 72/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5292 - accuracy: 0.8507 - val_loss: 1.1271 - val_accuracy: 0.7439\n",
            "Epoch 73/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5346 - accuracy: 0.8458 - val_loss: 1.1431 - val_accuracy: 0.7365\n",
            "Epoch 74/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5264 - accuracy: 0.8497 - val_loss: 1.0707 - val_accuracy: 0.7561\n",
            "Epoch 75/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5219 - accuracy: 0.8512 - val_loss: 1.1432 - val_accuracy: 0.7377\n",
            "Epoch 76/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4998 - accuracy: 0.8545 - val_loss: 1.0928 - val_accuracy: 0.7328\n",
            "Epoch 77/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.5163 - accuracy: 0.8562 - val_loss: 1.2683 - val_accuracy: 0.7451\n",
            "Epoch 78/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5101 - accuracy: 0.8518 - val_loss: 1.1681 - val_accuracy: 0.7537\n",
            "Epoch 79/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4988 - accuracy: 0.8566 - val_loss: 1.1579 - val_accuracy: 0.7451\n",
            "Epoch 80/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.5018 - accuracy: 0.8576 - val_loss: 1.1673 - val_accuracy: 0.7292\n",
            "Epoch 81/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4851 - accuracy: 0.8578 - val_loss: 1.1271 - val_accuracy: 0.7475\n",
            "Epoch 82/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4914 - accuracy: 0.8623 - val_loss: 1.2006 - val_accuracy: 0.7377\n",
            "Epoch 83/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4888 - accuracy: 0.8610 - val_loss: 1.0119 - val_accuracy: 0.7500\n",
            "Epoch 84/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4857 - accuracy: 0.8634 - val_loss: 1.1508 - val_accuracy: 0.7426\n",
            "Epoch 85/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4762 - accuracy: 0.8657 - val_loss: 1.2352 - val_accuracy: 0.7390\n",
            "Epoch 86/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4746 - accuracy: 0.8677 - val_loss: 1.2571 - val_accuracy: 0.7316\n",
            "Epoch 87/100\n",
            "248/247 [==============================] - 10s 42ms/step - loss: 0.4626 - accuracy: 0.8701 - val_loss: 1.1918 - val_accuracy: 0.7365\n",
            "Epoch 88/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4686 - accuracy: 0.8687 - val_loss: 1.2510 - val_accuracy: 0.7181\n",
            "Epoch 89/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4724 - accuracy: 0.8672 - val_loss: 1.1989 - val_accuracy: 0.7439\n",
            "Epoch 90/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4595 - accuracy: 0.8681 - val_loss: 1.1184 - val_accuracy: 0.7488\n",
            "Epoch 91/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4821 - accuracy: 0.8615 - val_loss: 1.0982 - val_accuracy: 0.7525\n",
            "Epoch 92/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4872 - accuracy: 0.8643 - val_loss: 0.9672 - val_accuracy: 0.7684\n",
            "Epoch 93/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4480 - accuracy: 0.8752 - val_loss: 1.3464 - val_accuracy: 0.7414\n",
            "Epoch 94/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4496 - accuracy: 0.8718 - val_loss: 1.2056 - val_accuracy: 0.7500\n",
            "Epoch 95/100\n",
            "248/247 [==============================] - 10s 41ms/step - loss: 0.4529 - accuracy: 0.8748 - val_loss: 1.0354 - val_accuracy: 0.7549\n",
            "Epoch 96/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4530 - accuracy: 0.8753 - val_loss: 1.1208 - val_accuracy: 0.7451\n",
            "Epoch 97/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4621 - accuracy: 0.8732 - val_loss: 1.3925 - val_accuracy: 0.7022\n",
            "Epoch 98/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4439 - accuracy: 0.8781 - val_loss: 1.2282 - val_accuracy: 0.7537\n",
            "Epoch 99/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4449 - accuracy: 0.8732 - val_loss: 1.2588 - val_accuracy: 0.7181\n",
            "Epoch 100/100\n",
            "248/247 [==============================] - 10s 40ms/step - loss: 0.4459 - accuracy: 0.8701 - val_loss: 1.3192 - val_accuracy: 0.7463\n",
            "Training completed in time:  0:16:57.144923 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "id": "PBrv3sA8X5Aa",
        "outputId": "8d4342f3-a5fc-4991-ed19-ec539e0921e5"
      },
      "source": [
        "show_results(history9)"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 76.8382 %\n",
            "\tMin validation loss: 0.83280\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xb8rY65wX71B"
      },
      "source": [
        "### fold-10 <a name=\"fold-10\"></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oapDZcbzX7Nt",
        "outputId": "c97522af-ae1a-4536-e19a-9450418641ac"
      },
      "source": [
        "FOLD_K = 10\n",
        "REPEAT = 1\n",
        "\n",
        "history10 = []\n",
        "\n",
        "for i in range(REPEAT): \n",
        "    print('-'*80)\n",
        "    print(\"\\n({})\\n\".format(i+1))\n",
        "    \n",
        "    history = process_fold(FOLD_K, us8k_df, epochs=100)\n",
        "    history10.append(history)"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "\n",
            "(1)\n",
            "\n",
            "Pre-training accuracy: 3.3453%\n",
            "\n",
            "Epoch 1/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 2.0758 - accuracy: 0.2180 - val_loss: 1.8123 - val_accuracy: 0.2640\n",
            "Epoch 2/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.8572 - accuracy: 0.2866 - val_loss: 1.7000 - val_accuracy: 0.3513\n",
            "Epoch 3/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.7047 - accuracy: 0.3516 - val_loss: 1.5606 - val_accuracy: 0.4050\n",
            "Epoch 4/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 1.5961 - accuracy: 0.3965 - val_loss: 1.4558 - val_accuracy: 0.4552\n",
            "Epoch 5/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 1.4946 - accuracy: 0.4533 - val_loss: 1.3869 - val_accuracy: 0.4421\n",
            "Epoch 6/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.4092 - accuracy: 0.4846 - val_loss: 1.2565 - val_accuracy: 0.5615\n",
            "Epoch 7/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.3362 - accuracy: 0.5188 - val_loss: 1.2145 - val_accuracy: 0.5508\n",
            "Epoch 8/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.2969 - accuracy: 0.5332 - val_loss: 1.1962 - val_accuracy: 0.5771\n",
            "Epoch 9/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.2506 - accuracy: 0.5469 - val_loss: 1.0921 - val_accuracy: 0.6511\n",
            "Epoch 10/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.2090 - accuracy: 0.5786 - val_loss: 1.1682 - val_accuracy: 0.5735\n",
            "Epoch 11/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 1.1847 - accuracy: 0.5807 - val_loss: 1.0380 - val_accuracy: 0.6368\n",
            "Epoch 12/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 1.1398 - accuracy: 0.5986 - val_loss: 1.1056 - val_accuracy: 0.6487\n",
            "Epoch 13/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.1082 - accuracy: 0.6189 - val_loss: 1.0352 - val_accuracy: 0.6093\n",
            "Epoch 14/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 1.0973 - accuracy: 0.6186 - val_loss: 1.2283 - val_accuracy: 0.5460\n",
            "Epoch 15/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 1.0705 - accuracy: 0.6205 - val_loss: 1.0078 - val_accuracy: 0.6941\n",
            "Epoch 16/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.0398 - accuracy: 0.6429 - val_loss: 0.9957 - val_accuracy: 0.6822\n",
            "Epoch 17/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 1.0342 - accuracy: 0.6431 - val_loss: 1.0132 - val_accuracy: 0.7109\n",
            "Epoch 18/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 1.0110 - accuracy: 0.6552 - val_loss: 1.0272 - val_accuracy: 0.6774\n",
            "Epoch 19/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.9822 - accuracy: 0.6684 - val_loss: 0.9067 - val_accuracy: 0.7073\n",
            "Epoch 20/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.9670 - accuracy: 0.6697 - val_loss: 0.9993 - val_accuracy: 0.6428\n",
            "Epoch 21/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.9682 - accuracy: 0.6752 - val_loss: 1.0024 - val_accuracy: 0.6703\n",
            "Epoch 22/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.9400 - accuracy: 0.6847 - val_loss: 0.9100 - val_accuracy: 0.6977\n",
            "Epoch 23/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.9217 - accuracy: 0.6918 - val_loss: 0.9173 - val_accuracy: 0.6822\n",
            "Epoch 24/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.9171 - accuracy: 0.6966 - val_loss: 0.9071 - val_accuracy: 0.6941\n",
            "Epoch 25/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.9051 - accuracy: 0.6970 - val_loss: 0.8788 - val_accuracy: 0.7240\n",
            "Epoch 26/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.8958 - accuracy: 0.7072 - val_loss: 0.9500 - val_accuracy: 0.7097\n",
            "Epoch 27/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8869 - accuracy: 0.7102 - val_loss: 0.8655 - val_accuracy: 0.7276\n",
            "Epoch 28/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8616 - accuracy: 0.7188 - val_loss: 0.9716 - val_accuracy: 0.6643\n",
            "Epoch 29/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8507 - accuracy: 0.7232 - val_loss: 0.8958 - val_accuracy: 0.7025\n",
            "Epoch 30/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8231 - accuracy: 0.7374 - val_loss: 1.0168 - val_accuracy: 0.6368\n",
            "Epoch 31/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8328 - accuracy: 0.7283 - val_loss: 0.9668 - val_accuracy: 0.6977\n",
            "Epoch 32/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8163 - accuracy: 0.7393 - val_loss: 0.9026 - val_accuracy: 0.7073\n",
            "Epoch 33/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.8054 - accuracy: 0.7376 - val_loss: 0.8846 - val_accuracy: 0.7085\n",
            "Epoch 34/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.7882 - accuracy: 0.7506 - val_loss: 0.8525 - val_accuracy: 0.7622\n",
            "Epoch 35/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.8051 - accuracy: 0.7483 - val_loss: 0.9751 - val_accuracy: 0.6631\n",
            "Epoch 36/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.7687 - accuracy: 0.7526 - val_loss: 0.7754 - val_accuracy: 0.7539\n",
            "Epoch 37/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.7636 - accuracy: 0.7541 - val_loss: 0.9535 - val_accuracy: 0.7228\n",
            "Epoch 38/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.7345 - accuracy: 0.7712 - val_loss: 0.8143 - val_accuracy: 0.7479\n",
            "Epoch 39/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.7515 - accuracy: 0.7623 - val_loss: 0.8225 - val_accuracy: 0.7336\n",
            "Epoch 40/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.7247 - accuracy: 0.7745 - val_loss: 0.8703 - val_accuracy: 0.7204\n",
            "Epoch 41/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.7207 - accuracy: 0.7752 - val_loss: 0.9171 - val_accuracy: 0.7145\n",
            "Epoch 42/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.7146 - accuracy: 0.7740 - val_loss: 0.8337 - val_accuracy: 0.7431\n",
            "Epoch 43/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.7145 - accuracy: 0.7778 - val_loss: 0.7639 - val_accuracy: 0.7539\n",
            "Epoch 44/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6894 - accuracy: 0.7856 - val_loss: 0.8005 - val_accuracy: 0.7706\n",
            "Epoch 45/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6909 - accuracy: 0.7819 - val_loss: 0.8688 - val_accuracy: 0.7360\n",
            "Epoch 46/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6878 - accuracy: 0.7859 - val_loss: 0.7524 - val_accuracy: 0.7611\n",
            "Epoch 47/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6750 - accuracy: 0.7906 - val_loss: 0.7994 - val_accuracy: 0.7515\n",
            "Epoch 48/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6554 - accuracy: 0.8018 - val_loss: 0.9509 - val_accuracy: 0.7025\n",
            "Epoch 49/100\n",
            "247/246 [==============================] - 10s 41ms/step - loss: 0.6739 - accuracy: 0.7954 - val_loss: 0.7690 - val_accuracy: 0.7503\n",
            "Epoch 50/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6736 - accuracy: 0.7915 - val_loss: 0.8818 - val_accuracy: 0.7324\n",
            "Epoch 51/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.6481 - accuracy: 0.8052 - val_loss: 0.8655 - val_accuracy: 0.7312\n",
            "Epoch 52/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6317 - accuracy: 0.8092 - val_loss: 1.0938 - val_accuracy: 0.6846\n",
            "Epoch 53/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6295 - accuracy: 0.8089 - val_loss: 0.7878 - val_accuracy: 0.7491\n",
            "Epoch 54/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6364 - accuracy: 0.8070 - val_loss: 0.7730 - val_accuracy: 0.7575\n",
            "Epoch 55/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6230 - accuracy: 0.8104 - val_loss: 1.0686 - val_accuracy: 0.6918\n",
            "Epoch 56/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6173 - accuracy: 0.8166 - val_loss: 0.8552 - val_accuracy: 0.7467\n",
            "Epoch 57/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.5948 - accuracy: 0.8200 - val_loss: 0.7362 - val_accuracy: 0.7897\n",
            "Epoch 58/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6099 - accuracy: 0.8194 - val_loss: 0.7891 - val_accuracy: 0.7575\n",
            "Epoch 59/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.6057 - accuracy: 0.8177 - val_loss: 0.8158 - val_accuracy: 0.7563\n",
            "Epoch 60/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5842 - accuracy: 0.8229 - val_loss: 0.7952 - val_accuracy: 0.7479\n",
            "Epoch 61/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5875 - accuracy: 0.8232 - val_loss: 0.8491 - val_accuracy: 0.7300\n",
            "Epoch 62/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5802 - accuracy: 0.8300 - val_loss: 0.7654 - val_accuracy: 0.7778\n",
            "Epoch 63/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5728 - accuracy: 0.8361 - val_loss: 0.8460 - val_accuracy: 0.7634\n",
            "Epoch 64/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5707 - accuracy: 0.8308 - val_loss: 0.9142 - val_accuracy: 0.7288\n",
            "Epoch 65/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5670 - accuracy: 0.8343 - val_loss: 0.9363 - val_accuracy: 0.7431\n",
            "Epoch 66/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.5783 - accuracy: 0.8294 - val_loss: 0.8745 - val_accuracy: 0.7384\n",
            "Epoch 67/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5568 - accuracy: 0.8337 - val_loss: 0.8782 - val_accuracy: 0.7658\n",
            "Epoch 68/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5677 - accuracy: 0.8337 - val_loss: 0.9270 - val_accuracy: 0.7419\n",
            "Epoch 69/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5483 - accuracy: 0.8390 - val_loss: 0.8766 - val_accuracy: 0.7742\n",
            "Epoch 70/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.5459 - accuracy: 0.8419 - val_loss: 0.7122 - val_accuracy: 0.7921\n",
            "Epoch 71/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5450 - accuracy: 0.8400 - val_loss: 0.9405 - val_accuracy: 0.7527\n",
            "Epoch 72/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5456 - accuracy: 0.8414 - val_loss: 0.9659 - val_accuracy: 0.7455\n",
            "Epoch 73/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5456 - accuracy: 0.8396 - val_loss: 0.8668 - val_accuracy: 0.7670\n",
            "Epoch 74/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.5419 - accuracy: 0.8381 - val_loss: 0.9145 - val_accuracy: 0.7252\n",
            "Epoch 75/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5295 - accuracy: 0.8485 - val_loss: 0.8846 - val_accuracy: 0.7802\n",
            "Epoch 76/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5118 - accuracy: 0.8509 - val_loss: 0.8641 - val_accuracy: 0.7718\n",
            "Epoch 77/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5079 - accuracy: 0.8545 - val_loss: 0.9544 - val_accuracy: 0.7742\n",
            "Epoch 78/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5208 - accuracy: 0.8480 - val_loss: 1.0188 - val_accuracy: 0.7180\n",
            "Epoch 79/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.5087 - accuracy: 0.8546 - val_loss: 0.8398 - val_accuracy: 0.7969\n",
            "Epoch 80/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.4991 - accuracy: 0.8529 - val_loss: 0.8967 - val_accuracy: 0.7384\n",
            "Epoch 81/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.5006 - accuracy: 0.8543 - val_loss: 0.9490 - val_accuracy: 0.7778\n",
            "Epoch 82/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.4876 - accuracy: 0.8581 - val_loss: 0.8264 - val_accuracy: 0.7957\n",
            "Epoch 83/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4891 - accuracy: 0.8604 - val_loss: 0.9031 - val_accuracy: 0.7885\n",
            "Epoch 84/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4896 - accuracy: 0.8603 - val_loss: 0.8179 - val_accuracy: 0.7849\n",
            "Epoch 85/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4901 - accuracy: 0.8584 - val_loss: 0.9374 - val_accuracy: 0.7479\n",
            "Epoch 86/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4736 - accuracy: 0.8642 - val_loss: 0.9064 - val_accuracy: 0.7563\n",
            "Epoch 87/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.4802 - accuracy: 0.8623 - val_loss: 0.8468 - val_accuracy: 0.7706\n",
            "Epoch 88/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4697 - accuracy: 0.8630 - val_loss: 0.7544 - val_accuracy: 0.7897\n",
            "Epoch 89/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4750 - accuracy: 0.8659 - val_loss: 0.8171 - val_accuracy: 0.7957\n",
            "Epoch 90/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.5065 - accuracy: 0.8504 - val_loss: 0.7829 - val_accuracy: 0.7921\n",
            "Epoch 91/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.4543 - accuracy: 0.8708 - val_loss: 0.8845 - val_accuracy: 0.7730\n",
            "Epoch 92/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.4625 - accuracy: 0.8695 - val_loss: 0.8213 - val_accuracy: 0.7981\n",
            "Epoch 93/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.4717 - accuracy: 0.8675 - val_loss: 0.9193 - val_accuracy: 0.7778\n",
            "Epoch 94/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4521 - accuracy: 0.8706 - val_loss: 0.8010 - val_accuracy: 0.7897\n",
            "Epoch 95/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4637 - accuracy: 0.8641 - val_loss: 0.8322 - val_accuracy: 0.7814\n",
            "Epoch 96/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4618 - accuracy: 0.8674 - val_loss: 0.9357 - val_accuracy: 0.7718\n",
            "Epoch 97/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4578 - accuracy: 0.8712 - val_loss: 0.8177 - val_accuracy: 0.7957\n",
            "Epoch 98/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.4513 - accuracy: 0.8675 - val_loss: 0.8744 - val_accuracy: 0.7682\n",
            "Epoch 99/100\n",
            "247/246 [==============================] - 10s 40ms/step - loss: 0.4415 - accuracy: 0.8763 - val_loss: 0.9692 - val_accuracy: 0.7682\n",
            "Epoch 100/100\n",
            "247/246 [==============================] - 10s 39ms/step - loss: 0.4565 - accuracy: 0.8695 - val_loss: 0.8899 - val_accuracy: 0.7778\n",
            "Training completed in time:  0:16:22.406434 \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "id": "SuYAxv77YBco",
        "outputId": "d3abf164-d307-4a54-e358-25ceb580d618"
      },
      "source": [
        "show_results(history10)"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "(1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\tMax validation accuracy: 79.8088 %\n",
            "\tMin validation loss: 0.71224\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "u-EnT2kFYSyN"
      },
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B30SvPqOYRSI"
      },
      "source": [
        "## 5. Results  <a name=\"fifth-bullet\"></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CX_LwbaxkTi0"
      },
      "source": [
        "### 10-fold cross validation:\n",
        "\n",
        "|fold|accuracy| loss\n",
        "|---|:-:|:-:|\n",
        "|fold-1| 0.82|0.739|\n",
        "|fold-2|0.72|0.998|\n",
        "|fold-3|0.70|1.055|\n",
        "|fold-4|0.70|1.134|\n",
        "|fold-5|0.80|0.741|\n",
        "|fold-6|0.72|1.027|\n",
        "|fold-7|0.70|0.962|\n",
        "|fold-8|0.74|1.040|\n",
        "|fold-9|0.77|0.833|\n",
        "|fold-10|0.80|0.713|\n",
        "|**Total**|**0.75**|**0.924**| "
      ]
    }
  ]
}