Course_2_Week_2_Horse_or_Human_WithAugmentation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Horse-or-Human-WithAugmentation.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "RXZT2UsyIVe_",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 399
        },
        "outputId": "6b2bc3d6-5272-4d0c-f1d5-16d5e408e9ce"
      },
      "source": [
        "!wget --no-check-certificate \\\n",
        "    https://storage.googleapis.com/laurencemoroney-blog.appspot.com/horse-or-human.zip \\\n",
        "    -O /tmp/horse-or-human.zip\n",
        "\n",
        "!wget --no-check-certificate \\\n",
        "    https://storage.googleapis.com/laurencemoroney-blog.appspot.com/validation-horse-or-human.zip \\\n",
        "    -O /tmp/validation-horse-or-human.zip\n",
        "  \n",
        "import os\n",
        "import zipfile\n",
        "\n",
        "local_zip = '/tmp/horse-or-human.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('/tmp/horse-or-human')\n",
        "local_zip = '/tmp/validation-horse-or-human.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('/tmp/validation-horse-or-human')\n",
        "zip_ref.close()\n",
        "# Directory with our training horse pictures\n",
        "train_horse_dir = os.path.join('/tmp/horse-or-human/horses')\n",
        "\n",
        "# Directory with our training human pictures\n",
        "train_human_dir = os.path.join('/tmp/horse-or-human/humans')\n",
        "\n",
        "# Directory with our training horse pictures\n",
        "validation_horse_dir = os.path.join('/tmp/validation-horse-or-human/horses')\n",
        "\n",
        "# Directory with our training human pictures\n",
        "validation_human_dir = os.path.join('/tmp/validation-horse-or-human/humans')"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2019-05-19 23:16:29--  https://storage.googleapis.com/laurencemoroney-blog.appspot.com/horse-or-human.zip\n",
            "Resolving storage.googleapis.com (storage.googleapis.com)... 74.125.142.128, 2607:f8b0:400e:c08::80\n",
            "Connecting to storage.googleapis.com (storage.googleapis.com)|74.125.142.128|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 149574867 (143M) [application/zip]\n",
            "Saving to: ‘/tmp/horse-or-human.zip’\n",
            "\n",
            "/tmp/horse-or-human 100%[===================>] 142.65M   160MB/s    in 0.9s    \n",
            "\n",
            "2019-05-19 23:16:30 (160 MB/s) - ‘/tmp/horse-or-human.zip’ saved [149574867/149574867]\n",
            "\n",
            "--2019-05-19 23:16:31--  https://storage.googleapis.com/laurencemoroney-blog.appspot.com/validation-horse-or-human.zip\n",
            "Resolving storage.googleapis.com (storage.googleapis.com)... 74.125.142.128, 2607:f8b0:400e:c08::80\n",
            "Connecting to storage.googleapis.com (storage.googleapis.com)|74.125.142.128|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 11480187 (11M) [application/zip]\n",
            "Saving to: ‘/tmp/validation-horse-or-human.zip’\n",
            "\n",
            "/tmp/validation-hor 100%[===================>]  10.95M  63.0MB/s    in 0.2s    \n",
            "\n",
            "2019-05-19 23:16:32 (63.0 MB/s) - ‘/tmp/validation-horse-or-human.zip’ saved [11480187/11480187]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "5oqBkNBJmtUv"
      },
      "source": [
        "## Building a Small Model from Scratch\n",
        "\n",
        "But before we continue, let's start defining the model:\n",
        "\n",
        "Step 1 will be to import tensorflow."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qvfZg3LQbD-5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import tensorflow as tf"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "BnhYCP4tdqjC"
      },
      "source": [
        "We then add convolutional layers as in the previous example, and flatten the final result to feed into the densely connected layers."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gokG5HKpdtzm",
        "colab_type": "text"
      },
      "source": [
        "Finally we add the densely connected layers. \n",
        "\n",
        "Note that because we are facing a two-class classification problem, i.e. a *binary classification problem*, we will end our network with a [*sigmoid* activation](https://wikipedia.org/wiki/Sigmoid_function), so that the output of our network will be a single scalar between 0 and 1, encoding the probability that the current image is class 1 (as opposed to class 0)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PixZ2s5QbYQ3",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 89
        },
        "outputId": "607f5ce8-4413-4aec-a3f9-536904887d9f"
      },
      "source": [
        "model = tf.keras.models.Sequential([\n",
        "    # Note the input shape is the desired size of the image 300x300 with 3 bytes color\n",
        "    # This is the first convolution\n",
        "    tf.keras.layers.Conv2D(16, (3,3), activation='relu', input_shape=(300, 300, 3)),\n",
        "    tf.keras.layers.MaxPooling2D(2, 2),\n",
        "    # The second convolution\n",
        "    tf.keras.layers.Conv2D(32, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    # The third convolution\n",
        "    tf.keras.layers.Conv2D(64, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    # The fourth convolution\n",
        "    tf.keras.layers.Conv2D(64, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    # The fifth convolution\n",
        "    tf.keras.layers.Conv2D(64, (3,3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2,2),\n",
        "    # Flatten the results to feed into a DNN\n",
        "    tf.keras.layers.Flatten(),\n",
        "    # 512 neuron hidden layer\n",
        "    tf.keras.layers.Dense(512, activation='relu'),\n",
        "    # Only 1 output neuron. It will contain a value from 0-1 where 0 for 1 class ('horses') and 1 for the other ('humans')\n",
        "    tf.keras.layers.Dense(1, activation='sigmoid')\n",
        "])"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Colocations handled automatically by placer.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "8DHWhFP_uhq3",
        "colab": {}
      },
      "source": [
        "from tensorflow.keras.optimizers import RMSprop\n",
        "\n",
        "model.compile(loss='binary_crossentropy',\n",
        "              optimizer=RMSprop(lr=1e-4),\n",
        "              metrics=['acc'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "ClebU9NJg99G",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "98f127a7-7e83-41b8-9d74-1d01940dd7b5"
      },
      "source": [
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "# All images will be rescaled by 1./255\n",
        "train_datagen = ImageDataGenerator(\n",
        "      rescale=1./255,\n",
        "      rotation_range=40,\n",
        "      width_shift_range=0.2,\n",
        "      height_shift_range=0.2,\n",
        "      shear_range=0.2,\n",
        "      zoom_range=0.2,\n",
        "      horizontal_flip=True,\n",
        "      fill_mode='nearest')\n",
        "\n",
        "validation_datagen = ImageDataGenerator(rescale=1/255)\n",
        "\n",
        "# Flow training images in batches of 128 using train_datagen generator\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "        '/tmp/horse-or-human/',  # This is the source directory for training images\n",
        "        target_size=(300, 300),  # All images will be resized to 150x150\n",
        "        batch_size=128,\n",
        "        # Since we use binary_crossentropy loss, we need binary labels\n",
        "        class_mode='binary')\n",
        "\n",
        "# Flow training images in batches of 128 using train_datagen generator\n",
        "validation_generator = validation_datagen.flow_from_directory(\n",
        "        '/tmp/validation-horse-or-human/',  # This is the source directory for training images\n",
        "        target_size=(300, 300),  # All images will be resized to 150x150\n",
        "        batch_size=32,\n",
        "        # Since we use binary_crossentropy loss, we need binary labels\n",
        "        class_mode='binary')"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 1027 images belonging to 2 classes.\n",
            "Found 256 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "Fb1_lgobv81m",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 5289
        },
        "outputId": "b25b6fc9-67b2-4a2f-9e2f-2537e60b42dd"
      },
      "source": [
        "history = model.fit_generator(\n",
        "      train_generator,\n",
        "      steps_per_epoch=8,  \n",
        "      epochs=100,\n",
        "      verbose=1,\n",
        "      validation_data = validation_generator,\n",
        "      validation_steps=8)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.cast instead.\n",
            "Epoch 1/100\n",
            "8/8 [==============================] - 1s 186ms/step - loss: 0.6550 - acc: 0.5039\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.6851 - acc: 0.5881 - val_loss: 0.6550 - val_acc: 0.5039\n",
            "Epoch 2/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 1.3880 - acc: 0.5000\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.6232 - acc: 0.6388 - val_loss: 1.3880 - val_acc: 0.5000\n",
            "Epoch 3/100\n",
            "8/8 [==============================] - 2s 191ms/step - loss: 0.5746 - acc: 0.6719\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.6964 - acc: 0.6641 - val_loss: 0.5746 - val_acc: 0.6719\n",
            "Epoch 4/100\n",
            "8/8 [==============================] - 1s 168ms/step - loss: 0.7321 - acc: 0.5156\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.6032 - acc: 0.7439 - val_loss: 0.7321 - val_acc: 0.5156\n",
            "Epoch 5/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 0.7762 - acc: 0.5195\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.6120 - acc: 0.6835 - val_loss: 0.7762 - val_acc: 0.5195\n",
            "Epoch 6/100\n",
            "8/8 [==============================] - 1s 169ms/step - loss: 0.6632 - acc: 0.5781\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.5305 - acc: 0.7566 - val_loss: 0.6632 - val_acc: 0.5781\n",
            "Epoch 7/100\n",
            "8/8 [==============================] - 1s 169ms/step - loss: 0.6905 - acc: 0.5742\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.5737 - acc: 0.7059 - val_loss: 0.6905 - val_acc: 0.5742\n",
            "Epoch 8/100\n",
            "8/8 [==============================] - 1s 168ms/step - loss: 0.6417 - acc: 0.6172\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.4949 - acc: 0.7868 - val_loss: 0.6417 - val_acc: 0.6172\n",
            "Epoch 9/100\n",
            "8/8 [==============================] - 2s 191ms/step - loss: 1.4190 - acc: 0.5039\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.5171 - acc: 0.7624 - val_loss: 1.4190 - val_acc: 0.5039\n",
            "Epoch 10/100\n",
            "8/8 [==============================] - 2s 189ms/step - loss: 0.6407 - acc: 0.6484\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.5519 - acc: 0.7361 - val_loss: 0.6407 - val_acc: 0.6484\n",
            "Epoch 11/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 0.7700 - acc: 0.6094\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.4986 - acc: 0.7517 - val_loss: 0.7700 - val_acc: 0.6094\n",
            "Epoch 12/100\n",
            "8/8 [==============================] - 1s 168ms/step - loss: 0.9544 - acc: 0.5781\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.4840 - acc: 0.8062 - val_loss: 0.9544 - val_acc: 0.5781\n",
            "Epoch 13/100\n",
            "8/8 [==============================] - 1s 160ms/step - loss: 2.1243 - acc: 0.5000\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.4860 - acc: 0.7760 - val_loss: 2.1243 - val_acc: 0.5000\n",
            "Epoch 14/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 1.1437 - acc: 0.5664\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.4631 - acc: 0.7790 - val_loss: 1.1437 - val_acc: 0.5664\n",
            "Epoch 15/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 0.7305 - acc: 0.6680\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.4740 - acc: 0.7838 - val_loss: 0.7305 - val_acc: 0.6680\n",
            "Epoch 16/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 1.3441 - acc: 0.5625\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.4162 - acc: 0.7955 - val_loss: 1.3441 - val_acc: 0.5625\n",
            "Epoch 17/100\n",
            "8/8 [==============================] - 2s 189ms/step - loss: 0.8358 - acc: 0.6406\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.4354 - acc: 0.8150 - val_loss: 0.8358 - val_acc: 0.6406\n",
            "Epoch 18/100\n",
            "8/8 [==============================] - 2s 191ms/step - loss: 0.9764 - acc: 0.5938\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.5326 - acc: 0.7897 - val_loss: 0.9764 - val_acc: 0.5938\n",
            "Epoch 19/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 2.2126 - acc: 0.5195\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.4089 - acc: 0.8306 - val_loss: 2.2126 - val_acc: 0.5195\n",
            "Epoch 20/100\n",
            "8/8 [==============================] - 1s 162ms/step - loss: 1.3037 - acc: 0.5859\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.4552 - acc: 0.7799 - val_loss: 1.3037 - val_acc: 0.5859\n",
            "Epoch 21/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 0.6135 - acc: 0.7461\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.3737 - acc: 0.8247 - val_loss: 0.6135 - val_acc: 0.7461\n",
            "Epoch 22/100\n",
            "8/8 [==============================] - 1s 168ms/step - loss: 1.0923 - acc: 0.6211\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.4092 - acc: 0.8150 - val_loss: 1.0923 - val_acc: 0.6211\n",
            "Epoch 23/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 1.3720 - acc: 0.5977\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.3627 - acc: 0.8208 - val_loss: 1.3720 - val_acc: 0.5977\n",
            "Epoch 24/100\n",
            "8/8 [==============================] - 1s 186ms/step - loss: 0.8341 - acc: 0.6992\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.3839 - acc: 0.8199 - val_loss: 0.8341 - val_acc: 0.6992\n",
            "Epoch 25/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 0.7478 - acc: 0.7031\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.3901 - acc: 0.8043 - val_loss: 0.7478 - val_acc: 0.7031\n",
            "Epoch 26/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 1.8302 - acc: 0.5547\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.3701 - acc: 0.8403 - val_loss: 1.8302 - val_acc: 0.5547\n",
            "Epoch 27/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 1.0272 - acc: 0.6523\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.3536 - acc: 0.8218 - val_loss: 1.0272 - val_acc: 0.6523\n",
            "Epoch 28/100\n",
            "8/8 [==============================] - 1s 176ms/step - loss: 1.0330 - acc: 0.6445\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.3655 - acc: 0.8306 - val_loss: 1.0330 - val_acc: 0.6445\n",
            "Epoch 29/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 1.6479 - acc: 0.5742\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.3312 - acc: 0.8520 - val_loss: 1.6479 - val_acc: 0.5742\n",
            "Epoch 30/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 1.5201 - acc: 0.5938\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.3341 - acc: 0.8335 - val_loss: 1.5201 - val_acc: 0.5938\n",
            "Epoch 31/100\n",
            "8/8 [==============================] - 1s 185ms/step - loss: 1.0257 - acc: 0.6641\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2840 - acc: 0.8617 - val_loss: 1.0257 - val_acc: 0.6641\n",
            "Epoch 32/100\n",
            "8/8 [==============================] - 1s 162ms/step - loss: 1.5573 - acc: 0.6016\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.3155 - acc: 0.8559 - val_loss: 1.5573 - val_acc: 0.6016\n",
            "Epoch 33/100\n",
            "8/8 [==============================] - 1s 168ms/step - loss: 0.8550 - acc: 0.7031\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.3202 - acc: 0.8598 - val_loss: 0.8550 - val_acc: 0.7031\n",
            "Epoch 34/100\n",
            "8/8 [==============================] - 1s 170ms/step - loss: 1.4147 - acc: 0.6211\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.3306 - acc: 0.8520 - val_loss: 1.4147 - val_acc: 0.6211\n",
            "Epoch 35/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 1.3067 - acc: 0.6367\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.3239 - acc: 0.8491 - val_loss: 1.3067 - val_acc: 0.6367\n",
            "Epoch 36/100\n",
            "8/8 [==============================] - 2s 189ms/step - loss: 1.0923 - acc: 0.6641\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.3466 - acc: 0.8384 - val_loss: 1.0923 - val_acc: 0.6641\n",
            "Epoch 37/100\n",
            "8/8 [==============================] - 1s 159ms/step - loss: 0.8424 - acc: 0.7383\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2696 - acc: 0.8812 - val_loss: 0.8424 - val_acc: 0.7383\n",
            "Epoch 38/100\n",
            "8/8 [==============================] - 2s 189ms/step - loss: 1.6172 - acc: 0.6016\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.3065 - acc: 0.8861 - val_loss: 1.6172 - val_acc: 0.6016\n",
            "Epoch 39/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 1.5028 - acc: 0.6094\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2720 - acc: 0.8724 - val_loss: 1.5028 - val_acc: 0.6094\n",
            "Epoch 40/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 1.5638 - acc: 0.6094\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.2570 - acc: 0.8822 - val_loss: 1.5638 - val_acc: 0.6094\n",
            "Epoch 41/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.2040 - acc: 0.5664\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.2989 - acc: 0.8773 - val_loss: 2.2040 - val_acc: 0.5664\n",
            "Epoch 42/100\n",
            "8/8 [==============================] - 1s 161ms/step - loss: 0.4493 - acc: 0.8359\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.3350 - acc: 0.8763 - val_loss: 0.4493 - val_acc: 0.8359\n",
            "Epoch 43/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 1.7386 - acc: 0.6055\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2630 - acc: 0.8987 - val_loss: 1.7386 - val_acc: 0.6055\n",
            "Epoch 44/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 3.0174 - acc: 0.5391\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.2475 - acc: 0.9007 - val_loss: 3.0174 - val_acc: 0.5391\n",
            "Epoch 45/100\n",
            "8/8 [==============================] - 2s 190ms/step - loss: 1.3051 - acc: 0.6484\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.3263 - acc: 0.8861 - val_loss: 1.3051 - val_acc: 0.6484\n",
            "Epoch 46/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 2.0998 - acc: 0.5898\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2399 - acc: 0.8832 - val_loss: 2.0998 - val_acc: 0.5898\n",
            "Epoch 47/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 1.8310 - acc: 0.6250\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.2151 - acc: 0.9143 - val_loss: 1.8310 - val_acc: 0.6250\n",
            "Epoch 48/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 2.7087 - acc: 0.5508\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.2555 - acc: 0.8802 - val_loss: 2.7087 - val_acc: 0.5508\n",
            "Epoch 49/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 1.7975 - acc: 0.6289\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2005 - acc: 0.9065 - val_loss: 1.7975 - val_acc: 0.6289\n",
            "Epoch 50/100\n",
            "8/8 [==============================] - 1s 161ms/step - loss: 1.5277 - acc: 0.6602\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.2129 - acc: 0.9007 - val_loss: 1.5277 - val_acc: 0.6602\n",
            "Epoch 51/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 1.7216 - acc: 0.6055\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.3450 - acc: 0.8773 - val_loss: 1.7216 - val_acc: 0.6055\n",
            "Epoch 52/100\n",
            "8/8 [==============================] - 1s 186ms/step - loss: 1.8668 - acc: 0.6328\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1961 - acc: 0.9211 - val_loss: 1.8668 - val_acc: 0.6328\n",
            "Epoch 53/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 2.3278 - acc: 0.5898\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2064 - acc: 0.9163 - val_loss: 2.3278 - val_acc: 0.5898\n",
            "Epoch 54/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 1.9748 - acc: 0.6211\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.3063 - acc: 0.8812 - val_loss: 1.9748 - val_acc: 0.6211\n",
            "Epoch 55/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.2643 - acc: 0.6094\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.3226 - acc: 0.8997 - val_loss: 2.2643 - val_acc: 0.6094\n",
            "Epoch 56/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 1.9685 - acc: 0.6445\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.1875 - acc: 0.9211 - val_loss: 1.9685 - val_acc: 0.6445\n",
            "Epoch 57/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 3.5432 - acc: 0.5430\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1650 - acc: 0.9348 - val_loss: 3.5432 - val_acc: 0.5430\n",
            "Epoch 58/100\n",
            "8/8 [==============================] - 1s 187ms/step - loss: 1.5685 - acc: 0.6680\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1790 - acc: 0.9260 - val_loss: 1.5685 - val_acc: 0.6680\n",
            "Epoch 59/100\n",
            "8/8 [==============================] - 1s 186ms/step - loss: 2.0373 - acc: 0.6250\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2281 - acc: 0.9085 - val_loss: 2.0373 - val_acc: 0.6250\n",
            "Epoch 60/100\n",
            "8/8 [==============================] - 1s 168ms/step - loss: 1.4753 - acc: 0.6836\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1762 - acc: 0.9289 - val_loss: 1.4753 - val_acc: 0.6836\n",
            "Epoch 61/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 2.1609 - acc: 0.6406\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1728 - acc: 0.9260 - val_loss: 2.1609 - val_acc: 0.6406\n",
            "Epoch 62/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 1.6366 - acc: 0.6562\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.2942 - acc: 0.9056 - val_loss: 1.6366 - val_acc: 0.6562\n",
            "Epoch 63/100\n",
            "8/8 [==============================] - 1s 187ms/step - loss: 1.8277 - acc: 0.6523\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.1672 - acc: 0.9309 - val_loss: 1.8277 - val_acc: 0.6523\n",
            "Epoch 64/100\n",
            "8/8 [==============================] - 1s 171ms/step - loss: 1.9448 - acc: 0.6406\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.2984 - acc: 0.9056 - val_loss: 1.9448 - val_acc: 0.6406\n",
            "Epoch 65/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.3746 - acc: 0.6133\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1969 - acc: 0.9163 - val_loss: 2.3746 - val_acc: 0.6133\n",
            "Epoch 66/100\n",
            "8/8 [==============================] - 2s 188ms/step - loss: 2.0485 - acc: 0.6523\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1624 - acc: 0.9299 - val_loss: 2.0485 - val_acc: 0.6523\n",
            "Epoch 67/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.7302 - acc: 0.5938\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1989 - acc: 0.9182 - val_loss: 2.7302 - val_acc: 0.5938\n",
            "Epoch 68/100\n",
            "8/8 [==============================] - 1s 162ms/step - loss: 3.1617 - acc: 0.5547\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1656 - acc: 0.9328 - val_loss: 3.1617 - val_acc: 0.5547\n",
            "Epoch 69/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 3.6691 - acc: 0.5391\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1504 - acc: 0.9299 - val_loss: 3.6691 - val_acc: 0.5391\n",
            "Epoch 70/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 3.0546 - acc: 0.5664\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.1938 - acc: 0.9153 - val_loss: 3.0546 - val_acc: 0.5664\n",
            "Epoch 71/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 2.3030 - acc: 0.6016\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.2647 - acc: 0.9279 - val_loss: 2.3030 - val_acc: 0.6016\n",
            "Epoch 72/100\n",
            "8/8 [==============================] - 1s 168ms/step - loss: 3.3331 - acc: 0.5508\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1613 - acc: 0.9241 - val_loss: 3.3331 - val_acc: 0.5508\n",
            "Epoch 73/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 3.1991 - acc: 0.5586\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1524 - acc: 0.9357 - val_loss: 3.1991 - val_acc: 0.5586\n",
            "Epoch 74/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 1.8471 - acc: 0.6367\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.3733 - acc: 0.8851 - val_loss: 1.8471 - val_acc: 0.6367\n",
            "Epoch 75/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 2.8477 - acc: 0.5781\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1375 - acc: 0.9455 - val_loss: 2.8477 - val_acc: 0.5781\n",
            "Epoch 76/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 3.2904 - acc: 0.5586\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1414 - acc: 0.9387 - val_loss: 3.2904 - val_acc: 0.5586\n",
            "Epoch 77/100\n",
            "8/8 [==============================] - 1s 170ms/step - loss: 3.1233 - acc: 0.5586\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.1483 - acc: 0.9338 - val_loss: 3.1233 - val_acc: 0.5586\n",
            "Epoch 78/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 1.5243 - acc: 0.6875\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1746 - acc: 0.9172 - val_loss: 1.5243 - val_acc: 0.6875\n",
            "Epoch 79/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 2.2195 - acc: 0.6367\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1624 - acc: 0.9299 - val_loss: 2.2195 - val_acc: 0.6367\n",
            "Epoch 80/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.2228 - acc: 0.6328\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.2100 - acc: 0.9309 - val_loss: 2.2228 - val_acc: 0.6328\n",
            "Epoch 81/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.3325 - acc: 0.6250\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.3463 - acc: 0.8929 - val_loss: 2.3325 - val_acc: 0.6250\n",
            "Epoch 82/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 2.7073 - acc: 0.6211\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1348 - acc: 0.9406 - val_loss: 2.7073 - val_acc: 0.6211\n",
            "Epoch 83/100\n",
            "8/8 [==============================] - 1s 165ms/step - loss: 2.1169 - acc: 0.6602\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1728 - acc: 0.9328 - val_loss: 2.1169 - val_acc: 0.6602\n",
            "Epoch 84/100\n",
            "8/8 [==============================] - 1s 185ms/step - loss: 1.5326 - acc: 0.6914\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1688 - acc: 0.9435 - val_loss: 1.5326 - val_acc: 0.6914\n",
            "Epoch 85/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 3.2405 - acc: 0.5703\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1394 - acc: 0.9396 - val_loss: 3.2405 - val_acc: 0.5703\n",
            "Epoch 86/100\n",
            "8/8 [==============================] - 1s 162ms/step - loss: 2.6419 - acc: 0.6211\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1491 - acc: 0.9289 - val_loss: 2.6419 - val_acc: 0.6211\n",
            "Epoch 87/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 1.1449 - acc: 0.7539\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1302 - acc: 0.9426 - val_loss: 1.1449 - val_acc: 0.7539\n",
            "Epoch 88/100\n",
            "8/8 [==============================] - 1s 162ms/step - loss: 3.8786 - acc: 0.5586\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1564 - acc: 0.9318 - val_loss: 3.8786 - val_acc: 0.5586\n",
            "Epoch 89/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 2.5629 - acc: 0.6172\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1686 - acc: 0.9270 - val_loss: 2.5629 - val_acc: 0.6172\n",
            "Epoch 90/100\n",
            "8/8 [==============================] - 1s 164ms/step - loss: 4.4178 - acc: 0.5273\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1761 - acc: 0.9348 - val_loss: 4.4178 - val_acc: 0.5273\n",
            "Epoch 91/100\n",
            "8/8 [==============================] - 1s 187ms/step - loss: 1.9289 - acc: 0.6641\n",
            "9/9 [==============================] - 25s 3s/step - loss: 0.1487 - acc: 0.9377 - val_loss: 1.9289 - val_acc: 0.6641\n",
            "Epoch 92/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.4224 - acc: 0.6328\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1400 - acc: 0.9406 - val_loss: 2.4224 - val_acc: 0.6328\n",
            "Epoch 93/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.5415 - acc: 0.6289\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1336 - acc: 0.9435 - val_loss: 2.5415 - val_acc: 0.6289\n",
            "Epoch 94/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 1.8339 - acc: 0.6523\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1838 - acc: 0.9182 - val_loss: 1.8339 - val_acc: 0.6523\n",
            "Epoch 95/100\n",
            "8/8 [==============================] - 1s 161ms/step - loss: 3.4769 - acc: 0.5625\n",
            "9/9 [==============================] - 24s 3s/step - loss: 0.1093 - acc: 0.9572 - val_loss: 3.4769 - val_acc: 0.5625\n",
            "Epoch 96/100\n",
            "8/8 [==============================] - 1s 162ms/step - loss: 2.9243 - acc: 0.6250\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1078 - acc: 0.9562 - val_loss: 2.9243 - val_acc: 0.6250\n",
            "Epoch 97/100\n",
            "8/8 [==============================] - 1s 166ms/step - loss: 2.6068 - acc: 0.6289\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1404 - acc: 0.9455 - val_loss: 2.6068 - val_acc: 0.6289\n",
            "Epoch 98/100\n",
            "8/8 [==============================] - 1s 187ms/step - loss: 3.0162 - acc: 0.5820\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1565 - acc: 0.9396 - val_loss: 3.0162 - val_acc: 0.5820\n",
            "Epoch 99/100\n",
            "8/8 [==============================] - 1s 167ms/step - loss: 2.7071 - acc: 0.6289\n",
            "9/9 [==============================] - 23s 3s/step - loss: 0.1004 - acc: 0.9611 - val_loss: 2.7071 - val_acc: 0.6289\n",
            "Epoch 100/100\n",
            "8/8 [==============================] - 1s 163ms/step - loss: 1.4930 - acc: 0.6953\n",
            "9/9 [==============================] - 22s 2s/step - loss: 0.1539 - acc: 0.9455 - val_loss: 1.4930 - val_acc: 0.6953\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7zNPRWOVJdOH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "outputId": "174a11e4-5bc4-4af6-a4bf-8c819ee0e37a"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['acc']\n",
        "val_acc = history.history['val_acc']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'r', label='Training accuracy')\n",
        "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "\n",
        "plt.figure()\n",
        "\n",
        "plt.plot(epochs, loss, 'r', label='Training Loss')\n",
        "plt.plot(epochs, val_loss, 'b', label='Validation Loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzsnXecFPX5xz8PB0g/jo4gRQ6RQ8Vy\nYMGCYsOCNQZijd3Iz9g1MRpjNMYeTYwl0VhiQ2MhBrEAsWAJIEWKwElvR4fjKMfdPb8/nn2c787N\n7M5svd37vl+ve+3t7Ozsd3ZnPvOZ5/t8ny8xMywWi8WSXzTKdgMsFovFknqsuFssFkseYsXdYrFY\n8hAr7haLxZKHWHG3WCyWPMSKu8ViseQhVtzzGCIqIKJtRNQjletmEyIqJqKU5+8S0fFEtMR4Pp+I\njgqybgKf9Xci+nWi77dYgtA42w2wOBDRNuNpCwC7ANREnl/FzK+E2R4z1wBolep1GwLM3C8V2yGi\nywFcwMxDjW1fnoptWyyxsOJej2DmH8U14gwvZ+ZP/NYnosbMXJ2Jtlks8bDHY/3ChmVyCCK6l4je\nIKLXiKgCwAVEdDgRfU1Em4loNRE9QURNIus3JiImol6R5/+MvP4BEVUQ0VdE1DvsupHXhxPRAiLa\nQkR/JqLJRHSJT7uDtPEqIiojok1E9ITx3gIieoyINhDRIgAnx/h+7iCi113LniSiRyP/X05E8yL7\n80PEVfttawURDY3834KIXo60bQ6AQ1zr/oaIFkW2O4eIRkSW7w/gLwCOioS81hvf7d3G+6+O7PsG\nInqXiLoG+W7CfM/aHiL6hIg2EtEaIrrV+Jw7I9/JViKaSkR7eoXAiOgL/Z0j3+dnkc/ZCOA3RNSX\niCZFPmN95HsrNN7fM7KP6yKvP05EzSJt7m+s15WIthNRe7/9tcSBme1fPfwDsATA8a5l9wKoAnA6\n5MLcHMAgAIdC7sL2BrAAwOjI+o0BMIBekef/BLAeQCmAJgDeAPDPBNbtBKACwBmR124EsBvAJT77\nEqSN7wEoBNALwEbddwCjAcwB0B1AewCfyWHr+Tl7A9gGoKWx7bUASiPPT4+sQwCOA7ADwAGR144H\nsMTY1goAQyP/PwzgvwCKAPQEMNe17nkAukZ+k59F2tA58trlAP7rauc/Adwd+f/ESBsPBNAMwF8B\nTAzy3YT8ngsBlAP4JYA9ALQBMDjy2q8AzATQN7IPBwJoB6DY/V0D+EJ/58i+VQO4BkAB5HjcB8Aw\nAE0jx8lkAA8b+zM78n22jKw/JPLaswDuMz7nJgDvZPs8zOW/rDfA/vn8MP7iPjHO+24G8Gbkfy/B\nftpYdwSA2QmseymAz43XCMBq+Ih7wDYeZrz+NoCbI/9/BglP6WunuAXHte2vAfws8v9wAPNjrPs+\ngGsj/8cS92XmbwHgF+a6HtudDeDUyP/xxP1FAH8wXmsD6WfpHu+7Cfk9Xwhgis96P2h7XcuDiPui\nOG04Vz8XwFEA1gAo8FhvCIDFACjyfAaAs1N9XjWkPxuWyT2Wm0+IaF8i+k/kNnsrgHsAdIjx/jXG\n/9sRuxPVb909zXawnI0r/DYSsI2BPgvA0hjtBYBXAYyK/P+zyHNtx2lE9E0kZLAZ4ppjfVdK11ht\nIKJLiGhmJLSwGcC+AbcLyP79uD1m3gpgE4BuxjqBfrM43/NeEBH3ItZr8XAfj12IaAwRrYy04QVX\nG5awdN5HwcyTIXcBRxLRfgB6APhPgm2ywMbccxF3GuAzEKdYzMxtANwFcdLpZDXEWQIAiIgQLUZu\nkmnjaogoKPFSNccAOJ6IukHCRq9G2tgcwFsA7oeETNoC+ChgO9b4tYGI9gbwFCQ00T6y3e+N7cZL\n21wFCfXo9lpDwj8rA7TLTazveTmAPj7v83utMtKmFsayLq513Pv3ACTLa/9IGy5xtaEnERX4tOMl\nABdA7jLGMPMun/UsAbDinvu0BrAFQGWkQ+qqDHzm+wAOJqLTiagxJI7bMU1tHAPgeiLqFulcuy3W\nysy8BhI6eAESklkYeWkPSBx4HYAaIjoNEhsO2oZfE1FbknEAo43XWkEEbh3kOncFxLkr5QC6mx2b\nLl4DcBkRHUBEe0AuPp8zs++dUAxifc9jAfQgotFEtAcRtSGiwZHX/g7gXiLqQ8KBRNQOclFbA+m4\nLyCiK2FciGK0oRLAFiLaCxIaUr4CsAHAH0g6qZsT0RDj9ZchYZyfQYTekgRW3HOfmwBcDOngfAbS\n8ZlWmLkcwE8BPAo5WfsAmA5xbKlu41MAJgD4DsAUiPuOx6uQGPqPIRlm3gzgBgDvQDolz4VcpILw\nW8gdxBIAH8AQHmaeBeDPAP4XWacfgG+M934MYCGAciIywyv6/vGQ8Mk7kff3AHB+wHa58f2emXkL\ngBMAnAO54CwAcEzk5YcAvAv5nrdCOjebRcJtVwD4NaRzvdi1b178FsBgyEVmLIB/GW2oBnAagP4Q\nF78M8jvo60sgv/MuZv4y5L5bXGjnhcWSMJHb7FUAzmXmz7PdHkvuQkQvQTpp7852W3IdO4jJkhBE\ndDIkM2UHJJVuN8S9WiwJEem/OAPA/tluSz5gwzKWRDkSwCJIrPkkAGfZDjBLohDR/ZBc+z8w87Js\ntycfsGEZi8ViyUOsc7dYLJY8JGsx9w4dOnCvXr2y9fEWi8WSk0ybNm09M8dKPQaQRXHv1asXpk6d\nmq2Pt1gslpyEiOKN0gZgwzIWi8WSl1hxt1gsljzEirvFYrHkIVbcLRaLJQ+x4m6xWCx5iBV3i8Vi\nyUOsuFssFkseYsXdYrFY0klNDfDss8DWrRn9WCvuFovFkk4++AC46irg/vsz+rFW3C0WiyUMkyYB\nhx4KnH02cPPNwBtvALEKMI4ZI49PPQVUVGSmjbDibrFYLOEYMwaYMQP4/nvgySeBkSOBF1/0XnfX\nLuC994DSUmDLFuBvf8tYM624WywWSxhmzhTnPncusG0bcPTRwC9/CazwmPb2o48k1v773wNDhwKP\nPQZUVWWkmVbcLRaLJSi1tcCsWcDAgfK8oAB4/nmguhq44oq64ZkxY4CiImDYMOCWW+QC8PrrGWmq\nFXeLxWIJyqJFQGUlcOCBzrI+fYA//hEYPx74xz+c5Tt3SkjmrLOAJk2A4cOB/fYDHnwwdow+RVhx\nt1gsiXPxxRJ3bijMmCGP6tyVa68FjjkGuOEG4H+RqYQ/+kg6UM87T54TiXufM0cyaNKMFXeLxZIY\nO3YAL70EjB4N3HNP+t3ohAlA167ArbcCmzZ5r1NbK1kpSwOVPPemshL4+9+BQYOAE0+Mfm3mTKBR\nI2DAgOjljRoBL7wAtGsnMfiXX5aQTLt2wHHHOeuNHAmcfLI4+XTDzFn5O+SQQ9hiseQw8+czA8x9\n+8rjzTcz19Ymv90ZM5h//nPmTZucZZs3M++1F3O7dsxEzEVFzI88wlxdHf3eN9+UtpSWMu/eHe5z\nZ89mHj2auU0b2UZRkTwuWuSsc/rpzP37+29j3TrmY4+V9xUUMF92Wbg2BADAVA6gsda5WyyWxFB3\n/Mwz4t4ffhj42c/8XXVQ3nxTYtcjRsjdASD55CtXAuPGAdOnS7bKTTcB993nvK+6GvjNb8QtT50q\nse147NoFvPqquO399pORpKedBnz+OfDVV7LO+PHO+jNn1g3JmHToAHz4oXwftbXAhReG3/9UEeQK\nkI4/69wtDYYVK7LdguTYuZP59tuZFy+OXv73vzvOtraW+d57xa3uuSfzBx8k/nk/+Qlzy5bi0E8/\nnfn99+Vzbrster1Ro5gbN2aePl2eP/ecrPfOO8znncfcpAnzrFnen7FkCfOttzJ36CDv6dOH+aGH\nxHkrtbXMvXszjxghzzdulHXvvz/YfmzcGG6/A4KAzt2Ku8WSTl54QU6zv/wlu+3YuJH53XeZt28P\n/95775V9eOCB6OV33ikCXFXlLJs6lbmkJJwIujngAOZTT2V+8kknvFFSwrxjR/R669czd+7MPHAg\n89atErYZPFhEee1a5o4dmQ8+OLp9zBJ+KSqS7Z51FvNHHzHX1Hi35Zpr5EKzcyfzf/8r7Rk3LrH9\nShFW3C2WbLNjhwgOkQjJxInJbe/dd5m//jr2OjU13gJ+wQX8Yxz5hhskXh6EH35gbtZM3nvFFdGv\nXXwxc7dudd+zY4c47latJFZuUlUl+/Dqq8y//z3zyy/XbX/z5sw33ijPf/972c4333i37913pW0H\nHCCPn3zivPbWW7LsvPMcF714sdxZdO0a7DsYO1a2MWEC8+OPy/8rV8Z/Xxqx4m6xZJvHHpNT7O23\npROufXsRSzdVVcxXXSXOeO1a723t2sXcujXzEUfE/swLL5RQwpYtzrIFC5gbNRKRO+88CWXssUdd\ncautZf7yS/ksfX7KKSKuffsyDx0avf6xxzIffrh3O6ZNk31/9NHo5WecIcv1r2lT5/OYmZculeVP\nPx2977E4/3x5z7BhdV/7wx9kf/fck/mVV2Q/2rb1D9e4qaiQNt5yC/Oll8rdQCo6jZPAirvFkk22\nbpV4rgrOwoXimvfbTwTDZNy4aLH72c/qxuknTZLXGzVi3rDB+zOnTnW2c/31zvKf/1zc9+rV8nzx\nYnHHF18c/f5nnpH3FhczjxnjON9HH2W+6CLm7t2j1997b+aRI/2/gyOPlAuNZrR88IFs75ZbJDSi\nMfKZM533fPyxLJs0yX+7bjZskH2ZM8f79WnTmAcMkO02b848eXLwbTPLb7jffhLiOf74cO9NA1bc\nLZZscs89cnqZ4YTx42XZY49Fr3vhheImv/2W+f/+T4T4wguj17n1Vke4X3/d+zNPOklSBS+4QC4C\n06dLZ2dBAfMvfxm97g03yHK9k9i8WVzpwIGOEDZqJM9373b2R0M+NTXSYXnrrf7fgaYlvvuubKOk\nRC4c6sTnzpXXX3zRec9f/sJpCX3s3Mn84IPMn34a/r0PPSRtatyY+aabUtuuBLDibrGkimXLxBUH\nZf16yZU+88y6rx12GHO/fs6tfWWlhD0uv9xZ57LLZJkZOz/gAOajjhLxvuSSutvVzr6HHpL4cqdO\n8lmXXSZ3A+47gVWrJDSjn6sXjylTxGk//7yEgKZMkddfe01enz1bnq9cKc+ffNL/e9i9W/ocjj3W\n6Rx95x3n9epqcdLmXcZ118m+Zzn0EcV33zkX1pdeynZrrLhbLCnjrLPkVLn66rohFS8efFDW/+67\nuq+9+KK8pp2rb7wR/ZxZOgUBCY0wizBrtsrIkcxdukRnd9TWihB36+ZcEF56yRGkX/zCu52jR4sb\nnThRLgAXXeS/T1OmOC6cmfmrr+T5v/8d+7v44x9lvTZtJGbvFu1DD42O5Z90koQ/6hO1tRKScoeQ\nsoQVd4slVfTpIyl3RBJnjhezLS1lHjTI+7Xt28V9/+Qn8vyMMyRzwxxpWV0tAq7O34xNa2ql5nYz\ni8ACEjNXamtFNJs0kU5KL5YvF1Fv1oy5RYvY+fia4/3QQ/JcL0rxOiY3bBB3ThTdZuWqqyQkpaLf\nu7fkr9c3rrpK9iNe524GCCrudoSqpWEwdiywfn349+3YIZUAr7kG+PRT8cJnnulfR2XxYhkd+ZOf\neL/evDlwySXAO+8A8+ZJAamRI6V0rFJQIMvGjQM2b5Z19twT2H9/4KSTZB0tPLV9O3DjjUBxMfDz\nnzvbIJKKhN9+C/To4d2W7t2BSy+V6oW33QZ06+b/PRQVAe3bA2Vl8lxHp/ptW2nXTiom3ndfdCVF\n5cADZR+XLpV2LFkC7LNP7G1mg/vvByZPBpo2zXZLghPkCpCOP+vcLRlj+XJxmb/6Vfj3Tp8eHSL5\n29/k+cKF3us/8IC87h7NabJggayjHZca1zb55hvHjRcWShqectBBzEcfLf/fcEPdsE4Y1qxhvuMO\nif3H47DDnOwfrcGSLF9/zT/G4mfPlv9ffTX57eYxsM7dYokwebI8ainWMMydK48lJfJYWiqPU6d6\nrz9mjFQT7NXLf5t9+wLHHy+lX/v2BQ45pO46gwZJnfC77pLp2YYPd147+WTZp/ffB/70Jyk3e+yx\noXcNANC5M3DvvUCLFvHXLS6Odu49eyb2mSb77y8VFWfMAObPl2X10bnnIFbcLfmPivuUKVLMyWTX\nrtilaufOlTBJ377yfMAAoFkzb3FftAiYNs2p3x2Lq6+Wx1GjJITihkiKcJWXy+cff7zz2vDhQE2N\nhH569ZKwRyYoLgaWLZPwybJl8UMyQWjRQsR8+nRgwQJZZsU9JQQSdyI6mYjmE1EZEd3u8XpPIppA\nRLOI6L9E1D31TbVYEmTyZBHLrVsdAQFE1AcNAgYP9p7/EhBx79vXibU2aSJx4ilT6q775pvy6Bdv\nNznzTODPfwauv95/nVGj5PHww4G2bZ3lhx0GtGkjIvv880CrVvE/LxUUF8t3tnixiHsqnDsAHHSQ\n49y7dgVat07Ndhs4ccWdiAoAPAlgOIASAKOIqMS12sMAXmLmAwDcA+D+VDfUYkmIbdukTOuZZ8pz\nMzQzfz7w3XfiwktLgS+/rPv+uXOdkIxSWiodlTU10cvHjJELRRDRKyiQsrBFRf7r9O8PXHedlLY1\nadJESts+8IBMupwp9O5l+nQp65sK5w7IxXLZMuDrr4F+/VKzTUsg5z4YQBkzL2LmKgCvAzjDtU4J\ngImR/yd5vG6xpJ7ly8W53nSTxKW9+OYbEeHLLhNHaIr7xx/L4/vvi/s99ljgX/9yXt+1S2LMbnEf\nNEguGhojBoAffhDBDxKSCcPjjzsXJpNbbpEZiTJJcbE8Toyc6qkS94MOksfvv7chmRQSRNy7AVhu\nPF8RWWYyE8DZkf/PAtCaiNq7N0REVxLRVCKaum7dukTaa7GIkO6/v4jLZZcBjz4KPPKI97oakhky\nRDouTXH/6CMRrFNPleX9+wN33um8vnChXBi8nDsQHXfXGe3PPTf5/auvtGsndxoTJsjzVIVlzMkv\nrHNPGanqUL0ZwDFENB3AMQBWAqhxr8TMzzJzKTOXduzYMUUfbWlwPP20uOZHHpGwytlni8PdvLnu\nupMnSydo27YSMpk5Uxx5VRUwaZIzR2a7dnKhmDdPRB1wMmX694/eZr9+QMuWjrjX1sodxLHHpk7w\n6ivFxZKLDqTOuXfqJHn8gHXuKSSIuK8EsJfxvHtk2Y8w8ypmPpuZDwJwR2SZx5lmsSQJM/D225I9\ncuONMjXanXdKZ+kTT0SvW1MjcdwhQ+T54MEi6rNmyRRqlZXREyCPGCGPY8fK49y54vrdbrKgQO4C\ntFP1v/+VTJnLLkv57tY7NO7euLF0fqYKDc1Y554ygoj7FAB9iag3ETUFMBLAWHMFIupARLqtXwF4\nPrXNtFgizJolQnr22c6yAw8UYf7Tn0TklTlz5Lkp7oCEYD76SETazA/v2VNCBO+9J8/nzgX23ltG\nlbopLZUMj927geeeAwoLo9uUr2jcvXv36FG1yXLkkRLyiTU+wBKKuOLOzNUARgP4EMA8AGOYeQ4R\n3UNEEauDoQDmE9ECAJ0B3Oe5MYslWd5+Wwa9qMtW7rxTMjiefNJZpvntKu7duwNdujjifvjhklJo\nMmKEvG/9egnRuOPtSmmppCJOniydsOef730RyDdU3FMVklFuukk6VJs0Se12GzCBYu7MPI6Z92Hm\nPsx8X2TZXcw8NvL/W8zcN7LO5cy8K52NtuQBZWXROedBeftt4KijJE5rUloKnHKKxOHXrpVlkyeL\nmPfuLc+JxL1/8okMNjrhhLrbP+MMiaG/957E9f3EfdAgebzpJonhX355+H3JRVTcU9230KRJ3d/U\nkhR2hKols1RXSxGmkhIR6W3bgr93wQJg9mz/8Mdvfyudqj16ABdeKCl7Q4ZEjwAdNAhYtUpi92a8\nXTn4YCmg9dhjEnLxE/c+fSQU8+23Ei/WmHG+ozH3VDt3S8qx4m7JHAsXSmz1178Gjj5aHLa7E/Tj\nj6WjdOPGuu9/5x15POss7+1rNszll4vzXr1aLiDudQDJntGURhMiCc3MmSPP/cSdyHl/Q+hIVTp0\nAP7xD+Cqq7LdEks8glQXS8efrQrZABkyRGp3v/aa1O8+/XSpeKhzgi5aJJUGAZnQeNy46PcPHuxf\nJ91NRYXMXG/OZsQsnwUwn3OO/3t1Ojwg9uQcv/udTFq9cWOwNlksKQC2KqSlXlFbK8PWL7lEapUT\nSY3vrVuBBx+UEIguf/ttyZw45RQZ8fnII8DLL0tHaNCMlFatgNNPr9vJ2a6dhFx+/Wv/9w4dKqNZ\ne/SIXbflttskLh+rhIDFkiUaZ7sBlgbCkiUyscSAAc6y/feXyodPPAGsWyfi/eabEnY55RTg7rtl\nwJIW5AJSk24Yq1gXAOyxh6zjriDptV4qc70tlhRCHKvcaRopLS3lqX41sS35x7//LbHsL7+UFETl\nhx+AffeVjtarrwaeeir6fcyS4vjDDyK2hx6a2XZbLPUMIprGzB4dRtFY527JDNpB6R7K36ePhEg+\n+0xqxLghklBKu3bpb6PFkkdYcbdkhrlzJcXQrEuu/O53mW+PxZLn2A5VS2aYMyc63m6xWNKKFXdL\n+qmtlaH8Vtwtloxhxd0Sm+XLZbDOq6/WfW1XwCoTixcDO3ZYcbdYMogVd0ts3n1X6rCcf77kddfU\nSFGt66+Xolu33hp7gmnA6Uy14m6xZAzboWqJzaRJUiRq+HAZbPTppxJi2bZN6rA89BDQvr0Ivx9+\nmTIWiyVtWOdu8ae2VsT8uOMk//ypp2SU6dChMgPSN98Ao0YBt98O/O1v/tuZO1fK7RYWZqzpFktD\nx4q7Rfj+e+Dii2UUqTJrlhTw0gktrr4aqKiQolwlJVJX/YUXgJNPltfuuENi9G5spozFknGsuFuE\nJ54AXnopeqj/pEnyaM5W1LRp9PuaNgXeektKBtx/v8ykc8YZzvyjNTU2U8ZiyQJW3C0iwP/6l/z/\n9787yydNkskZuneP/f6WLUXgFy2SEM0XX4jAb98umTI7d1pxt1gyjBV3iwz9X7sWOOwwEeb580Xw\nP/ss2rXHo1cvqfT4r3/JTEt33GEzZSyWLGHF3QKMGQO0aAG88opMevz889JxumVLOHFXhg4Frr0W\nePxxp6PVZspYLBnFintDYNw4mYLOi+pqcdqnnw7svTdw2mnAiy/KBNKACHUi/PGP4uT/8x9gr73q\nTkRtsVjSihX3hsBf/gLcc4/3hNSffSa11H/yE3l+2WVAeTnw8MNSijfReuWtWsl0bIANyVgsWcCK\ne77DLCNMAeC55+q+PmaMdIgOHy7Phw8XQd+0KbGQjMkxx0gGzm9+k9x2IPNQn3YaUFWV9KYslgaB\nFfd8Z+VK6Sxt2lTCLbt3O6+ZIZkWLWRZ48aS7w4kL+4AcOGFwJAhSW9m8mSJ8JSXJ98ki6UhYMW9\nPsEM/PnPwIYNqdumuvabbxZlHDfOeW3iRKkToyEZ5frrgdGjZaq7eoLWKNu5M7vtsFhyBSvu9YmZ\nM4HrrgP++c/UbXPaNBlJetttEm7R0MzGjcAvfgF06eKEZJTOneUi07Jl6tqRJFbcLZZwWHGvT3z3\nnTyWldV9Ld5kzX5MmyalAtq0kXDLuHHAsmXAeedJqYC33waaN0+8zRlCxT1olWGLpaFjxb0+4Sfu\nzEDfvsAtt4TbnnamHnKIPL/0UhmcNHQoMGEC8PTT0ZNV12O0I9U6d4slGFbc6xMq7gsXRi9fsUKG\n9j/8sNRXN/n2W3HfM2ZIUS+TVaskzq7i3rcvcPTRUhLg+uuBn/88PfuRBmxYxmIJhxX3+sTs2fK4\nZEl0Vsu8efLYvr2472XLxJU/+igwaBBwzjnAQQdJ6OWKK5z3aWeqijsAPPKIlAV46KG07kqqseJu\nsYTDint9YdMmcej9+0voZOlS5zUV97FjJX3xZz+T+PlNN0k1xilTpJrjyJFS+EvruWhn6oEHOtsq\nLQXuvVdSHnMIK+4WSzisuGeDhQulSJdZ+1xd+1lnyaMZd583Dygqkvj4M89I0vfLL8uo0zFjRLDP\nPVdGorZoIeEbQMS9f38nhz2HseJusYTDins2ePppmcXorbecZRpv9xP3/v0BIpn56K9/layXO+8U\nZ660by/lA155RQYvmZ2pOY4Vd4slHFbcM011tYgvAIwf7yz/7juZhu6QQ6Qui9mpquKuXHNN3dx0\n5cYbJW3yttuANWvyRtxttozFEg4r7olSU5NY0vVHH0kGS//+Mj+pTmv33XfA/vuLOy8udpz7hg1S\n2CtoydxevSSHXS8geSLu1rlbLOEIJO5EdDIRzSeiMiK63eP1HkQ0iYimE9EsIqo/49bTxc03A0ce\nGf59L70k4ZOHHxbFmjRJMl9mzxZxByRlUcVdO1PD1EPXfHh3Z2oOY8XdYglHXHEnogIATwIYDqAE\nwCgiKnGt9hsAY5j5IAAjAfw11Q2td0yZAkydKqGPoGzeLHnqo0YBxx0nHZ3jx0vH6pYtjrgXF0su\nenV1YuJ+0EEStjn44HpVQiAZrLhbLOEI4twHAyhj5kXMXAXgdQBnuNZhADobQyGAValrYj1l8WJ5\n/OKL4O95801RqYsuApo1k6qLH3zgZMqY4r57t4j+vHlSHqBnz3Dte/NN4OOPw72nHmPF3WIJRxBx\n7wbAyNnDisgyk7sBXEBEKwCMA/B/XhsioiuJaCoRTV23bl0Cza0n7Nghoz8B4PPPg7/vpZdkAozS\nUnk+fDjwww/AO+/I8/32k8fiYnlcuFDEvV+/6KyYILRsCbRtG+499RjboWqxhCNVHaqjALzAzN0B\nnALgZSKqs21mfpaZS5m5tGPHjin66CywZIk8EgUX9x9+EJd/8cXyPgA4+WR5fOklmYpOxbhvX3ks\nK6ubKdNAsc7dYglHEHFfCWAv43n3yDKTywCMAQBm/gpAMwAdUtHAeomGZI4/Xsr0bt0a/z133gns\nsQdwwQXOsj59RMirqpyQDCBleFu0kG0vXWrFHVbcLZawBBH3KQD6ElFvImoK6TAd61pnGYBhAEBE\n/SHinsNxlzgsWiSPF10kOeVffRV7/U8+AV57Dbj9dqB79+jX1L2b4q7pkDqxhhV3K+4WS0jiijsz\nVwMYDeBDAPMgWTFziOgeIhq05OmNAAAgAElEQVQRWe0mAFcQ0UwArwG4hJk5XY3OOosXSyfnGWcA\nBQWxQzO7dgHXXitifXudLFJnMNIBB0QvLy6WWjOAFXdYcbdYwhKoehQzj4N0lJrL7jL+nwsg+Yky\ns8mtt0qd8yBTyy1aBPTuDbRuLWmHscT9wQeBBQtk8FKzZnVfP+kkGXB0zjnRyzXuXlDg/N+AsR2q\nFks47AhVQEIrjz4KPPBAsPUXLQL23lv+P+ooqRPjNVr1u++A++4DfvpT4IQTvLfVqJFUedxjj+jl\nmjHTp49Mbt3Asc7dYgmHFXdA5hOtqZFqi5s2xV6XWcIyprjv2iUDmpTaWuCxx4DBg6XG+qOPhm+T\nirsNyaCmRv4AK+4WS1CsuANS6wUQBfnoo9jrbtggMx717i3PtQSBhma+/VYGJ914I3DiicCsWcCe\ne4ZvkxX3HzFviqy4WyzBsOIOOOIOAO+/H3tdzZRR596xowxMevll4NBDpVDXjBnACy9IqYEuXRJr\nU7duwP33SwnfBo4Vd4slPFbcAWDtWnk8+GApB6AxAC80x13FHZA6MXPniqN/4gnJTTcHKyUCkWTX\nqINvwGhnKmDF3WIJihV3wHHul14qYZf//c9/XXXuvXo5y/7wB5kYY84c4P/+L6+G/dcH1LnvsUdi\nVZYtloaIFXdAxL2gQKo1FhTEDs0sWgR06iQTaiiFheL6k3HqFl9U0AsLrXO3WIJixR2QsEynTkC7\ndsCQIcB//uO/rpkpY8kIVtwtlvBYcQfEuXfuLP+fdprUdDEnrzbRAUyWjGGKe3W1/FkslthYcQei\nxf3UU+Vx3Li661VXA8uWWeeeYUxxN59bLBZ/rLgDIu6dOsn//ftLZ6mXuC9fLpk0VtwzimbLtIlM\nB2NDMxZLfKy4M0vMXZ07kdR7mTRJZkMy0UwZG5bJKG7nbsXdYolPwxL3qirg7LNlFKlSUSFqoeIO\nyMjSigqpGWPiHsBkyQhW3C2W8DQscV+8WKa0e+89Z5nmuGtYBpBBSY0a1S1FsHgx0Lhx3ZrslrRi\nxd0SFGbgN7+RGSobOg1L3FevlseyMmeZirvp3Nu2lVICbnEvK5OJqgsK0ttOSxRW3C1BWbdOCrG+\n+262W5J9rLhr6QFT3AEp0TtlilSMBORx3DinUJglY9gOVUtQdMbLHTuy2476gBV3r7AMIHH32lpg\n4kR5/tRTQGUlcNNN6W+nB3PmAFu2ZOWjs4517pagWHF3yE9xZ5ZRprW10ctV3DdudBy5invHjtHr\nai32jz+WI+WJJ2RKPHOu0wwyZEhiZeHzASvulqBYcXfIT3GfPFlGmn7ySfRyFXfAce/l5UD79kCT\nJtHrNmkiHasffgi8+KKEb269Nb3t9qGqSlz7mjVZ+fisY8XdEhQV9+3bs9uO+kB+irtOLP3DD9HL\nV68GiorkfxV3rSvjxYknSvneu+4CBg0CjjkmPe2NQ2WlPOqB29Cw4m4JSkWFPFrnnq/irqGWZcui\nl69eDRxxhAxUMp27uzNV0XlP160T156lqo/btsljrsfcr74auO228O+rqpLM1JYt5bkVd4sf1rk7\nNM52A9KCn7ivWgUMGwbstZeTCFteLrMnedGnjwxYIgLOOit97Y2DOvdcF/dPPklsxsFdu6SWe7Nm\n8tyKu8UPG3N3yG9xX7rUWbZjh6hj164yu5EZlvFz7kTAv/4FNG2a1dx2de65HJZhlhsnswx+UKy4\nW4JinbtDwwnLaGdq165A374i7jt3ytHgF3MHgAMPBEpK0tfWAOSDc9+2TU44jYmGwYq7JSjWuTvk\nt7ivXOkU/zLFvbgYWL8eWLBAlvk593pCPoi7fv2J3H2ouDduLLF3K+4WP6y4O+SvuBcUSJ77qlWy\nzC3ugKRMAvVe3DUsU1FRN3U/V9CvP1Hn3rSpRMmaNbPibvFHjy8blslHcWcWcdfBRhqaiSXuscIy\n9QB17syO0Oca+vXv2uWUEwhKVZU4d8CKuyU21rk75J+4b9kiajBokDzXTtXVq+W+vkMHyYIBcs65\nA7kbmjEHYIV17xqWAeTRirvFD9uh6pB/4q7xdhV3de6rVgFdukjQtnlzKdu7ZIm8liPOHcjdjBlz\ncHAy4t6smZ1mz+KPde4O+SvuvXpJvRjTuXft6qynoZlWrYAWLTLaxLDkg3M3xT3sBcot7ta5W/zQ\nY2vXrtztn0oV+SvunTsDPXpEx9y9xL2eh2SAaOeeq+KeqrCMFXdLLEzj0NDde36Le8+eeSHupnPP\n5bBMly7yf9h9qKqSbBnAirvFn5oaMUIdOshzK+75Rnm5xNU7dBDnvnSpqMP69dHi3revPOaAuFdW\nSjcBkLvOffVqYJ995H/r3C3pQE2QntINvVM1P8W9Y0fJc+/RQ5Rx3jx5zcu51/POVEAOWq3Jkovi\nXlUFbNhgxd2SXvSOUMXdOvcAENHJRDSfiMqI6HaP1x8johmRvwVEtDn1TQ2IWeWxZ095/OYbeTSr\nVvXpIzXbe/TIbPsSoLJSdqlRo9wMy2ikTMXddqha0oEV92jiijsRFQB4EsBwACUARhFRVLEVZr6B\nmQ9k5gMB/BnA2+lobCBMcVfh/vpreTSde8uWwJdfAqNHZ7Z9CbBtmyT1tGkTzLnPmgWMGAFcd136\n2lRVBRx9NPD55/HX1UwZjYRZ525JB25xt2GZ+AwGUMbMi5i5CsDrAM6Isf4oAK+lonEJ4eXcvcQd\nAEpLnVmX6zGVlSLuhYWxxX3VKuCSS6TW2b//LfN5p4uVK0XYP/00/roq7t27S9ap7VC1pAM9rrTj\n3jr3+HQDsNx4viKyrA5E1BNAbwATfV6/koimEtHUdevWhW1rfLT0gIp7hw7SE/n991KYJAfi615s\n2yY3Gm3axBbGq68GXnsNuPlm4Kc/TayOS1DcU9DGQtMgu3aVfbDO3ZIO9Liyzl1IdYfqSABvMXON\n14vM/CwzlzJzaUf3hNSpYNs2uVzrr0skoRlmEfbGuVm+PqhzX7wYOOUU4MEHxSWnsw6NivvatfHX\nXb3auba2bh3OuTNbcbcEIxdi7jt3AlddFSycmSxBxH0lgL2M590jy7wYiWyHZIDo9EaNu7tDMjlE\nZaU493jivnEj0K6d/N+6tTiXGs/LbPKEFfcOHaT/unXrcM5dKzZbcbfEIxfEfdUq4Nln607vnA6C\niPsUAH2JqDcRNYUI+Fj3SkS0L4AiAF+ltokhiCXuiczvVg+oqZGDVDtUY7leU9x1xiNzdGsqCROW\nMcePhQ3LaB0ZU9x3707fRctSfykrAw47TKY09kLPDY2+1sewzIoV8tjNM7CdWuKKOzNXAxgN4EMA\n8wCMYeY5RHQPEY0wVh0J4HVm5vQ0NQBe4q6dqjnq3PUAjefcd+wQR+sW93SFZsI49zVrnK8/bFhG\nywObHaqALR6WS1RUpGZ8xjvvSFbz7Nner2/dKh32miNRH527inv37un/rEBBaGYeB2Cca9ldrud3\np65ZCZKHYRkV53jirmJrhmWA9HWq6udt2CBOukkT/3VXr3ZmKkyFcwfkQlbP671ZIlx2mRy3H36Y\n3Ha0Qvdmn1E0FRVyfOlo7vro3FdGAtqZEPf8GqFaXi49d2ZnbY47dw2raFimqsrbtarYtm/vrA+k\n37kDUtnBj9ra5Jx7LHHPJ557LliIKxdZuhSYPj25bTDLsBTA3+Bs3SrnSJMmMkC9vjr31q0d85VO\n8k/c27ePzorZf3+gXz/g8MOz164kcDt3wPvgdjv3TIp7rNDMxo1AdXVqY+5Afon7ihXA5ZcDzzyT\n7Zakh61bJU6eTGimrMyJtccSdxXNFi3qr7hnwrUD+Sju7kJg7dtLnvtBB2WnTUliOvcw4p6JsIwK\nbizHqQOYdGBJ69YizJoFE4+GIO46Z8ycOVltRtrQ4zWZDBENyZjbc6POHZDQTH0Ny2SiMxVoCOKe\n45jOXQ9cr7BGNpy71oqJ5dzNqWuB8BcdFXd3h2o+ivvcuVltRtrQ47WsLPFtTJ4MtG0rjjyIuFvn\nbsW93pOoc1dxT6dz33df+T+MuOvJF7Rdmi2Tz85dJwubPz/4HU2uoDXWgeTE/csvJbLatm3uOvfq\najkfrLiHZOdO4NkVp6C2U5dsNyWlhIm5N2ki6wGOQ06Hc2eWLJnevcVRxwrLaOkBMywDBO9UbQhh\nGRX33buTE8D6iHkRT3TfNm6Uu5ohQ+QciJctA4i41zfnXl4uCQY2LBOS8e/twlW7nsCMmv2z3ZSU\n4s6WAfzDMu3aSbIQ4Ih8OsS9slKEqH17GTASz7m3auXcSYR17g1B3JcscdI68y00YxqRhQsT28ZX\nkWGRRxzhnw7MXLdDtb4590zmuAN5JO5bV4jibW1Rf5z7tGnAmWcmd6sdxrlrSAYQF7/HHukJy5hp\nl506xe9QNbNQG7Jz37gROOEEZ+ZHZelSYOhQuTDnW6eq/s6FhYk798mTJbVx8GB/cd+1S86z+uzc\nM5njDuSRuG8vFxXb1jwNBckSZMIE4L33nB81EdS5myPvgog7IEKaCudupj2az9u1ky6OWM7dzHHX\nNgHJO/dcHKE6cybwySfRg3mYRexLSiTMla/ifvDBciwkcjx++aUku7Vs6R9z189JpEN1x47MXAgy\nWXoAyCNxr1wtv25ls/ZZbomDiuCmTYlvY9s2cSEFBeLGmzePHZYxadUqeXGfO1fGhE2ZEv1ZgHxe\nvLDMkiXRZX0S7VDVbBkV+Vx07ipKZuilvFz2pWdPYMCA/BZ3IHw65O7dwP/+J/F2wN+5u8U9TIfq\nqFHAyJHh2pUIK1bIcawTeKeb/BH3edIrta3dXnHWzBwqgm7nGwYt96v4HdwbNng792TDMnPnSifQ\nrFnOMrdzLy8XB+pmyRJxpeb4sfoclvnii+j9TDX6u5kCrp2pPXuKe1+wIL8yZnSfVdzDhmamTxdX\nfcQR8jyMuAd144sXy112dXW4toVl5UoJyWi/WLrJD3FnxvYFcs+zbUf9qdm+YYM8JuPctdyv4ndw\np8u5r1olj8uN6Vrczr2qylusJ0yQx+OPd5alKiyTanFnFvd2662p3a6JfkemuGuOe69e4tzzLWPG\n7dzD7tv//iePahAKC+W3d4fl9HhKpEO1okLOs5kzw7UtLCtWZC4kA+SLuM+di8qKWgDpnaAiLKlw\n7jp/quJV9nfXLjk427siUpkSd8A7NDNhgsTb+/d3ljVu7B9a8iJT4r5smTirdNbZ1ovyqlVOOp/p\n3AcMkP9V/JmlLMENN8Su31Of0d+5e3c5VsJmzMycKceZdkL6JRUk49z1veYo2HSQyQFMQL6I+wcf\noBJib9NVvzwRUhFzD+LcdftBwjK1teFCNSruZobHxo0iss2bO2PG3BkzzMDEicBxx9W9DQ1TX8Yt\n7hp7T7W464m9dGn6asWbv5vG3ZculU7CNm1kUJiZMTNxohQU+9OfgD59gD/8Ib4b3b27fmWJbN0q\n+9SypUyQHta5z5wJDBzoHENBxV07VIMUINdjUQuTeWEOxkoE5syWHgDyRdzHj0dlG+m1y3fn7iXu\n7tGpipdzf/RROcDMDtJY+Dl3/Sw/5z5njgj+sGF1txmmMqR2qGpJYaL0zMak4r57d3LZTbHYuhVo\nFDnjVMCXLJGQDCCC1Lu3I/wPPSSDv779VlIl77gDOOec2IJ1663AkUemp/2JoKNGiYDi4nDiXlMj\ntdsHDnSWtW0rj0GcOxD/ONm1yznGYjn33//eKVudCBs2yGdZ5x6GbduAzz/H9g49fnyaSqZOBb77\nLrH3psO5e4VlYom72yGXlcmy4cOlnlo8THFXUfESd7dz13i7n7iHce5NmjiiCPiLOzPw6qvOyRqG\nL790vudFi8K/PwhbtgB77y0iruK+dKlTlRpwMmZmzpSUyeuukzTA994DHnkEGD8eeOst/8/49FPp\nFK4vM1Vt2eK47eJiuXAGjYUvXCju2xT3MGEZIP5djB6HfftK2MQ9BgGQ4+qVV+S1RO+KMj2ACcgH\ncZ84EaiqQmUbSaZOtbhfey3wi1+Ef9/Onc5BnGwqZKLO3SvPfcMGCaUUFAAnnhjtyL1YtUri5Nu3\nR9+J6Gdp6Xy3c//kEzmZda4Uk7BhGQ3JKH7iPnMmcP75MmNPGCoqRBDPPlueL14c7v1B2bJFnGdJ\nibhzZm9xX7AAuP9++d2vvtp5TYX++uu973x275YLQ3W1c1HONma9l+JieQx68dQOzqDiXlDgiLqO\n+I13IdHj8OST5dHLvX//vXPHkWjfh94N2rBMGMaPB1q2RGUT+dVTHXPftEkOstra8O9Tkk2FdMfc\nKyqinVks515VFe1kN26Uk2z8eDlBTjnF3+VVVso6Bx4oz/VCYIp7kybyvynu1dXiIL1cOxAuLBNG\n3LXed1hx/vpr+X1HjpQ7hGSd+/z5wE9+UreNW7fK71dSIiK8caNcfDUsA8hru3cDb7wBXHEFUFTk\nvNa4MfD00zLq9+67vT9Xf2vNwsk2XuIeNDQzc6bssxkOUXF315epqJDjSmPzQZ27HodHHinnmVfc\nfawxY3Si4m6de1iYgQ8+AIYNw/Ydsiupdu4VFfIX9mQxBd3LuQeNGXtly+hy92d5ibt73Q0bJKvm\noIMkjjh7tlO50Y0uP+wwedRbVnfapbsEwZQp8p35iXu6nLt+z5qBEpTJk0XUjzxS7jSSde6vvy6h\nE7eIaYhiwABx1upM3c4dEBd6/fV1tz14MHDVVcATT9RN3TOfh/0O0oWXuAfNmJk5UzqZzd8/lnPX\nzwEc5x40LFNUJMe5l3MfO9Zpg9/k3PFYuVKOsS4ZrI6S2+K+YIGo7vDhPzr2VIu7bi9sDqwKbteu\ndZ37zJniMuIViWL2du5A9MG9caOIgXlwA96VIVXcAYn/Av638Lr80EPlUZ27e8CUuwTBhAnioI49\n1nu7YZ27Zsgo8cQ97IX4yy9lwq42baRDM1nnPnWqPOo4B8UUd0B8CRAt7vvuK3dDo0Z5h7QAyZpp\n1046WE1mzXI6nuuLuJsx97ZtZXRmGOduhmQA/xIcbnEPOo+qHoetW8tAqZkzo41HebkULtOQXTLO\nvUuX6Eni0k1ui/unn8rjCSekRdxra5MX9+Lius5d46LffBN7G5rK5Y65A3XFvaiobsqhn3NXYdb4\nn5+4a5zwwANFYJcvlzbt3BmdU+927hMmyHv8hlmHce5VVel17jU1EpbR4e17752cuDM74u6+qKsA\naZhhXGTKeTMs06IF8NlnwF/+4v8ZRUUSQpo4MXowz8yZcuHo3Dm9YZkPP6x74fLDLbrFxXL8x0tR\n3LBBjj+3uBcUiBAHFfegzr1NGzkGamujz8v//Efaetll8jxR557pHHcg18V94UI503v3/vEKncqY\nu7mtRMW9Tx858MyhzVrjfP78YJ/vzpYBop2v1+hUoO5oUC2QpMKsNV/iOffu3eVv2TLvEJDp3Jct\nAz7/3Omg8qJ1a2lHkOHeYcIy5sCgIPnNgGRCVVQ4w9v33lsuVImWi1250vl9TXHX8QWFheLUW7SQ\nO7eWLev+docd5lzE/Rg2TL7Dr792lqnT7dkzfc69okL6aZ58Mtj6btE96igJfRx6qOPNvPDqTFW8\nkgr8wjJBO1Rbt5bvnSg67j52rNxBDR0qYZVkOlStuIdh4UKgTx8wNUqLcze3lYy4A9EdQOpy44m7\nWe5X8XPuXuLudu5mqV5AMl0KCmKLe/Pm8pl77SXO3UvcO3WS/auqkgE3QHSWh5swJQgSiblv3x78\nJNQYqzr33r3l0Yy7//Ofkr0SBHXtQLS4V1TIBaewUERC3XvPnonVGjnmGNmOppyuXSsXlXSL+5o1\ncqEKcnejA39M0b3/fuAf/5D+nKFDgfPO867wmYi463EFhO9Qbd1atrv//tJfsnixvPejj4ARI+Q8\nad8+ubBMJjNlgFwX97IyoLgYO3fKiVNQIEIW1LXFQ8WnpER+7KBxYkBuKwsKnHiqGZoJ69yDhGVi\nibvuh95Kq7g3aiR9An6DdlatEndPJO4llrgD0gXy7LOx48VAuMqQiYg7EFzcvvxS9lF/J+2HMMX9\n/vuBBx4IdlxNnSq/e+PG0eLuzsNWcTdDMmFo2xYoLXXE3RRDFfewGV5BUGMSJOzj3mdAvptLLpFj\n5Xe/A958E7jwwroZWzNnyh2h16yZXrMxmbMwAeE7VPXCcMst4hn33Rc491x5/4gR8lrHjvHDMl53\noxUV8l1Y5x6U2lopBFJc/OOtV8eOcpCkqta3/vA64i9MxUAVXBVB80TXE6SsLHZoIkxYxl1XBqjb\noeoWd0CELZZzV7ex117iPvTgdodlADlZKyuBm2/23yezXUEulmE7VPW7Ciru33wjRanUPatzV2e6\nerWET7Zs8c8qMpk6FdhvP/mOzd9cL8Z6cdZOVbMzNSzDhkn7t26NFvdeveR7i1WKOVHUmAT5fs2J\nOtw0bw7cdZeMwn3zTWD06OiLp1dnqhIkLBOmQ7VFC7noAMAFF4i4X3CBpAsXFspdEiB9SH7OffZs\n4LTTxFC5v5tM13FXclfcV62SM7xv3x9FUEUmVXF3t7iHCc24xd3Lue/eHdsBqSjHc+5e5X7N97nF\n3Vw3nrhrXH6vveTCqRk+Xs79rbeAk07yPymVMM49TIfq5s3AAQfI/+7vdeLEuh3bzHLiaegMEIPQ\nsqUj7hMnOq/Fq7WunamlpfL9BBH3RJ07INU2a2qkA3bmTBGP9u2dC0Y6QjNqTFasiN9n4uXc3dx8\ns5RMePpp4Fe/km3u3i3HWVBxr6mRYzzRDlV3+7p3l5o+s2fL76/mwsu5b98uxd0GDpSBe7t2RYfm\nALlLAWQUbCbJXXHXZNni4h/FXEUmVXF33c6++0qGQlhxb9/eGYRiCkt5uXNbHis04+Xc1WXowb17\nt5xEiYRlAH9xZ44Wdw2zzJghj17iDshtbTzCOvcwYZleveRkNYWtvFyE8JlnotffskW2b976E0lo\nRsMyEyY4t/jxxH3pUvmOY4m7CklpqfyvaaaJcMQR8l1MmBDtdFXc05Exo8YkyCjYIOIOAH/8owjk\nAw/Ixfnxx+Wi7ifu7tmY9DxNpEPVHas36d/fKVUMeDv3l1+WC8Ho0c657D6n9Xm/frHbkmpyV9w1\nWdYIy6Ra3M143MCBiTl3FXc90Wtq5HZZb/ViibuXcyeSg1hjjvroJe7uSbLdHaqAiPfGjd6jKbdv\nj3bugHwHTZpEX3BUHA8+WKpAxiOdMfeiorodilOnysXKXWpBXah7YInmujOLcA4fLt+Ze1zCp59K\naQqNF2sxtkGD6oq7O0TRubMIlB4HidCsmXQEf/ABMG9eXXFPp3MPsn33Bc0PIumreecd+S7VIAR1\n7l4XkaZNZbuJOHc/OnaUi7fZl7FokXzWY4/J996tm7e4d+4cPwMq1eS2uDdtCnTvXicsk4i4e3WW\nucX9u++CF2Ryi7s6dz04BgwQwYhVvMvLuQPiAFRI/EanAtJh2rJldFimRQunJjrgiLc7nqyuzC3u\nixbJZ5kZHq1aSd2Txx8PlvmRDudeWysXuqIice+ma9XbZPc+qgt1d9qpc//hB0ntHDbMKRlg8vTT\nwFNPyZ9+TtOmEnOPF5ZJFcOGiXhUVzti2KaNfA/pcu7qiuOJe6yYuxsimUx+9mz5Pq+5JnoeAJPC\nQnH2egzosWreQRIFm0dVyxYEoUMHOf/NztzlyyWMo4Xt+vWrK+7ff5951w7kurj36QMUFNQJy4SN\nuTNLCtSjj0YvV1FUcd+xI/joOhX3pk1FYPVENwXF60Dw+nzTuQPSez9lihzUscRd32uGZdzr+eW6\nu8W9sNA5CdzbIBJhD1pqNhWpkFVV0Q6qokKeezl3vRDqd6/Ecu6VlcBrr8nzYcOcao2mCdA0yjvu\nkAvH1KkSVthjj/hhmVRhlngwnW660iHLy4FDDpH/4108goZlTJo0kTTav/7V6eR0464vo+ekO6bt\nnke1tFRCQO42Bm2fDsozQzPLljnGB3DOafM4mT/fins4Fi78sVhFsmGZdevkxJ09O3q5ik/Lls6J\nEyQ0s3u3vFdFsF07x7mbghJP3P2cu6Zm/fvf8cXdrAxplh5Qgoo7kXMQ+31WUMKIe1WVd7YMEJ0V\npSd627YibFu2yDJzxGgY5w5ILLV7dxGNAQOiM2aWL5e/0aPFQd54IzBtmoRkAPmOtm1zCnlp1UL3\nb5kshxwiYte8ebS49eqVHnFfs0Yufp06BXfuqb6guZMKFi6U41MznRTTue/cKb+P9hkpYZy7VkA1\nO1WXL49O++3XT447XWfDBvnbd99gn5FKclPcmX/McQeQdFhGb7e9cme1A7OkRB6DiLt7ZqSiImeZ\n27mXl3vPiar70bSpUy9EKSkRARo7Nrxzd4u7pme5c931uYo74BzEyYp7kyYi0MmEZYDo0Ix+vxqW\nAUR8Vq6U77hNGxFm01GVl0s+unt/VNyXLhVnTFR3CjwdxXjJJcDtt0uxsC1bxB0CdVNgt2xxJq1I\nJQUFwFlnAUcfHe10e/YUZ52qMR+AbKu8XI7dIHcGW7Y4ocFU4hb3sjIxHma4EYh27trf4u4QjdWh\n6sbt3Gtq5PhyO3fACbdmqzMVyFVxX71aLskZEHf94Zs1k6uv+8rvhVtwi4qcZW7nDvi7d3fRMIUI\nOOMM6ezTSo1BnLtXPnxRkYinl3MvLIz+/FQ5dyC6vsw//ym15fXvttuc9WKJu+ncTXE3OxTVtZ90\nkqxvXkjXrBEH2sh1FpjpiRr20OwmPVYmT5YL/8CBksKn6ZQq7vo9m+Kerg61555z6tQoPXvK8ZNM\nuWk3ml3UpUuwOwNzFqZU4p6NqazMO83QnEdV2+pOZQzboWpuY/VqEXi3cwfqZs5YcQ+KK8jmDsuE\njbn7ibu73O4hh0j8Np4bcou7GZZZs0YOulatnFs1P3F3f77JiBFyor3+upw8fsJhTrXn5dyJvNMh\nzTRIRcXda8BUWLQy5D1nZp0AABplSURBVGuvyQjFJUuknQsXAg8+KL9hba10FIZ17qa4T5ki7lxr\n3ZihmfJy7xKsLVo4y1XcO3WS/TbF/dBDZdvNmgEvvghcdJHj8N3OPUxsNyyNGvlfoFIZmlFjYjr3\nWKNg07XPXs5dywmbtGjhaIP2D5jOvapKzqFEnbveDZjOvUcPOR70nP7+e7lTTWY8Q6IEEnciOpmI\n5hNRGRHd7rPOeUQ0l4jmENGrqW2mCyPHHXDEXK+sYZ27pri5B7m443FDhshVO1496njOvXNnJ5+6\noCC8cwek87KoSPoJ2rb173zSsAyzf5mCoOKeqrAMICf9F1+IIB5zjIz+/fJL4OGH5fXvv687Obai\n4m52lumFuahIjoPmzeWE1hGjGo81xX3NGu/h7YAcWv37R/c5DBggx8q2bRKe03o0gPz/4ovO7+AV\nlslkKlw6ct01pNili2w/3ijYTIj7pk1iWrzE3c+5qzkzK0IGoXlzOR/VuXuJe0GBeE7TuRcXZ7bU\nrxJX3ImoAMCTAIYDKAEwiohKXOv0BfArAEOYeQAAj2kGUkhZmVwOI99qZaUIQLNm8hhG3JmDhWUA\n52SONUs6EN+5qyts2lQEPhHn3rgxcOqp0Z/jhYZltmyRW0gv1x3WuadC3Fu3lpDSfvvJ/KAq2GZs\nWzsj3R2qXvO26vfbtq0IsTpLHTHaVWZhDOTcAUnH02wZRTNmvvlGvktT3N3UF3EP49yZZZSl352p\n6dyD3BloP0OqMbNlfvhB/vdz7m5x12QHoG5dmSCYA5k0JOquo2QmSmQrUwYI5twHAyhj5kXMXAXg\ndQBnuNa5AsCTzLwJAJg5DVUtDMrKxIpFLofbtzu5t2YYIghr18qVX0e9mbeZ27ZF//D9+8t6sWZJ\nB+oO8y8qcuqgq3NX+vXzz3WP5dwBJ2smltjq9+E1OlVxi7t7dKpSUiIi7Jd/HIa99wb22cep36EU\nF8t1e84cf+euBZi0Zgcg4t6okfN79ewpw/I3bowWd3WfZuegF/vtV3cQjWbMvPWWXEB0hiovMhmW\n8ft8rzonsfjwQ+CEE4BJk7xfN/uLgtwZpGufdTq9LVuixjLWwexQNb8HFedEsnlMcV++3KkmadKv\nn4wH2bFDLj7ZyJQBgol7NwDm2L4VkWUm+wDYh4gmE9HXRORZzZuIriSiqUQ0dV2iVe+BOj0opgi2\nbBku5q6u/fDD5YQ30/MqKqKdc6NGsl48cd+4MToObg5kMp07IAfCwoXeg6NiOXdAOgmbNIkdA9ew\nTDxx1+kEAVl39+664t6li5xQQ4f6f15Q/vY3CcW4xbVxY/lOYom7Zvi4xb1tWyf23LOnc/usQ/2b\nN3ec+6ZNso9hpj3TTtVXXhGh1449L9q0kVv0bDl3vXsJE5bRSSr8zMaaNbJP7doFuzPQOWNTjV7E\nTXHXDCcTMyyzZIlzwdXjIhHnbtaXcee4K/36yfn8ySdyjNVn5x6ExgD6AhgKYBSAvxFRnUOfmZ9l\n5lJmLu2oAfKwMEfluAPR4h7WuWu8XSdrMEMzXjmwQ4bIUO9YWQg6M5IKjXlQrV8fLWj77isiprd4\nJvGce5s2wJ13SoldP1q3lrsRTW30E3fAET518V5V7NwhkkRp3LiuaCsa2/YT9xYt5Ps10ze19ICi\nYYOmTWWAGpEIue6jX457LDRkVFEROyQDyOdpXwtz5sUdCD+QSQd7+Q3U0zudRo2cUbDxxD1ddyt6\np11WJsep3rmbaIdqdbUcK5rJlErn7lXaWsX8vfein2eaIOK+EoB5feoeWWayAsBYZt7NzIsBLICI\nfeopLxfVM8Q9mbDMnDlyoGiowexU9RN3QOZV9MPdcamis2CBnOhu566vuamsjO3cARH3iy7yf13f\nryehX4cq4Ii6Fs1yO/dMMWCAtEF/C6+LQPfu0c5dSw8o6iwHDnQuSF27OuLuNzo1Fp06ORkT8cQd\ncEap7tolDi6TYRlALnCLFwfLdTcHe/mJu7sDOt6dQbpi7oBTX8bl86JQ575ypThpHVmbKue+fLm/\ncwdkkKH5PNMEEfcpAPoSUW8iagpgJICxrnXehbh2EFEHSJgmyWmGffAIsiXj3OfMETFRYVDnzuwd\nFhk0SG5NY4Vm3OKu/8+bJ4/mCaIHh9eEGdu2JT8AxC3usZy7tuGtt+SCd9BByX12omj4Q8cUeN0t\ndOvmHZZRVNzVrQEi7urYE3HugOPew4h7uurKxKN/f/lsv8lYTHSwF5F/Npi7AzpWrnt1tZiudIu7\nX4474HSo6gXIz7mH7VDVBIW1a73FvbDQmXqyQ4fUJCAkQlxxZ+ZqAKMBfAhgHoAxzDyHiO4hokiX\nHj4EsIGI5gKYBOAWZg44hW5IPApJJBpz10yZkpK64r59u7zu/uFbthTRiyfupojqtjUEZJ4g+r/X\nRBBBnHs8tP1LljihAjemc6+oAN5+G/jpT+uO+MsUKqDffiuPfs49VlimXz9xbmbtlWSdOyCi3rdv\n3aHuXmRb3MOUzFDXftRR0hno1Qfk5dz95qtVV5yufS4sFOe8dm1s515b61ys9t9fjILbuYe5AGk0\nWY2H34xj2omaLdcOBIy5M/M4Zt6Hmfsw832RZXcx89jI/8zMNzJzCTPvz8yvp63FO3bIEWZ8q5WV\niYVl1q6Vk8/sHFNxj3XLNmQI8L//ya22F35hGe2oMk+QZs3kdbe4V1XJ9lPp3P3y4Vu3ls9ZtUqE\nffv22KGedFNcLCfh9Ony3E/cy8uddEm3uHfoIK+ffbazrEsX+X137BChato0dqeoF7/7nZzYQUZd\nqrinq8ZKPHTikqDiXlAAnHOOfKfmXRHgZBeZF8OePeVc8+p/Svc+FxbGzpQBnAk79I55r72iY+Z6\njocxUBqWU+Ph5dwBR9SzlSkD5OII1WuukTPTKLiyfXtiYRnNlPESd7+KjICI+86djvi4cYt7YaGI\ngea+ukMBZkefotktXk47DHpxWrrU//aQSMIcq1YBL70kQ+kPPzy5z00GzZjRaQ29xL1bNxEcrRfj\njrkDTsqcoumQ5eXRg8nCts2r886Ldu3kd8yWcy8slNBJEHGfMkXSP/WC4I67a3aReezGynXPhLgr\nfuKuv9O8efLbN2sWLe46xV6YAUZhxb3eO/f6TqIxdxX3khLnINROvFjOXTNrvEIzWu/ZFNKCAjkY\nd+yQdrovGGYsWFHnlOykuvpZfvOsKnvuKSf4pEni2lNdDyQsJSVOGpufcwckNLNjh7jNeBdCcyBT\nrNGpqaJdOxEQvVBnWtyBYJPMmNMDqlC6xd0rjBUrHTJdJY4V87s0p0k0Uef+/fdOW80O0TAVIRUN\ny6i4+52fmqBhnXuSmGEZjbn7ZQjMmOEceJop07WrI8BBwjLdusnB4iXufjMj6XOvGK8ZC1Y0npwq\ncQfii7tmVlxwQXKfmQo07g54d6iaA5nM0amxMMU91ujUVKG/uYpftsR94cLY080tWSIX/0GD5Dho\n1qyuuHt1QMcayBRmoo5E0N+6Sxf/sIpqwpIlTlvdzj3sxUed+/ffO2UuvDjhBCnoNnx4uO2nkpwX\nd+a6YRlm7xlY1q4Vd9Knj0wuMXOmiIi61LZtg4k7IEWj9Opt4leCV12ll1tUcTcvSKmaMd1sfzxx\nB6RDzWtASKYxxd0vLANEi3s85252XmfCuev3ramlmY65AyLutbV15yow0c7U0lLJYe/TJ5hz15nG\nvMKTmQrL+IVkAEd4mZ0QUrLOXcev1Nb6h2QACfVceml2asooOS/uO3fKj2eKO+Admikrk7BJu3bA\n9dcDX3/tpN0B0eIeK+YOyA/rFmTAX9xjOfcuXWQ/zHK0K1ZIt0KiY72UMM4dyG5Hqon5u3iJe9u2\n4sxWrgwu7h07yom5apVc6DPl3LMt7oDTf+HFlCnO9ICACGYQ504kHbBvv103Qy1T4u6XBglEu2rT\nuW/e7EwsH7Z9OkIX8M+UqS/kvLjrQWVmywDe4q63x++8I3U0TjpJUv6UMM69c2cRZPdsQok6dyA6\n7r5ypbhTdynXsDRrVnekrBfDhklZ3PPOS+7zUoVmzADe4k7kDGQyK0LGoqBAvv/Zs+Uin4mYOyDi\n3rJldlxc795yTsSKu5vTAwKOuJt1lsrLxWy4v+OLLpJz8J13opdnKuYey7mbHd9mzB2QfpBEnLu5\njVjOvT6Q8+KusUQzzx3wznVXce/ZUyaFGD8+Og+6bdtgHaqA4/rcHaHuomGKnhR+MXcgOu6+YkXy\nIRlARFD3IZZzP+AA4IMPsuMuvdCMGcC/TIEOZArq3AH5rjVHOVPOfcmS7H2vjRrJb+sn7rW10dMD\nAiKYO3dGF5PTMJa7o33IELmAvPRS9PKtW9MzC5Oi322QsAwQ7dwBibsnKu66DSvuacY9z2gs575k\niQicX6ilqCh4WEZdn1l2Fogflonl3N3inmxnqqL7kIpJNjKJhmb8xF0HMgXtUAVE0LWOT6ac+65d\n2elMVQYOlLCMV5JBWZkIsTmS1ytjxq8DulEj6YD/5JPoQWXpmoVJGTwYeOYZmZHMDy/nbop7orVv\n1LnbsEyaCRuW0R/ZC3dYplkz/1tpP+e+fr0c0G6hieXc3dtilhOloYv7EUc40wB60a2bfE96txRE\n3PVCCqTfuev4Bv0/WwwcKGESr5TFr7+WR9O5axzbFPdYHdAXXijH7CuvOMvSXeK4USPgyiv9jw3A\nce6moTOnyrPOvZ7jDsvEE/dY0121bSs/eHV1/B/ez7mvXCmi4b4oxHLuhYVyITHL0e7YkZqwDBAs\nLFMfufZaSePzu8B27y6/1YIFTondeJjinm7nXlDgXHCyGe6KVYZgwgQRKzM7qXt36e8I4twBuRgc\nfriEZvTuIJ1Fw4Ki4m4aOhXm1asl9JRIG624Zwh3WMYv5s4czLkD4jri1VLv0EHcg9u5a0eomyOP\nBI47LvokUoiic91TNYBJ0f3IVgGjRCkoiH1B0u9n9uzgI3lVoJo1y4z46HeeTeeuJY/d4s4s4n7s\nsdEd9wUFkg6r4l5bG3tiE0A6VufMcfozMj05iRd6N28aOhXmRZGyhok49xNPlHBQtqqmBiXvxN3P\nua9fLy4/lribk2rEc+4FBXKL53bufrHyfv3kRPLbpinuGrts6M49Hvr9zJ8fXNzVuSdSeiAR6oO4\nt2wpcXS3uM+fL8fa8cfXfY+ZDjljhmQXxQpjnXeeuP1DD5XPmzgxfN2eVKOZYqa4N2kiv4WKeyIX\noGOOAd59N9idYjbJYop9atCwTLyYu8Yb44VlAIm7B4nHdeniXTbg2GPjNttzW1pYLB3OvUmT5CtM\n1jf0+6muDi4kKu7pjrcrKu7ZdrEDB9YdbDRhgjyaGWNKcbGUoli8GDj9dHGp55zjv/127SQsowOi\nAOCss5JvdzIQyTy4gwdHL+/Y0Rl7kIhzzxVyXtzdzl1F3k/cg4RlVNzjOd3OnaOdu9Z5TsRxd+3q\nzF25YoUTqkkFPXrIqMNs14tJNZ06STy+ujox554J6oNzB2SiirfeElHTcsUTJsj54DUiubhYzq2j\nj5b+n88/jx+G+OlPo8eN1Ae8xm106OAM6sr2RTed5F1YpqBAOlLcMXetfxFU3OPF3IG6zj2ZejBd\nu0o4aOdO2U7nzlGFL5Pi7rvjz/uaizRq5AhO2Jh7pp17tsX9wgvleHr0UXleUyNmYtgw74u+pkNu\n2AD85z/efUW5SseOzh1/Pjv3nBd3d1gG8K4MuXSpXKVj3b6HDcuoc9cMgWTFHZCLRSpz3AGJPeZa\nZ2pQ9HsKKu7NmombO9lzCvfUU1/EvVs34PzzgeefF8GePl2Oc6+QDCB574MGSWmBbJZ/TgfaqQrk\nt7jnRVhmjz2iOzf8xL1nz9ihiTAdqoC4v6oqp5Z4MsW+zFz3lStjj7yzOIQVdwB44430tMWL+hJz\nB4CbbwZeeAH461+du8LjjvNet317mZAmHzHFvT78Luki5527Wctd8RJ3s+ynH61aya3+pk3BwjLu\nXPdkxN0cpZqq0gMNAf2esp2Z4Yf222TbuQMSWjn1VOCJJ5xQS6bCU/UJsxhfPjv3nBf37dvrzozj\nNY9qvAFMgDOydM0aiUkGce6AI+4rV4pTCzpTj4mKe1mZ3AmkMiyTzyTi3DOJdl7WlwEvt94qacFf\nfOEfksl3GkpYJufFPYhz37xZsljiOXdAxH35cvk/SMwdcDpVk3HcWo5WU8mscw9GfRf3I4+UDJVs\nTrdmctRRTmpgQxV3de7Nm2e33nq6aRDiHiQNUikqcsQ9SLYMEB2WSdRxFxRIap+Ku3XuwTjsMEnz\n0yH29ZF4d4yZhAi47z7pLE1kPEY+oM49n+PtQB6Iu1dYxk/cg5xkYZx7u3Yiyurcky321bWrM3LO\ninswevSQC2J9Hwpenzj+eOkszeeQRCzUuef7/ue8uHs5d3fMPYxzb9vWeW+8H79RIycdsqpKHpMJ\np5iDlmxYxmJJD9a55whBwjJLlkh8LciUdWbWRZAre+fO4ty1Lkyyzh2Q0FAinbIWiyU+bdpIKqh1\n7vUcv7DM9u2S8QIEy3FXTHEPUoulSxdx7KmoB6MxfOvaLZb0QSTu3Yp7PcfPuQPO6NV4pX5NzKyL\nMM49mRx3RZ27jbdbLOnl9NPzv0M55xOB/GLu+lrr1hKWOfjgYNsLG5bp0gVYu9bphE1FWMaKu8WS\nXp55JtstSD857dyZ/cMygMTdKytl0EbQdLSwYZnOnYHdu4HvvpOLSjIjEVXcbVjGYrEkS06L+86d\nIvB+YZnJk4Gzz5b/99032DZV3Js2lb94aJx82jQR5WTK6vbsKRk4++yT+DYsFosFyPGwjLvcr6Li\nfsklkov+yCOxZ0k3UXEP2tmio1TnzQOGDg32Hj/23FPuAOrLaEaLxZK75IW4u8MyJSUyae/ZZwO3\n3x6uqJR2qAadtUide21tamLlJSXJb8NisVhyWtw1G8bt3Lt1AxYsSGybiTp3/VyLxWKpD+R0zN0v\nLJMMYcW9qMipjW2zXCwWS30hL8Q9laM5mzcPN3qNyHHvVtwtFkt9IZC4E9HJRDSfiMqI6HaP1y8h\nonVENCPyd3nqmyrMng389rcyU5JfWCYZiMSNB425A3ZkqcViqX/EjbkTUQGAJwGcAGAFgClENJaZ\n57pWfYOZR6ehjVF88AFwzz3A00872SmpFHdAtjtoUPD1rXO3WCz1jSAdqoMBlDHzIgAgotcBnAHA\nLe4Z4ZZbgKOPlhllxoyRZakushV2js0uXSSUE6QwmcVisWSCIOLeDcBy4/kKAId6rHcOER0NYAGA\nG5h5uXsFIroSwJUA0KNHj/CtjXDoocB//yvzQH79tTOVWba48kqZLKJRTvdgWCyWfIKYOfYKROcC\nOJmZL488vxDAoWYIhojaA9jGzLuI6CoAP2Vmn3nVhdLSUp6q0w5ZLBaLJRBENI2ZS+OtF8RrrgRg\nTu/bPbLsR5h5AzPvijz9O4BDgjbUYrFYLKkniLhPAdCXiHoTUVMAIwGMNVcgImMOIYwAMC91TbRY\nLBZLWOLG3Jm5mohGA/gQQAGA55l5DhHdA2AqM48FcB0RjQBQDWAjgEvS2GaLxWKxxCFuzD1d2Ji7\nxWKxhCeVMXeLxWKx5BhW3C0WiyUPseJusVgseYgVd4vFYslDstahSkTrACxN8O0dAKxPYXNyhYa4\n3w1xn4GGud8NcZ+B8Pvdk5njFjvJmrgnAxFNDdJbnG80xP1uiPsMNMz9boj7DKRvv21YxmKxWPIQ\nK+4Wi8WSh+SquD+b7QZkiYa43w1xn4GGud8NcZ+BNO13TsbcLRaLxRKbXHXuFovFYomBFXeLxWLJ\nQ3JO3ONN1p0PENFeRDSJiOYS0Rwi+mVkeTsi+piIFkYei7Ld1lRDRAVENJ2I3o88701E30R+7zci\nZafzCiJqS0RvEdH3RDSPiA5vIL/1DZHjezYRvUZEzfLt9yai54loLRHNNpZ5/rYkPBHZ91lEdHAy\nn51T4m5M1j0cQAmAUURUkt1WpYVqADcxcwmAwwBcG9nP2wFMYOa+ACZEnucbv0T0fAAPAHiMmYsB\nbAJwWVZalV4eBzCemfcFMBCy/3n9WxNRNwDXAShl5v0g5cRHIv9+7xcAnOxa5vfbDgfQN/J3JYCn\nkvngnBJ3GJN1M3MVAJ2sO69g5tXM/G3k/wrIyd4Nsq8vRlZ7EcCZ2WlheiCi7gBOhczmBSIiAMcB\neCuySj7ucyGAowE8BwDMXMXMm5Hnv3WExgCaE1FjAC0ArEae/d7M/BlkjgsTv9/2DAAvsfA1gLau\niZBCkWvi7jVZd7cstSUjEFEvAAcB+AZAZ2ZeHXlpDYDOWWpWuvgTgFsB1EaetwewmZmrI8/z8ffu\nDWAdgH9EwlF/J6KWyPPfmplXAngYwDKIqG8BMA35/3sD/r9tSvUt18S9QUFErQD8C8D1zLzVfI0l\nhzVv8liJ6DQAa5l5WrbbkmEaAzgYwFPMfBCASrhCMPn2WwNAJM58BuTitieAlqgbvsh70vnb5pq4\nx52sO18goiYQYX+Fmd+OLC7X27TI49pstS8NDAEwgoiWQMJtx0Fi0W0jt+1Afv7eKwCsYOZvIs/f\ngoh9Pv/WAHA8gMXMvI6ZdwN4G3IM5PvvDfj/tinVt1wT97iTdecDkVjzcwDmMfOjxktjAVwc+f9i\nAO9lum3pgpl/xczdmbkX5HedyMznA5gE4NzIanm1zwDAzGsALCeifpFFwwDMRR7/1hGWATiMiFpE\njnfd77z+vSP4/bZjAVwUyZo5DMAWI3wTHmbOqT8ApwBYAOAHAHdkuz1p2scjIbdqswDMiPydAolB\nTwDw/+3cPWqCQRSG0ZPKOllCVuASspesJGuxsEiTwtIVBDtTBKIrsbH4bANBkI+Mz6mmmztceGF+\nmB9s8TR3rTda/ws2l/EzPnHAOxZz13eD9S6xu/T7A4/30Gu84RtfWGExWr+xNt0pnEy7tNffeosH\n02vAI/aml0RXz933A0kyoP92LJMk+YPCPUkGVLgnyYAK9yQZUOGeJAMq3JNkQIV7kgzoDEMSNKiZ\ntwgwAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAEICAYAAAB/Dx7IAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ntyAIQj1DhFsQBKGe8f/8ywkGS+yrfQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}