MNIST_CNN_Remove_Noise.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "MNIST CNN Remove Noise.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bS-7UEeGsQH7"
      },
      "source": [
        "#Section 3\n",
        "\n",
        "\n",
        "---"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ia6vseNSqGlM"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from keras.utils import np_utils, to_categorical\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Conv2D , Activation, Flatten , Dense, MaxPooling2D, UpSampling2D\n",
        "from tensorflow.keras.layers import Dropout\n",
        "\n",
        "from keras.layers.normalization import BatchNormalization\n",
        " \n",
        "from keras.constraints import max_norm\n",
        "\n",
        "from keras.datasets import mnist "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Qg2XHNfIqOiu"
      },
      "source": [
        "def load_data():\n",
        "    \n",
        "    return mnist.load_data()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nQYDghHoqdXo"
      },
      "source": [
        "def prepare_data(train_images, train_labels, test_images, test_labels):\n",
        "\n",
        "    x_train = train_images.reshape(60000,28,28,1)\n",
        "\n",
        "    x_test = test_images.reshape(10000,28,28,1)\n",
        "\n",
        "    #one hot encoding\n",
        "    y_train =  to_categorical(train_labels) \n",
        "    y_test = to_categorical(test_labels)\n",
        "\n",
        "    return (x_train, y_train, x_test, y_test )"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Og3r6NdLWEel"
      },
      "source": [
        "def add_noise(x_train, x_test, noise_factor_train, noise_factor_test):\n",
        "\n",
        "    x_train_noisy = x_train + noise_factor_train * np.random.normal(loc=0.0, scale=1.0, size=x_train.shape) \n",
        "    x_test_noisy = x_test + noise_factor_test * np.random.normal(loc=0.0, scale=1.0, size=x_test.shape) \n",
        "\n",
        "    x_train_noisy = np.clip(x_train_noisy, 0., 1.)\n",
        "    x_test_noisy = np.clip(x_test_noisy, 0., 1.)\n",
        "\n",
        "    return x_train_noisy, x_test_noisy"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MWcd-zOmXIbu"
      },
      "source": [
        "def model(x_train, y_train, x_test, y_test):\n",
        "    \n",
        "    model = Sequential()\n",
        "    #encoder\n",
        "    model.add(Conv2D(16, kernel_size=3, activation='relu',padding='same', input_shape=(28,28,1)))\n",
        "    model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "    model.add(Conv2D(8, kernel_size=3, padding='same',activation='relu'))\n",
        "    model.add(MaxPooling2D(pool_size=(2,2),padding='same'))\n",
        "    model.add(Conv2D(8, kernel_size=3,padding='same', activation='relu'))\n",
        "    model.add(MaxPooling2D(pool_size=(2,2), padding='same'))\n",
        "\n",
        "    #decoder\n",
        "    model.add(Conv2D(8, kernel_size=3, activation='relu',padding='same'))\n",
        "    model.add(UpSampling2D(size=(2,2)))\n",
        "    model.add(Conv2D(8, kernel_size=3, padding='same',activation='relu'))\n",
        "    model.add(UpSampling2D(size=(2,2)))\n",
        "    model.add(Conv2D(16, kernel_size=3,padding='same', activation='relu'))\n",
        "    model.add(UpSampling2D(size=(2,2)))\n",
        "\n",
        "    model.add(Flatten())\n",
        "    model.add(Dense(64, activation='relu'))\n",
        "    model.add(Dense(10, activation='softmax'))\n",
        "\n",
        "    model.summary()\n",
        "\n",
        "    model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "\n",
        "    history = model.fit(x_train, y_train,\n",
        "            batch_size=2048, epochs=50 , verbose=1,\n",
        "            validation_data=(x_train, y_train))\n",
        "    \n",
        "    score = model.evaluate(x_test, y_test)\n",
        "    print('Test loss :', score[0])\n",
        "    print('Test accuracy:', score[1] * 100)\n",
        "\n",
        "    return history"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "esTYYnLclKqq"
      },
      "source": [
        "def show_result(x_test_noisy, test_images):\n",
        "    n = 10\n",
        "    for i in range(n):\n",
        "        # Display original\n",
        "        ax = plt.subplot(2, n, i + 1)\n",
        "        plt.imshow(test_images[i].reshape(28, 28))\n",
        "        plt.gray()\n",
        "        ax.get_xaxis().set_visible(False)\n",
        "        ax.get_yaxis().set_visible(False)\n",
        "\n",
        "        # Display noisy \n",
        "        ax = plt.subplot(2, n, i + 1)\n",
        "        plt.imshow(x_test_noisy[i].reshape(28, 28))\n",
        "        plt.gray()\n",
        "        ax.get_xaxis().set_visible(False)\n",
        "        ax.get_yaxis().set_visible(False)\n",
        "\n",
        "    plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3Q044s_9Zqvq"
      },
      "source": [
        "def plot(history):\n",
        "    plt.plot(history.history['accuracy'])\n",
        "    plt.plot(history.history['val_accuracy'])\n",
        "    plt.title('model accuracy')\n",
        "    plt.ylabel('accuracy')\n",
        "    plt.xlabel('epoch')\n",
        "    plt.legend(['train', 'test'], loc='upper left')\n",
        "    plt.show()\n",
        "    # summarize history for loss\n",
        "    plt.plot(history.history['loss'])\n",
        "    plt.plot(history.history['val_loss'])\n",
        "    plt.title('model loss')\n",
        "    plt.ylabel('loss')\n",
        "    plt.xlabel('epoch')\n",
        "    plt.legend(['train', 'test'], loc='upper left')\n",
        "    plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m8ABVKs_qPNJ"
      },
      "source": [
        "def main(noise_factor_train = 0.4, noise_factor_test = 0.7):\n",
        "\n",
        "    (train_images, train_labels), (test_images, test_labels) = load_data()\n",
        "\n",
        "    x_train, y_train, x_test, y_test = prepare_data(train_images, train_labels, test_images, test_labels)\n",
        "    \n",
        "    x_train_noisy, x_test_noisy = add_noise(x_train, x_test, noise_factor_train, noise_factor_test)\n",
        "\n",
        "    history = model(x_train_noisy, y_train, x_test_noisy, y_test)\n",
        "    \n",
        "    show_result(x_train_noisy, x_test_noisy)\n",
        "\n",
        "    plot(history)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "sD1Lh8lFVyyM",
        "outputId": "f81928f8-7478-453a-969b-e96e1d233f2b"
      },
      "source": [
        "main(noise_factor_train = 0.3, noise_factor_test = 0.3)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_8\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_36 (Conv2D)           (None, 28, 28, 16)        160       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_24 (MaxPooling (None, 14, 14, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_37 (Conv2D)           (None, 14, 14, 8)         1160      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_25 (MaxPooling (None, 7, 7, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_38 (Conv2D)           (None, 7, 7, 8)           584       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_26 (MaxPooling (None, 4, 4, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_39 (Conv2D)           (None, 4, 4, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_6 (UpSampling2 (None, 8, 8, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_40 (Conv2D)           (None, 8, 8, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_7 (UpSampling2 (None, 16, 16, 8)         0         \n",
            "_________________________________________________________________\n",
            "conv2d_41 (Conv2D)           (None, 16, 16, 16)        1168      \n",
            "_________________________________________________________________\n",
            "up_sampling2d_8 (UpSampling2 (None, 32, 32, 16)        0         \n",
            "_________________________________________________________________\n",
            "flatten_2 (Flatten)          (None, 16384)             0         \n",
            "_________________________________________________________________\n",
            "dense_4 (Dense)              (None, 64)                1048640   \n",
            "_________________________________________________________________\n",
            "dense_5 (Dense)              (None, 10)                650       \n",
            "=================================================================\n",
            "Total params: 1,053,530\n",
            "Trainable params: 1,053,530\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 1.4555 - accuracy: 0.5314 - val_loss: 0.8159 - val_accuracy: 0.7308\n",
            "Epoch 2/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.5908 - accuracy: 0.8090 - val_loss: 0.4589 - val_accuracy: 0.8540\n",
            "Epoch 3/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.3772 - accuracy: 0.8798 - val_loss: 0.3517 - val_accuracy: 0.8861\n",
            "Epoch 4/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2912 - accuracy: 0.9071 - val_loss: 0.2654 - val_accuracy: 0.9149\n",
            "Epoch 5/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2493 - accuracy: 0.9207 - val_loss: 0.2384 - val_accuracy: 0.9239\n",
            "Epoch 6/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2206 - accuracy: 0.9296 - val_loss: 0.2004 - val_accuracy: 0.9367\n",
            "Epoch 7/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1956 - accuracy: 0.9370 - val_loss: 0.1968 - val_accuracy: 0.9359\n",
            "Epoch 8/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1848 - accuracy: 0.9404 - val_loss: 0.1701 - val_accuracy: 0.9450\n",
            "Epoch 9/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1637 - accuracy: 0.9481 - val_loss: 0.1548 - val_accuracy: 0.9498\n",
            "Epoch 10/50\n",
            "30/30 [==============================] - 67s 2s/step - loss: 0.1567 - accuracy: 0.9502 - val_loss: 0.1418 - val_accuracy: 0.9548\n",
            "Epoch 11/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1420 - accuracy: 0.9544 - val_loss: 0.1538 - val_accuracy: 0.9508\n",
            "Epoch 12/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1393 - accuracy: 0.9551 - val_loss: 0.1287 - val_accuracy: 0.9592\n",
            "Epoch 13/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1246 - accuracy: 0.9607 - val_loss: 0.1155 - val_accuracy: 0.9635\n",
            "Epoch 14/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1172 - accuracy: 0.9633 - val_loss: 0.1115 - val_accuracy: 0.9646\n",
            "Epoch 15/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1125 - accuracy: 0.9644 - val_loss: 0.1056 - val_accuracy: 0.9662\n",
            "Epoch 16/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1101 - accuracy: 0.9638 - val_loss: 0.1105 - val_accuracy: 0.9645\n",
            "Epoch 17/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1058 - accuracy: 0.9658 - val_loss: 0.1000 - val_accuracy: 0.9675\n",
            "Epoch 18/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1038 - accuracy: 0.9672 - val_loss: 0.0935 - val_accuracy: 0.9701\n",
            "Epoch 19/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0942 - accuracy: 0.9704 - val_loss: 0.0918 - val_accuracy: 0.9708\n",
            "Epoch 20/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0940 - accuracy: 0.9703 - val_loss: 0.0950 - val_accuracy: 0.9697\n",
            "Epoch 21/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0914 - accuracy: 0.9713 - val_loss: 0.1011 - val_accuracy: 0.9671\n",
            "Epoch 22/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0878 - accuracy: 0.9719 - val_loss: 0.0794 - val_accuracy: 0.9748\n",
            "Epoch 23/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0819 - accuracy: 0.9738 - val_loss: 0.0747 - val_accuracy: 0.9767\n",
            "Epoch 24/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0809 - accuracy: 0.9740 - val_loss: 0.0759 - val_accuracy: 0.9762\n",
            "Epoch 25/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0789 - accuracy: 0.9744 - val_loss: 0.0777 - val_accuracy: 0.9752\n",
            "Epoch 26/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0746 - accuracy: 0.9761 - val_loss: 0.0674 - val_accuracy: 0.9785\n",
            "Epoch 27/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0706 - accuracy: 0.9774 - val_loss: 0.0743 - val_accuracy: 0.9762\n",
            "Epoch 28/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0721 - accuracy: 0.9771 - val_loss: 0.0719 - val_accuracy: 0.9772\n",
            "Epoch 29/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0677 - accuracy: 0.9786 - val_loss: 0.0638 - val_accuracy: 0.9792\n",
            "Epoch 30/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0642 - accuracy: 0.9796 - val_loss: 0.0592 - val_accuracy: 0.9811\n",
            "Epoch 31/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0620 - accuracy: 0.9804 - val_loss: 0.0605 - val_accuracy: 0.9802\n",
            "Epoch 32/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0621 - accuracy: 0.9799 - val_loss: 0.0620 - val_accuracy: 0.9799\n",
            "Epoch 33/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0624 - accuracy: 0.9801 - val_loss: 0.0528 - val_accuracy: 0.9837\n",
            "Epoch 34/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0561 - accuracy: 0.9822 - val_loss: 0.0572 - val_accuracy: 0.9818\n",
            "Epoch 35/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0584 - accuracy: 0.9815 - val_loss: 0.0600 - val_accuracy: 0.9799\n",
            "Epoch 36/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0571 - accuracy: 0.9816 - val_loss: 0.0506 - val_accuracy: 0.9842\n",
            "Epoch 37/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0532 - accuracy: 0.9834 - val_loss: 0.0525 - val_accuracy: 0.9830\n",
            "Epoch 38/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0517 - accuracy: 0.9833 - val_loss: 0.0486 - val_accuracy: 0.9844\n",
            "Epoch 39/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0507 - accuracy: 0.9838 - val_loss: 0.0456 - val_accuracy: 0.9852\n",
            "Epoch 40/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0479 - accuracy: 0.9847 - val_loss: 0.0469 - val_accuracy: 0.9855\n",
            "Epoch 41/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0481 - accuracy: 0.9845 - val_loss: 0.0487 - val_accuracy: 0.9844\n",
            "Epoch 42/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0464 - accuracy: 0.9848 - val_loss: 0.0448 - val_accuracy: 0.9857\n",
            "Epoch 43/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0439 - accuracy: 0.9858 - val_loss: 0.0441 - val_accuracy: 0.9856\n",
            "Epoch 44/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0441 - accuracy: 0.9857 - val_loss: 0.0430 - val_accuracy: 0.9861\n",
            "Epoch 45/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0425 - accuracy: 0.9864 - val_loss: 0.0378 - val_accuracy: 0.9877\n",
            "Epoch 46/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0421 - accuracy: 0.9868 - val_loss: 0.0428 - val_accuracy: 0.9857\n",
            "Epoch 47/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0404 - accuracy: 0.9870 - val_loss: 0.0462 - val_accuracy: 0.9839\n",
            "Epoch 48/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0418 - accuracy: 0.9862 - val_loss: 0.0441 - val_accuracy: 0.9856\n",
            "Epoch 49/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0371 - accuracy: 0.9883 - val_loss: 0.0397 - val_accuracy: 0.9870\n",
            "Epoch 50/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0361 - accuracy: 0.9881 - val_loss: 0.0358 - val_accuracy: 0.9881\n",
            "313/313 [==============================] - 4s 14ms/step - loss: 0.0776 - accuracy: 0.9767\n",
            "Test loss : 0.07757900655269623\n",
            "Test accuracy: 97.67000079154968\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9YjrYk42esla",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "463381d2-1b3d-4691-863b-89ed9cc5810a"
      },
      "source": [
        "main(noise_factor_train = 0.5, noise_factor_test = 0.5)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_10\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_48 (Conv2D)           (None, 28, 28, 16)        160       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_30 (MaxPooling (None, 14, 14, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_49 (Conv2D)           (None, 14, 14, 8)         1160      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_31 (MaxPooling (None, 7, 7, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_50 (Conv2D)           (None, 7, 7, 8)           584       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_32 (MaxPooling (None, 4, 4, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_51 (Conv2D)           (None, 4, 4, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_12 (UpSampling (None, 8, 8, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_52 (Conv2D)           (None, 8, 8, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_13 (UpSampling (None, 16, 16, 8)         0         \n",
            "_________________________________________________________________\n",
            "conv2d_53 (Conv2D)           (None, 16, 16, 16)        1168      \n",
            "_________________________________________________________________\n",
            "up_sampling2d_14 (UpSampling (None, 32, 32, 16)        0         \n",
            "_________________________________________________________________\n",
            "flatten_4 (Flatten)          (None, 16384)             0         \n",
            "_________________________________________________________________\n",
            "dense_8 (Dense)              (None, 64)                1048640   \n",
            "_________________________________________________________________\n",
            "dense_9 (Dense)              (None, 10)                650       \n",
            "=================================================================\n",
            "Total params: 1,053,530\n",
            "Trainable params: 1,053,530\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 1.3472 - accuracy: 0.5620 - val_loss: 0.6330 - val_accuracy: 0.7966\n",
            "Epoch 2/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.4858 - accuracy: 0.8466 - val_loss: 0.3751 - val_accuracy: 0.8825\n",
            "Epoch 3/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.3373 - accuracy: 0.8939 - val_loss: 0.2870 - val_accuracy: 0.9100\n",
            "Epoch 4/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.2684 - accuracy: 0.9163 - val_loss: 0.2370 - val_accuracy: 0.9256\n",
            "Epoch 5/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.2319 - accuracy: 0.9269 - val_loss: 0.2103 - val_accuracy: 0.9337\n",
            "Epoch 6/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.2006 - accuracy: 0.9362 - val_loss: 0.1877 - val_accuracy: 0.9406\n",
            "Epoch 7/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1795 - accuracy: 0.9438 - val_loss: 0.1658 - val_accuracy: 0.9483\n",
            "Epoch 8/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1661 - accuracy: 0.9484 - val_loss: 0.1508 - val_accuracy: 0.9528\n",
            "Epoch 9/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1529 - accuracy: 0.9522 - val_loss: 0.1384 - val_accuracy: 0.9572\n",
            "Epoch 10/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1418 - accuracy: 0.9559 - val_loss: 0.1405 - val_accuracy: 0.9553\n",
            "Epoch 11/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1358 - accuracy: 0.9564 - val_loss: 0.1359 - val_accuracy: 0.9567\n",
            "Epoch 12/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1269 - accuracy: 0.9590 - val_loss: 0.1348 - val_accuracy: 0.9568\n",
            "Epoch 13/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1191 - accuracy: 0.9631 - val_loss: 0.1154 - val_accuracy: 0.9638\n",
            "Epoch 14/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1125 - accuracy: 0.9645 - val_loss: 0.1065 - val_accuracy: 0.9665\n",
            "Epoch 15/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1098 - accuracy: 0.9649 - val_loss: 0.1142 - val_accuracy: 0.9638\n",
            "Epoch 16/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1078 - accuracy: 0.9659 - val_loss: 0.1037 - val_accuracy: 0.9674\n",
            "Epoch 17/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1039 - accuracy: 0.9667 - val_loss: 0.1040 - val_accuracy: 0.9668\n",
            "Epoch 18/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0963 - accuracy: 0.9692 - val_loss: 0.0896 - val_accuracy: 0.9719\n",
            "Epoch 19/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0936 - accuracy: 0.9707 - val_loss: 0.0930 - val_accuracy: 0.9706\n",
            "Epoch 20/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0898 - accuracy: 0.9708 - val_loss: 0.0825 - val_accuracy: 0.9734\n",
            "Epoch 21/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0858 - accuracy: 0.9725 - val_loss: 0.0819 - val_accuracy: 0.9733\n",
            "Epoch 22/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0857 - accuracy: 0.9726 - val_loss: 0.0847 - val_accuracy: 0.9724\n",
            "Epoch 23/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0856 - accuracy: 0.9723 - val_loss: 0.0833 - val_accuracy: 0.9731\n",
            "Epoch 24/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0821 - accuracy: 0.9735 - val_loss: 0.0756 - val_accuracy: 0.9758\n",
            "Epoch 25/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0739 - accuracy: 0.9762 - val_loss: 0.0710 - val_accuracy: 0.9771\n",
            "Epoch 26/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0723 - accuracy: 0.9768 - val_loss: 0.0675 - val_accuracy: 0.9784\n",
            "Epoch 27/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0718 - accuracy: 0.9765 - val_loss: 0.0703 - val_accuracy: 0.9769\n",
            "Epoch 28/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0703 - accuracy: 0.9774 - val_loss: 0.0661 - val_accuracy: 0.9783\n",
            "Epoch 29/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0653 - accuracy: 0.9786 - val_loss: 0.0621 - val_accuracy: 0.9800\n",
            "Epoch 30/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0619 - accuracy: 0.9798 - val_loss: 0.0616 - val_accuracy: 0.9798\n",
            "Epoch 31/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0604 - accuracy: 0.9800 - val_loss: 0.0714 - val_accuracy: 0.9767\n",
            "Epoch 32/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0616 - accuracy: 0.9797 - val_loss: 0.0576 - val_accuracy: 0.9813\n",
            "Epoch 33/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0586 - accuracy: 0.9809 - val_loss: 0.0654 - val_accuracy: 0.9783\n",
            "Epoch 34/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0604 - accuracy: 0.9798 - val_loss: 0.0557 - val_accuracy: 0.9823\n",
            "Epoch 35/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0540 - accuracy: 0.9824 - val_loss: 0.0506 - val_accuracy: 0.9838\n",
            "Epoch 36/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0509 - accuracy: 0.9838 - val_loss: 0.0476 - val_accuracy: 0.9853\n",
            "Epoch 37/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0512 - accuracy: 0.9839 - val_loss: 0.0465 - val_accuracy: 0.9847\n",
            "Epoch 38/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0483 - accuracy: 0.9844 - val_loss: 0.0453 - val_accuracy: 0.9850\n",
            "Epoch 39/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0486 - accuracy: 0.9841 - val_loss: 0.0511 - val_accuracy: 0.9829\n",
            "Epoch 40/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0477 - accuracy: 0.9841 - val_loss: 0.0467 - val_accuracy: 0.9845\n",
            "Epoch 41/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0470 - accuracy: 0.9845 - val_loss: 0.0572 - val_accuracy: 0.9798\n",
            "Epoch 42/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0466 - accuracy: 0.9848 - val_loss: 0.0418 - val_accuracy: 0.9866\n",
            "Epoch 43/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0429 - accuracy: 0.9858 - val_loss: 0.0368 - val_accuracy: 0.9887\n",
            "Epoch 44/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0412 - accuracy: 0.9868 - val_loss: 0.0401 - val_accuracy: 0.9867\n",
            "Epoch 45/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0429 - accuracy: 0.9861 - val_loss: 0.0362 - val_accuracy: 0.9883\n",
            "Epoch 46/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0370 - accuracy: 0.9885 - val_loss: 0.0362 - val_accuracy: 0.9887\n",
            "Epoch 47/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0361 - accuracy: 0.9886 - val_loss: 0.0345 - val_accuracy: 0.9891\n",
            "Epoch 48/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0350 - accuracy: 0.9890 - val_loss: 0.0346 - val_accuracy: 0.9890\n",
            "Epoch 49/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0350 - accuracy: 0.9890 - val_loss: 0.0389 - val_accuracy: 0.9876\n",
            "Epoch 50/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0343 - accuracy: 0.9889 - val_loss: 0.0296 - val_accuracy: 0.9908\n",
            "313/313 [==============================] - 5s 15ms/step - loss: 0.0788 - accuracy: 0.9765\n",
            "Test loss : 0.07884909957647324\n",
            "Test accuracy: 97.64999747276306\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAACkCAYAAACpb/RaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9d1SU1/Y//JnegBmYobcRBkREQBiVACpEUAlYiD2WhKtJdGkUrxr1amKMSfwmxn5jjDEmxtiuLZYYYmwRU2xYMCoqKhCKIEjv8Pn94TvPciJjbt71W+/3j9e91rPEmfOcZz/77LPLZ59zRkQSz+gZPaNn9Iz+vyfx/zYDz+gZPaNn9P9XemaAn9EzekbP6H+JnhngZ/SMntEz+l+iZwb4GT2jZ/SM/pfomQF+Rs/oGT2j/y0i+V9fer2e3t7eBEAAVCqVDA4Opkwmo6urK3U6HaVSKe3t7SmXy4V2j19OTk7s3Lmzze8tl4ODA5VKJTvi4/F2YrGYAKjVaikSiQiAcrmcIpGIdnZ29PT0pIODA0UiEd3d3enp6dnh88RiMe3t7QmAEomEAKhWqwmACoXCJh/e3t5UKpXUaDQEQL1eT1dXV2q1WorFYmo0GioUCoE3sVhMrVZLBwcHq+eLRCLhuZZ2Li4uVm2kUmmHfPy5L8sll8vp6upKjUZDqVT6xPcSiYSenp4MCQmhs7MzJRIJtVotDQYDFQoFvby8nujP8ndHfGg0mifka2dnR5FIRAcHB0okEtrb29Pe3p7u7u709vYWvn987FxcXOjh4SHIQywWU6fTUSQS0c3NjXq9np06dbLJh+XZzs7OfPxvqVRqJePHL4VCIbyfWCy2ave4Lv35Pn9/f5t8ODo6UqVSUSaTCfPl8Xu7dOlCOzs7SqVSOjs708PDg2q1mjKZzOq+x/n482cABL39K3lYLpVKRTc3N2HuAKCvry81Gg39/PzYqVMnBgQE0M3NTRhDg8Eg6JNFvxUKhcCXSqX6W3zIZDJ2796dKpXKSlckEgnFYrEVbxKJhBKJhAqFgnZ2dvxzP5a5p1arqdfrCYCdOnWiSCT6Sz58fHwE/r28vIR5otFo6Ovra/U8hUJBpVL5hH5IJBLqdDqBBz8/P8EmhYSE2JQHyb8XAZeXl6OgoACTJk2CRCJBY2Mjrl+/jl69eqGmpga1tbVITk5GSkoKXFxcrO6NiopCcnIyKioqcPPmTTQ3NwMARCKRVTuJRIKRI0eipqYGKpWqQz6USqXwt1gsxtatWwEAMTExOHfuHAICAjB16lT4+/ujpKQE1dXVIIni4mIUFhZCJBLB0dER3t7eAACNRoPQ0FA4OTmhrKwMAQEBAID6+nrI5XI0NTV1yIednR3++OMPREVFoaGhAdOnT8e9e/cgl8tRVVWF0NBQREdHo3v37tBqtTAajUhMTETnzp3R3t7+xDs999xzWLlyJRYtWgSxWCzwJxaLsXLlSrS2tnbIR319PQBg6NChEIvF+Pjjj/HGG29g2LBhkMvlaGxs7PBeOzs7BAQEwGw2w9fXF21tbaiqqoJOp8PKlSvxxx9/wNXVFc3Nzfjhhx+EMbNFJpMJ69evF/gIDw9HY2PjI0UTP1K1mpoa1NTUoLi4GO3t7dDr9VCpVPDx8UFwcDA6deqEkSNHolu3blCpVOjZsyf0ej369esHkigrK0NoaKhlIj2VysrKMH/+fKSlpaGmpgaxsbEYOnQodDqd0CYtLQ1JSUmQyWTC+7W3tyMgIACpqakAYHFymD17NlQqFdLS0mA0GuHp6Ync3Fybz29paUFDQwMCAgIwfvx4zJo1C0ajEWKxGL6+vhgxYgQmTpwINzc3lJWVobGxEampqejbty8mT56MtWvXWvWn1WoxePBgDBw4UPhMLBajV69eWLFixVNlkZaWBpPJBA8PD0yePBkmkwmTJk2CTqeDm5sb+vbtC2dnZyQlJWHEiBGIjY3Fhg0bsGzZMri5uSEoKAjNzc1wdXXF4sWL4e3tjfDwcEyaNAkikQidOnXC7Nmz4enp+ZfjYpHNpUuXoFAoYDabMXbsWCgUCnTr1g0hISHQaDTo3LkzRowYgeeeew5KpRJmsxk7d+7E9u3bYWdnh6CgILS0tKCxsRH+/v4YOXIkqqurERcXB5lM9l/pSH5+Pvz9/ZGbm4vu3btjyJAhmDFjBnQ6HT799FMMHDgQMpkMvr6+SEtLw+DBg0ES7e3tiImJwZ49ezB58mRs2bIFX331Ferr61FcXIwlS5agqqoKd+7ceToDfycCtnhfNzc3jh49WvDGDg4Ogsd6PJqxXCKRiGKxmAqFgiEhIUxMTKRWq2VsbCyDgoLo6OjIJUuWWLV3d3enj4+PTQ82c+ZMJiUlUa/Xc8aMGVSpVFZRZEhICHU6nfDMxyN3qVRKpVJJk8nEMWPGCJ8ZjUbOnz+fa9asYb9+/SgSiZiWlka9Xt8hHxKJhBMnTuRbb73FgoICuru7UyqVUqFQMDExkTqdTvCiMTEx3LhxI0eNGmXlfSMiIgiAOp2OZrP5Ce8vlUo5a9YsSiQSGgyGDvlwdnYWIkT8PxGcSCSi0Wh8ImIymUx0d3dnUlISY2NjaTKZOHDgQCYlJXHixIlMSEigo6Mj+/fvTy8vL165coVarZb79u0T+rDFh1wup0QioUgkopeXF0NCQrh27Vo6OjoK9+r1eqakpNBoNLJnz55ctmwZx48fz6SkJM6fP59ms5l2dnbU6/WUSqUMCQnhypUrmZiYyK5du1KhUDAjI4NSqZQBAQH/VYbk4OBAvV5PrVbLoKAguri4cPny5bx16xYzMzPZr18/of2kSZO4fv16hoaGcvTo0dy3bx/feustymQyISMQiUSUy+VctWoVg4ODqdFoOuTD29ubISEhwpwYOHAg5XI5582bx2vXrjEkJIRyuZzOzs5UKpX8/PPPuXPnTkZGRtJoNHLq1Kl87bXXCICOjo6Mjo6ml5cX4+Li6OjoyO+++44Gg4Emk0nQmY74sLe35969eymRSBgVFcVjx47R29tbmCvTp0+n0Wjk8OHDGRQUxI0bNzI9PZ1ms5m9evWil5cXQ0NDGRgYyIkTJ1rp5+M6FxERwZiYGJt8dOnShe7u7k9E8Gazmc7OzuzcuTN9fHyEyFqj0VCpVArP8PLy4pgxYzhy5EjhXq1WK/zf0s7yr0QisakfEonEKsMrKyujt7c358+fT29vb7799tuMiIigQqFgVFQUAwMDGRUVxdjYWHbt2pX9+/cX5m1YWBjFYjEdHR35xhtvUKFQcN68eULUbEseJP+eAQbAqqoqSqVSisVi3r9/nwaDgVu2bBHSs4EDB/Kdd95hWFgYe/Xqxc6dOz8xGSypocVQ6PV6bt68mWfPnuWWLVvo6upKkUhER0fHDhnXarWUSqVctmwZvb29KZPJuGTJEiENSE9Pp5+fH/fv38+GhgY2NjaysbGRDQ0NwrV582YaDAYOGTJEMIoymYxRUVH86aefBEP49ttvs3v37jYH0sPDgz4+Phw2bBjNZrOQwojFYm7YsIGRkZEUi8VMS0tjcnKy4HTs7e3Zq1cv9unThw4ODjSbzRw7dixXrFhBT09P3rlzh5s2bSLwKC1PTk5+amr32muvUSQS8eHDh1yyZAnVajVTUlJob2/P6OhoOjo6ct++fTxz5gw//PBDjh07lomJifT396ezszONRiNTUlJ49OhRqlQq9uvXj1u2bOH7779PuVzOtLQ0q0nzV4YvKCiIEomkw5TZ29tbcEwWWMDd3Z1+fn4UiUR0cnLikCFDOG/ePKakpFAmkwmwklgsZp8+fQiArq6uf8kHAC5dupTl5eU8fPgwy8vLuXv3bsHQy+XyJ9LdcePG8eTJk0xLS6OHhwcHDBhAmUxGHx8fSiQShoeH88cff+T7778vwBMd8WEwGCgWizlr1iy+/fbbQrptcdKxsbFWzkkmk1Emk3HhwoX87bffOG7cOOF5H330EcPCwiiRSOjr68s33niDcrmccrmcmzZtYlxc3FMhuwkTJtDJyYnJyckCtHbr1i2aTCYBGrJAZTKZjLNnz+akSZPo6upKb29vqlQq6nQ6arVams1mJiUlcfr06VaG2Gg0PtURWAyjyWTiK6+8wpEjR1Kv1/POnTtctWoV3d3dOXfuXCFtHzhwICMjI62c9+PPsLOzo9lspouLC1evXk2RSGQFOz1NT9euXSs4IKVSSaVSycWLF/PBgwcMDw+nVCplUFAQMzIyOHHiRMHmWcZILBZTKpXSZDJx1KhRjIyM5OrVqxkeHs7FixcLdiA0NJRqtfr/jgFWqVQcOXKk4DkGDRpEuVxOlUrFiIgIqtVqSqVSGgwGJicnMygoiGKxmB4eHoyJiWFwcDANBgODgoKshKRQKBgfH8+BAwfSw8OD3t7eAq7TER8W3M2CU8bFxdFoNPLMmTMcPnw4nZyceOTIEba0tNAWNTc3s6ysjMePH+eQIUOoUCgYHBxMiUQiKKhlcDw8PDrkw93dnaNGjeLgwYMpl8vZs2dP9uvXj7m5uTSZTAJeZJlcFmMTHh7OQYMGcenSpayrq2N1dTWvXbvGGTNmMCoqisHBwbSzs6NarRYcm2VAO+LDgpkCYEVFBS9evCjg8QAYHh5OBwcHqtVqjh07lmfPnmV9fT3r6upYU1PD6upqbtmyRWij1+uZmprKwMBAFhQUUKPRCPjf0xTbwkNSUhKlUim/+uor2tvbU61W8+LFiwwNDeWgQYPo4+NDNzc3btiwgZMmTbLKnCzOLCYmht7e3pw6dSpDQkKEMZVKpZwyZYpwj60JtnjxYi5evJhffvkldTqdwMfy5csZGBhoZXTt7Ow4cOBA9uvXjyEhIUxLS+OECRPo6urK0NBQSiQSSqVS4f3FYjHVarUQ3diSh1KpZE5OzhNtFyxYQGdnZ/r6+lKlUjEkJIQffvgh4+PjBb1OSUnh2LFjhecplUqGhYXx559/ZmJiIuVyOe3t7XnhwgWqVCpKpVK6uLjYlIcFa581axa3bt3KWbNm0d3dnWKxmKNHj+bEiROtMO6QkBAGBgYyNDRUcHh6vZ79+/fnzJkz+fnnn9PX15ejR4/moEGDBOdliUaf5hg1Go1gyCwObOfOnTSZTFYOMSkpiX369GFMTAxTUlLo7e3NkSNHsqqqilVVVayurhb0t7m5mVVVVbx//77wDnK5vEM+LNi8pd27777LHTt28JVXXhFkkpGRwQEDBgiY/PDhw5mUlMQxY8Zw3rx5DA8PZ1paGl1dXZmamkqTyUSDwcCpU6dy3bp1lEqlnDx5MiUSic1x+dsG2JJ6WQS0du1awXBu27aN/v7+QgHOko5aFMrV1ZUGg4He3t4cMmQIR4wYwdmzZwvFtqSkJI4bN46xsbH08PAQotmO+IiIiGBsbKzgicaOHUuVSkVHR0cqFAoeP36czc3NNo2vhX744QcGBgZyxIgR7N+/PzMzMymXyzlo0CDm5ORYRe4d8WGBVeRyOWfOnCkYKr1ez549e3Lu3LlMTk6mTqdjbGysULCRSqWUy+WcPHmywEtmZiaDg4O5cuXKJ4pxFmMTGBhokw9LdFFfXy8UFi5evEh/f38uWbKEnTt35vbt27l69WpmZmY+IYvbt29z2rRpBEBPT0++8847HD16NE0mE3/88UeqVCoWFxc/1eCIxWK6uroKxsbBwYE//vgj/f396eHhIUSyO3fu5IMHD/j6669TqVRSJpPRaDQKhlahULBv37588OABb9y4wZiYGIpEIqtCR1xcnE0+VCoVp0+fTk9PT16+fFmQBwD27NmTdnZ2XLBgAYcMGSIUb5OSkhgfH0+j0cjly5ezqqqKDx48YEVFBX/88UchU0tISBDGJT4+XnB8tgzOjBkzuHnzZsbHxwvZgEaj4YgRI3jp0iWWlpayoqKC9fX1rKysFCCHAQMG8OWXX7YKVDQaDUeOHMnKykqWlpby7t27jI+P55YtW1haWmozUwsNDRXeX6VSMTY2lg4ODpw3bx6NRiN/++03xsXFcceOHVbG1BLlyWQyqtVqOjs7Uy6XMywsjN9++y3NZjNdXV25a9cupqSkcPr06UxKSrIpD61Wy6VLl/LUqVMsKipir169hIDExcWFb731ljBHAHDdunVcunQp165dK4xHVVXVX87roqIiurm52YSG3N3dmZiYSCcnJ0GuSUlJjIqK4vz585mUlEQfHx/K5XIOHjyYsbGx1Ov1lMvlVCqV9PLyYnBwMN3c3AQ4SiaT0c/Pj19//bVQuFOpVPT29rYZKPDvGmBLtd1sNjM3N5c+Pj40mUy0s7Njv379qNVqOW/ePIaFhQkD7u/vz/DwcOr1etrb29PFxYUqlYr+/v6cMGECFQoF/fz8qFQq6ebmxry8PMpkMr7xxhsMCgrqkHE7Ozt6eXkJEzIqKoofffQRdTodT5482aHxfeedd7hy5UqrzxoaGrhnzx6qVCoqlUoajUYmJydz/fr1zMjIENLHp02wzMxMSiQSYVVDTEwMN23aRKPRSAcHB6pUKgH/tkSkU6ZM4axZs/jqq68KvJw8eZL29vZMSEhgSkqKkPJbUu6YmBibqYzF26enp/Ozzz7j6tWrqdFomJqaSrPZzBUrVnDx4sXs2bMnd+zYwaampifkk5eXxzfffFOADzIzM/n+++/T0dGRBw8eZExMDGNjYwk8WlFgiw+Lwzhy5AgPHjzIzz77TFB0y3X48GGStII1QkNDOWXKFCYlJVGj0XDs2LEkyVu3btHT05N6vd4qc7KsKOiID5FIxNGjR1OtVtPNzY0LFixgWFgYpVIp3333XWZmZnLu3LlMTU1lUlISPT09GRsbS7PZzClTprCmpsZKNufPnxcw4MfT1djYWI4fP/6p+qHRaPjDDz9YrRCwzItr165ZPWf27Nn09fWlo6MjPTw8aDQa6eTkxKioKEZERDA0NJRlZWVC+8rKSspkMiGw0Wq1HfLxOH6dlJRElUrFy5cvMygoSIjszWYzJ0+ezBMnTjA/P1+4Dh48yKSkJLq5uTElJYUJCQmMiIjglStXBFjIYDDQzs6OaWlpVCqVVKlUNsdlzpw5dHd3p5eXl7Dq5OTJkxw6dCjd3d25f/9+Zmdnc8WKFfztt99YU1PzxHg8jRoaGujl5SXUIjriQ61WW2WnliDRycmJZrNZqOUsX76cycnJjIqKoslkorOzM11dXa3ghzFjxjA4OJj79++nVCqll5cXx48fT6lUyoULFwqrNzrig3/XAFsYtqQma9as4ccff8xLly4xLy+Pt27dYnl5OfPy8piTk8MxY8Zw1KhRTE5OFvBeS3RnSaE8PT25bt06YVJ98MEHnDx5Mp2cnOjg4GBzIC0e2iK81NRUqtVq5uTkWA3I9OnT+T//8z8cOXIkFy9ebPXd2bNnGRUVRYVCwfDwcL722mucOnUq4+Pj2bVrVyHtsqVQAGg0GoWIwZL6WhQbgDCBLOmrvb09vb29OWfOHJaUlJAkT5w4wejoaKalpVGn0zEmJoa9evUSvOjjy2ts8WGJ0GbOnMm0tDTGxsZSp9NRqVQyKCiISUlJPHDggE1lbmlpYVlZGTMzM5mens6goCAaDAaq1WpOnz6dHh4eVCgUHDJkyFOLG++//z7FYjF9fX3p4+PDsLAwKxx469atAg9paWmcOXMmJ06cSIVCQUdHR/r6+jIoKIgJCQkkH0FFp06dokaj4ZgxYyiXy/ndd99xxIgRNuXh4OBAnU7H69evU6FQ0GAwCAbbgnc7OjqyT58+7N+/P93c3GgymWg0Gjlr1qwnZHPx4kUBr7bI2s3Njf369aPZbLaJvVqe6eXlxVdeeYUymYxbt25lUlISf/nlFytHOH36dGGpXnp6OufPn0+JREKz2cxp06ZRqVQyMjJSaP/gwQOaTCaBn2HDhtmM+B6HsCzR2d69e7lu3Tq+9tprdHJyolKp5M6dO9nY2Gj17hcuXOC4ceNob29POzs7Ojo60s3NjZ07dxaeLxaLGRwcLAQhOp3OJh9Tpkyht7c3AwMDuXDhQqrVavr6+nLp0qWMiYnh0KFDGRsbSy8vL27ZsqVDXX0aNTc3c+bMmfTz87PpkCxzCQDj4+MFSColJYUrVqxgVFQUx4wZwyVLljA0NJRKpZKLFi1ibm4uc3NzmZOTw5ycHN69e5clJSU8ePAgfXx8qFarOWrUKE6ZMoVjxoyhs7Mz/f39bWbQ/LsG2DJ4Fs8XGxvL+Ph4lpaWdiiMgoIC5uXlMTc3l1lZWVbXxYsXmZWVxezsbN66dYv+/v7UaDR0d3fnggULeOHCBZsezMHBgQcPHuSRI0cEL9q1a1fOnz+fe/fuFdKUJUuWcN26dTx48CAHDx78hAE+fPiwUF2ePHkyXVxc6OzsTJVKJeDaIpGIXbt2talQb775Jvv37y8Y4MDAQKs1t0uXLuXw4cOp1WppMpk4ZcoUDhgwwGqiV1VVcdeuXTSbzdTr9Zw9ezaTkpL42WefUSqVMjo6mmKx2OZqjMe9eGhoKL///nt6enpyzJgxdHBwoFQqpUaj4ZEjR/5SgVtaWnj06FGh4NC1a1fq9XohorCsGumID41Gw8GDB3Pp0qXctm0bExMTrfD0hQsXsri4WHjWli1bOHDgQKv1zo+vRrHQvXv3KJPJGBsby9DQUA4cOPCpDskyFmPGjOGMGTMEuCczM5PLly9nZGQkFQoFNRoN7e3tBUho+vTpLCgosJLH+fPnaTabGRUVJRg7S0Rvb28vTN6O+HhcD5ydnanVahkeHk4nJyfeunXL6jnDhw+nn58fjUYjjUYjw8LC6OnpKeDQkZGRvHHjhtC+pKSEarWaO3bs4MSJEzls2DB6enra1I/x48cLRSqRSERfX1/OmDGDJpOJUqmUX3zxRYfpfX19PVesWCGsfY2KiuKGDRvo5+fH48ePC8VyBwcHvv3220IwYosPg8HAqKgoXrlyhePGjeOQIUOEVTnHjh3jzZs3mZ2dzVOnTrGwsPAv9fXOnTsMDw9neHg4e/bsyUWLFtHd3Z0KhcImH5ZxmTt3LmNiYrh+/Xra2dlx8uTJNJvNtLe3p8FgoJubG9esWcOsrCybNs4io6ysLF6+fJm///47R48ezdmzZzMqKoo//PAD3d3dO+SDf9cA/3kRu1KppFqtfkJp/y41NTXxu+++ExZcv/7664yNjX2qANetW8d9+/ZZLXK/dOkSExIShOUicXFxjI6O5rRp07h7927m5eVZPffgwYOUyWR0d3dnUFAQhw8fzu+++44RERGMj48XPLqtKrejoyP9/f1pMBj49ttv08XFRSiGWGQVFBTECRMm0GQyUa1W08fHhwaD4YlIKzMzU1h87+vrS2dnZyYkJFCj0QiKLZPJOuTD3t6eOp2OYrGYqampHDVqFBUKhZBKubi48PPPP2dFRcV/NR4PHz4UqsTu7u709fXl0KFD2bVrV6Eq3REfEomEy5cvZ2pqKr/88kvu2rWL48aNo0ql4qhRo3jt2jW2trYKz5kzZw7DwsJoNBqpVCppNps5YsQI4XkWunfvnhDhL126VIAibBVZLMvO1q5dS5PJJFToo6OjeebMGa5atYqpqal8PBLy8PDg+vXrreTw888/C5CQTqejRqPh6NGj2adPH86YMYOvv/46N2zYYFMeloDFos8JCQmcM2cOjxw5wvr6euE5GRkZjIyMpFqtZp8+fZiYmChs8OnTpw+///57XrlyRWjf2NjIb7/9lhKJhL169RIKl7YmOgDm5uZywoQJdHR0ZFxcHGfOnMno6GgqFAqbxtdCxcXFzMzMFGC+kJAQqlQqRkVF0c3NTcCJly5d+tQageU7BwcHxsTE0MnJiWPHjqVOp+PBgwfZ0NAgPHPx4sUcMmQIDx8+zDNnzgh1i6ysLKalpXH8+PE8e/YsL1y4IDxTLBbTaDRSp9Nx8eLFNvmwBG2WmtXo0aOp1WqZmppKrVZL4FF9KzMz0wry+W+ooaGBn3/+Ob28vOjo6Mj09PSnQhB/ayNGW1sb3N3d4ePjA5FIhMbGRtTX12P8+PEYM2YMPv30UyxcuBDDhw/H6dOn/+t+6+rq8O2336KtrQ3t7e04dOgQTp8+DYlE0mH79vZ2rFixAhqNBiKRCB4eHnB1dcXs2bNRW1sLb29v3Lx5ExcvXkR2djZ8fX2RkJAAHx8foY9ff/0VK1asQEtLC0pLS5Gbm4vz58/jq6++Ql5eHm7evInq6moAjxbAd0QSiQR5eXno3r07Dhw4gOrqarS2tuLNN9/EhAkTsGXLFkilUhw7dgx1dXVwcHBAYGAg5s+fj0mTJgEAfvrpJ/zzn/9EWVkZBg4ciOHDh+PBgwdISEjA6dOn0dTUhP/85z/QaDQWJX6CGhsbERgYiLFjx+LChQsIDAyERCLBw4cPAQAffPABRo4cCUdHx/9qPLRaLfr3749u3bpBJpMhOjoaWVlZKCoqsrkpBQBkMhm+/vprvPLKK1i5ciWio6MhlUoBAGfOnBFkZuHpu+++Q21tLerr62E2mzFnzhzEx8fDaDRi3rx5VnL28PBAS0sLUlJSUFRUhPT0dJsbU5ycnBAXF4fS0lLI5XJhnEwmE44cOYLdu3fj/PnzqK2thUKhQGpqKjp37iyMt4UqKipw+vRpXLx4ETU1NXBwcEBBQQGuX7+OY8eOgSQ2btxoUx6NjY0AgPPnz6Nr165ITU3FiRMnEBoaarXJqGvXrnB3d0djYyP0ej3UajXKyspgMBggl8vx008/oVu3bgCAgoICDB48GO+++y5iYmIQERGBgoICVFZWory8vEM+VCoVpk2bhhs3bqCmpgY3btxAREQE8vPz0dzcjO7du8PBwcHme7i5uSE2NhZdunSBRCJBQEAARo0aBalUiqqqKrS2tqKtrQ379++32QcA+Pn5YcOGDVCr1ejatSt69+6Nn3/+GZ988gni4uKsNlh5e3vDzc0NmzZtwtSpU5Geno7Bgwdj27ZtiIiIQENDAzZu3Ihdu3bBw8MD9vb2+OCDD1BQUICGhgZB9h2RZR7l5ubCbDYL8/P69etobGzE2rVrMXbsWMTGxsJgMFjde+nSJcydOxcfffRRh30rFAr069cPf53kvsMAACAASURBVPzxBx4+fIht27Y9dc5InyqxDqimpgbTpk3Dxx9/jPT0dHzyySc4efIkFAoFsrKy0NbWhqKiIhQWFgoGz83NDWKxGEVFRZBIJOjRowdmzpwp9FlXV4fdu3dDJBJBp9OhqKgIZrMZN27c6JAHkqiursaaNWvQ2toKtVqN8vJyHDt2DAaDAZWVlQgLC4OdnR169OiBIUOGWBnRY8eOYdGiRTh37hzEYjHs7e2h1+uRl5cHsViM8vJyyGQyTJo0CRs3bkRVVVWHfNTV1WHEiBFISEiASCTCBx98gPLychQWFuLll1/G999/j7KyMtjZ2QF4tNvtwYMH0Gg0CAoKAgAUFRVh8+bNOHPmDBobG+Hs7AyVSoV+/frhP//5D5577jmcO3cOer0eJSUlHfKh0+mQl5eH+vp6lJWVoaCgAO3t7WhpacHKlSsxbNgwmxPs+vXr2LdvH+zs7DB9+nQAj3Ynent7Y9WqVSgqKsJ7770n7EQ7fPiwTd0Qi8W4fPkyPv/8c1y9ehW7d+/GjRs30NraivT0dLi7uwMAli5dim3btkEkEqGlpQUVFRW4e/cuvvjiC1RXV8NgMKBXr15PjHdjYyMWLFiAmpoanD59+ondhBaSy+Xw8fHB9u3bYTAYEBMTg19++QXl5eUoKSnB5cuXUVNTg1dffRUVFRUoLi5G3759kZycbNWPRCKBv78/XnjhBXz88ceoqqrC5cuXUVtbi4qKCgQEBMDV1dWmPCwOonv37vj555+Rl5eHbt26QS6XW7XbsmULrly5ApI4f/48XnzxRYwYMQI1NTWYMWOG4MTu3r2LN954A0eOHIFUKhWM9rvvvovm5mYcO3asQz5aWlrg4uICo9GI/Px83L9/H5988gnKysqecOpFRUXQaDQdBh2RkZGYM2cONm3aJOyw7Nu3LzIyMtDe3o5Lly4BAJydnTvkQyQS4fvvv4eDgwNMJhOuXLmCd955B6mpqVYOaenSpcjIyMCIESOQk5ODpqYmBAUF4eLFi9i5cyf8/PxQUFAgBAcqlQrvvvsutm/fjrCwMGRlZWHnzp02d9LKZDJ88sknmDx5MgoKClBeXo729nbMmzcPGo0GCQkJVrslAWDz5s04fPgw7t+/jxs3biAxMdHmOzo5OQn/Ly0thUaj6bAt8P/CAAcFBeHMmTOYMWMGCgsL4e3tjdzcXDQ1NeHmzZuIjY1FSkoKfvzxR+zfvx9yuVxQ1IqKCvj6+iI8PFzor7i4GB988AHee+897N69G+fOnUNoaCjkcrkwYf9MJNHU1ARnZ2eIRCKkpqYKEWhZWRnKysrQ3NyMF198EYMGDYKfn5/V/Xfu3EFxcTHi4uJw4sQJ6HQ6+Pv7w9HREdOmTcOePXvw3Xff4dq1awAgGNA/k0gkQklJCXbt2gW9Xo8PPvgA06ZNw+7du6FQKPDrr7+ipqYG06dPR2lpKX799Vfcvn0b9+/ft+qnoqIC5eXlGDduHLZv345FixZh69atIInKykp8+umnmD59OhQKRYd8NDQ0QCaT4erVq/D19cWRI0fQ3NyM8ePHY9SoUU8oEwDs2LEDR44cQU1NDZRKJUwmE7Zv344xY8YAeBQ19e3bFxUVFVizZg2uX78OnU6HTp062fToTU1NcHFxwaFDhwAAu3btQnFxMVJSUjBy5EiBj8zMTOTm5sLe3h6urq5ob29HaWkpLly4gJ49ewqOAABqa2uRlZWF6dOn4/3338fBgwcREBCA8+fPw97evkM+SkpK8OOPP+L27duoqqqCVqtFe3s7ysvLERsbC4lEgj59+iAvLw8uLi7IycmBi4sLgoODhT4uXryIzz//HBUVFbhz5w7i4uKQmZkJe3t7iMViBAQEYMSIEfjggw865OFxunLlCioqKjBz5kxUVFQIBtVCkZGRWLRokfD/iIgISKVSVFZWIjY2Vvi8vb1d2Hbe3t6Ouro6VFdXo6WlBdevX7f5/NbWVkRERGDnzp0QiURwcXHB2bNn8eGHH6K1tRUeHh5C26+//ho5OTlITk7G888/b2VM3N3dERUVhXfffRcBAQFoaGhAfn4+FAoF5s6di9zcXNjZ2QlHA3Q0LiRRUVGB2tpaXL9+HSkpKVaG8tKlSzh58iSCg4PRu3dvFBQUID8/H01NTSgtLUVlZSV++uknAI90VKFQQKfT4ccff8Tx48fh6+uL9PR0rFq1CjKZrEM+RCIRTpw4AZK4e/cugoODcf36dTz//PPC9n8Lffnll/jtt9/w888/4/fff8eAAQOwdOlS+Pv725S3XC5HWloajh8/jrfffhtz5syx2fZvG+Dq6mphQqxZswbOzs64d+8e2traIJFIUFZWhoSEBFRXV6OoqAgtLS24f/++kBJ4enoiPj5e6K+pqQk3btxAcHAwKioqUFdXh6qqKvzxxx82IQgLH+Xl5XBwcMDVq1cxe/ZszJ07F8CjNKC0tBQBAQHw8fGx6icnJwfnzp1DfX09SktL0dbWhubmZsjlclRWVuL8+fPw8vLCRx99hN27dwOAcI7Bn6m1tRWenp6CsXRwcEBNTQ1KS0uxefNmiMViTJkyBbW1taiurkZdXR0SEhIwYMCAJ/qqra3FtWvXUFxcLET+IpEISqUSDg4O0Gq1ePDgQYd8yOVytLS0CP1UVFRg3Lhx+Mc//tFhJLNjxw58/fXXOHfuHGpqagSHWF5eLhhgC6nVaowfPx7Tpk1DQEAARCKRzciira0NL774Ij777DOQxNWrV9HS0oLXX39dmMhffvklSEKpVKKsrAxisRghISHw8PBAU1MThg0bJkQXBQUFWLNmDfR6PXJzc4X3bGxsxMqVK/Hee+91yEd1dTUKCwsBQHDIaWlp+PXXX5GUlASpVIrCwkJkZGTAYDCgS5cuQkZiodu3b+P48eOQSqW4dOkS0tPTcerUKUHOSqUSlZWVuH//vk2ICgDGjh2LyMhIXLx4EWazGbNmzcKECROsnEf//v1t3v84tbS0oLq6GkFBQTAYDLh8+TLq6urQ0tKCe/fuPfXePn364J///CeMRiOSk5Oxbds2lJWVoby83Cpdv3DhAvbv3w+RSISgoCArA3z//n388ssvwpkhdXV1uH//PlQqFYYNG4a5c+eira0NtbW1HfJQV1cHAKisrMSZM2cwdOhQq6Di6tWr+Oqrr5Cbm4ukpCT89NNPOHLkCJycnAQHGhYWhvz8fFRUVMDLywv5+fno3r07du7cCeCRzlhk8Wf4wEIuLi7Yvn07UlNTsX//frz++uuYP39+h229vLwQEhKCLl26AHjkLHv37m1TzjU1NViwYAFKSkowatQoXLlyRXCaHdHfwoD1ej1u3ryJ9vZ2fPPNN+jfvz/Ky8vR1tYGAOjcuTMKCwtx+PBhODk5wcXFBfX19SgqKsK1a9egVqvRt29foT9L+j106FDI5XLk5uZi4cKFaGhoQHl5OUpLSzvkw3LQxtGjR6FUKpGRkQGtVgt3d3eIxWL07dsXL730EqKjo61S77KyMhw8eBBZWVmws7NDQUEBunfvjvr6ety+fRutra04efKkgMVZIIHHsSkr4YnFEIvFGD16NAwGA9avX4+uXbtiwoQJqK+vR2VlJSorK7Fr1y6cPXsWTU1NiI+PF9LrCxcu4Ntvv4VSqURhYSH27dsHd3d3nDhxAi0tLXBwcEBtbS2WLVsGX19fYfL/mdRqtZAttLe3Q6lUor6+Hnv27LHCNb/55hssWrQIH374Ic6dOwcvLy906dIF5eXluH79eoeGRCKRwM/PT8D7bt26ZRN7tYypu7s7Jk+ejPr6evj7+yM8PFyYaDt37kSnTp3w3HPPQavV4v79+6iqqoJYLIZWq7WKLMrKyrB37144OzsjIyMDwKPsx4Kv2cLEAeD5559HdHQ0RCIRNBqNgAseOXIEKpUKBw4cgEQiQWFhIfLy8p6QbWtrK1paWoSxl8vlSExMhFQqhUajQX5+Pj777DO0t7dbGanHSSQSQSqV4vXXX0dkZCT279+PX375Bfv377dppGxRQUEB1q1bhytXrkAikSAyMhI1NTUoLCzEpUuXbDpnC+3duxfPPfccGhoaEBQUBKPRiJqaGuzduxcVFRVCu1GjRmHhwoVISkp64r0kEgkUCgXa29vR2NiIuro6BAcHIzo6GuvXr8fx48dx+/ZtmwGLSqUS5rRMJkNdXZ3VGMrlcoSHh8PHxwcZGRn4+eefAQBSqVQ4tKuyshJSqRQ+Pj7o2rUrfHx8UFhYKOiNBYseNGiQYJf+TBaH4+DgAFdXV9y7dw92dnZPHAwGAImJiXjjjTeQnp6O9PT0pxrfqqoqLFq0CP/+97+RlZWFO3fu4JNPPnkqHv23ImBL5NOvXz/s27cPcXFx0Ov1CAwMhIuLC27evIn8/Hzs378fgYGBaGhoEO4ViUSIjIzEkCFDhM+Ki4uxadMmfPLJJ7h48SJEIhHs7OxgMpmgUCiQn5/fIR8ikQgzZ87EqlWrhFPX/v3vfwue+eHDh5g4ceIT0MPp06exf/9+VFZWwmAwQKfTQSKRoKmpCcXFxejVqxdEIhEqKyuxbNkyPHz4EGKxuMOBsdAvv/yCGTNm4MSJE48EKpVCqVRCJBLh+eefx2+//YaKigpIJBIkJCQgIiJCuPfBgwcoLCyEnZ0dVCoV4uLiEBISgrNnzyI3Nxfh4eFCv4GBgTYjraqqKtjZ2UGr1UIsFmPcuHH49ttv0djYiDfeeEPA5LZt24acnBy0tLQgKCgIzs7OKCgoQOfOnREZGQk3N7cn+m5oaBBSytraWqs0+M+k1+tx4MABvPrqqwLuNXz4cCsMLCUlBTdv3hRwwNu3byMnJwcikQj9+vWD0WgU2iqVSri7u+PWrVvCaVhHjx4VnO/TFDsqKgpHjx5F3759cenSJezcuRNhYWHYvHkzVq9ejU6dOqG8vBydOnWC0Wi0gmmys7Nx7tw5BAYG4s6dO1AoFLh69So0Gg1kMpmgn1VVVVCr1aipqemQBzs7O+zfvx9+fn5QqVQ4dOgQXFxcsG7dOhQWFj4BoahUKgQEBCAsLOyJ0wSrqqoEqOTu3bvIz8+Hj48PXFxckJ+fj7CwsKdGwUuWLEH//v1RUlKCK1euwM3NDW1tbQgNDbUan+HDh9vsw2AwIDIyEjKZTKi1SKVSdOvWDdevX0d0dDRKS0ufgFgsJBKJBMdjZ2eH7OxsK2ceGBgo2JKzZ88iJCQELS0tQnDUs2dPNDY2oqioCL/++ivu3buHpqYmoRBub2+PS5cuQS6XQy6X2wzgamtrkZqaiqNHj0Imk2Hjxo0QiUTYtWsXRCIRXnrppSfk/1fU0NCAo0ePYs2aNejcuTMaGxuxb98+m87IQn/LAFtSCBcXF3h5eWHv3r1QKpWYM2eOkG7eu3cP1dXVyM3NRWVlpWC8QkND0adPH6Gvqqoq/P7773BwcMCePXvwzTffQC6X41//+hdGjx6N6OhoLFu2rEM+7O3t0aVLF4hEIrS1taFnz564desW6uvr0dbWBrVaDaPRCLVaLdxz9uxZ7Nq1C/fu3UNDQwOqq6vh7OyMa9euQSwWw87ODlVVVZDJZLh9+zZqamoQHR2NgoICAQv+M7W1taG8vByXLl1CWVkZFAoF3NzcsGXLFmi1WpjNZtTU1KC8vBxeXl4YN24czGYzgEeT/O7duzCbzSgsLERrayv8/PwEo6xWqyGRSODi4oLS0lLcvHnTZsGnpqYG/v7+aG1tRVFREcaPH49Tp07Bz8/PSgYWHL6yshLOzs5obm7GjRs30L17d7z66qv4448/nui7qakJhw4dAklMnjwZy5cvtwkNyeVy2Nvbo729HV988QVGjx4NmUyGrVu3YvTo0dBqtZg2bRrWrFkjVNaLioqE6K13795WBlgkEqGurg579uyBSCSCv78/jh07hsjISCFD6YjEYjF0Oh3EYjHc3NzQo0cPnDx5Eg8fPkR9fT1+//13qNVqFBcXQ6PR4OWXX7bCWn/77Tfs3r0b3t7eqKurg729vRAlu7q6QiqVIiAgAG1tbcjOzhaKT38miUSCyspKLFq0CHPmzMHNmzfh4uKCq1evorm5GWKxGHfu3EF9fT06d+6MtrY29OjRA9HR0ejRo4eQKT18+BDXr1+HRqOBSqVCcXExfvnlF3h6esLJyQnV1dVWOO6fSSQSYfz48di9ezeUSiWkUikePHiAX375BWFhYbh48SJcXV07dPDZ2dmorKwUIj93d3ekpKTgp59+QktLC27duoWXX34ZlZWVqKqqws8//2wzM7E40czMTPz222+IjIzs0Fi/8MILeOGFF2y+T3Z2Nv71r3/h0KFDUCgU8Pf3R15engAjiUQi7Nmzx+b9MpkMI0eOhJOTE3bs2IGoqChcuXIFGzZsgK+vL8xmM06dOoWgoCCUlpYiPz8fLS0tiIyMtAqgLPTgwQNs3boVmZmZEIvF0Ov16NatG7744gtERkba1A/gbxpgS/Fl+/btCAkJQW5uLgwGA/R6PdauXYsBAwbAyckJLS0tiIiIQHl5Oa5cuYKamhokJCTgpZdeAvCoMrhv3z5s27YNJLF9+3ZERkYiNzcXAQEBqKqqwuDBg20WnXQ6Hf75z3/C29sbJSUliImJgZ2dnVAZf+utt2AymazuOXz4sLAUqra2Fk1NTSgrKwPwaNnS888/j++++w6enp64d++eIGyZTIbs7OyOhSeVoqmpCefOnUN7ezt69+6N5ORkODo6Ijs7G1lZWaipqUF9fT0SEhIQEhIieETLcqju3bujubkZDx8+FCCDhw8fwt7eHnfu3IHJZBI8eWVlpc2xGT58OBYuXAitVotDhw7hueeeg0wmszKWdnZ2yMnJgV6vx927d+Hs7IwePXpAp9PBy8sLXl5eVn3W19cjIyMDr7zyCnbu3Inw8HC4uLjYTO2Ki4vh7e2NP/74AwqFAh999BGWLl2Ko0eP4oUXXhAm+ONFtqdRc3MzCgoKIBaLBShJrVbD19cXhw8ftrmsTqPR4MqVK/D19cXmzZvxyiuvQCqV4vjx45DJZNi7dy/a29sRHByMS5cuCcXjx+nhw4dQKBQwGo2ora1FTk4OhgwZgtjYWLz99tsoKCiAp6enzSwNeJTqWhxodnY2amtrQRJtbW3w8vKCk5MTunXrhpMnT8LDwwPFxcXIyclBe3s7dDqdYIArKipw7NgxHDhwQJB9aWkpWltbIRKJUF5ejk8//dQmH1KpFL179xZWBnh4eCA7OxsqlQqnTp0S+hs+fLhV4fvWrVvYuHEjZDKZYIADAwPx5ptvYu/evdBqtWhra4O9vT1Onz4NOzs79OzZU4AO/kw6nQ5BQUEoKSlBW1sb8vLybOrS06hbt2746KOP0NLSgtOnT8Pb2xsymQwnT56EXq+3uRzPQi0tLTh8+DCGDBmCr7/+GuHh4UJg1qNHD5w5cwb//ve/0adPH2E5a9euXbFgwQIrA1xWVobz58+jpKQEu3fvxr1790ASWVlZeOmll9Da2oqAgICnnhn9tzBgS+ppNBpx584ddO3aFYmJiTh48CCcnJywadMmXLlyBUFBQQgLC0NcXBwiIyMRGRkpgNjAIw+2atUqKJVKeHp6Ijw8HOPGjUNVVRW6deuGP/74A/v27bNZ7Pnjjz8QERGBTp06wc3NTcAQe/XqZRXxWejKlSsoLi5GQ0MDSkpK0NraarUUSKfTIS4uTjg83NHREZ6enqiqqsKePXtsYsANDQ2IiIgQjGp0dDQ2btyIcePGQSqV4syZMwgNDYWHhwfu3r37hGKQRH19PR4+fIimpibU1tYiODgYMTExcHNzg0KhwP379yGTyeDn5/fUouTZs2dRVFSExsZGLFu2DM899xyys7OtYCBLVhIdHY1p06Zh0qRJGDt2rFVm8vi7ZWRk4K233kJSUhJiY2Mxb948ODs7P1XBq6urkZKSIqyhdXV1Re/evXHu3DkcOnTICm+00M2bN3Ho0CErR1dRUYFff/0VWq0WDQ0NqK2txfPPPw+ZTIY9e/ZApVLZXN6jVquxdu1aYQnTwYMHER0dDaVSicTERKSnp0OlUgnrwx+vfOfn5+P69etwcHBA79698fLLL8PZ2Rk5OTnYtGkTXF1d4eDggMLCQpw9e/apOLRKpRJgnczMTLS2tmLEiBHQaDQ4cuQI7t69i3/84x/o1q0bTp8+jbt37+Lu3bvo2bMnxo0bB+BRsPDgwQPodDpotVqEhIQAeBTl2dvbIzExEZ06dXqiev84SSQSLFy4UIjIDx8+jObmZjx48ABtbW0oKCjA/PnzceLECZw9exbff/89zp07h6+//hoZGRmorKzExYsXhf50Oh0SEhLg5OSE2tpafPzxx4iKisKwYcM61CULFRYWYsOGDfD390dycjKys7Nx+PBhHD9+HIcPH0Z2dvZTYaXHqUuXLnj//fdRV1eHc+fOYfz48dDr9TCZTIITsQXZOTg4YMuWLZg2bRqcnJzw66+/Ii8vD3l5efjiiy/w9ddfo76+Hlu3bkVhYSGioqKwaNEivPjii1b9ZGdnY8qUKfjkk08QGxuLhoYG9O7dG507d8a//vUvYX/A/9V1wADQo0cP3Lp1C8XFxSguLsb169cxePBg1NTU4N69e9BoNHByckJzczPi4+PRr18/REdHA3hU3KioqIBSqcTChQvR1NSErKws/P7772hpacGOHTswb948ZGZmPrVQERgYiB9++AGhoaFYtGgR/vGPfyAtLQ23b99+ArPNyMhATk4OwsPDIRaL0d7ejocPH+LmzZvCAvcdO3YgMTERgwcPxu+//w4vLy+cPn1aWCxui7p164bAwEAcOHAA27dvR2lpKb755hvExcXhyy+/FNLenTt3on///ujevTsACJDH/fv30b9/fxQXF0MsFqO1tRUvvfSSsKYYeKQwZrP5qXxYUrmHDx/Cx8cHU6ZMwfjx460cUrdu3VBfXw93d3f07NlTWAL2Z+y3trYWJ06cwIEDB6BUKjF16lS8/PLLQiTWkZMDIPyqxJ49e9DW1oZ33nkHVVVVKCoqAvAIslmzZg1CQ0Ot7vvyyy+xefNmvPrqq9iwYQOAR0WnzZs3C0Wu6upqAcYAIEApHZFlA8rMmTNx9+5ddOnSBcuXL0d1dTWef/55dOrUSSjgjBw50iryP3r0KLZt24b79+/j+vXr6NGjB6qrqxEcHAwnJyesWrUKnp6eIImqqir4+/vbzJAePnyIlpYWGAwGiMVi1NfXIyUlBQcOHEB9fT20Wi2OHz+OlpYW6HQ6KJVKKBQKq+Dg2rVrWLVqFQoKChATEwM/Pz/cunULvXr1QlhYGDIyMnD58mX84x//wJdfftkhH42NjYiMjBSw36KiIrS3t8PT0xNKpRJ1dXVobm5Gfn4+tmzZgtOnTyMiIgJ1dXWor69HZmYm6uvrsW3bNgCAv78/Vq5ciZCQEBgMBty4cQNisRglJSUoKSmxiQFbDNHVq1eFiHrq1KlQq9Vobm7G8OHDMXDgQOh0OgQHB9tcT/w4qdVqmM1mFBQUwMPDA21tbZg0aRLee+89m8tYy8vL4eHhgaKiIkRFRWHo0KG4fPkyfHx8IJVKkZOTAycnJ8jlcowePRrTpk2z2sQFPMoYzp49i7y8PBQUFKCwsBApKSmIjIyEWq3GsmXLMGnSJKxYsQK3b9+2/QK2tsh1dD2+tz05OZlSqVQ4LMZkMjExMZFRUVGMiopieno6R4wYwW3btllt1bt79y5nzZpFg8HAmJgYGo1GLlq0SOjXctBJZWWlzS3AlpOdJkyYQLPZzNGjR9PDw0M4nCcrK+svtwxmZGQwPT39L9u1t7c/dWulZW97VlaW8MsE69ev5+rVq+nr60u1Wk2JRCIcEG+hgwcP8v3332dmZib379/PjRs38pVXXrHa6m05f8HDw4NOTk50cnLqkA+ZTMb4+HjhBDrLqVQhISFPbL/+b+jSpUuMjIzkyZMnBT4sZyH4+/vbPBbT2dmZ4eHhPHXqFDt37kyRSMRevXpZ/Zac5SQpe3t7enl50cPDQ/g1k9dee03goaamhkePHuWUKVOEk/KCgoKsDr1/2rgAYHR0NOfOnSscHGM5Z2LgwIEcNWoUg4KCeODAAat337FjB4OCgoQt7mKxmCaTiWvXruXBgweFLdDx8fHC4ehPOxi+f//+jI+PZ3h4OHU6Hd3c3CiRSITzVMLDw9mrVy+GhoYyIiKCZrOZn3/+OUmyrKyMS5YsoUKhYP/+/bljxw46OzvT29ubhYWFXLt2LUNDQ2k2mzlo0CCbZ6eoVCqGhoYyKCiIQ4YMoUqlokQi4ZgxYzhz5kympqbS2dmZS5YsYWxsLA0GA7VarfALDwsWLOCAAQOszq+4e/cue/XqxaSkJIpEIuGHEbp27WrzEBzLMaUvv/wyFyxYQHd3d0E/LD9I4O/vT7lczt27d/+lnp4/f55+fn786aefqNFoKBKJaDKZmJ2dLZxT3hEfPj4+HDNmjHDWb3JyMuVyOTds2MBNmzbR29ubJpOJOp2uQz6qq6u5evVqK50OCQnhyZMnqVKp+Prrr9PFxYWJiYlP1VPyb54FYTmw2XIMpOUKDw+nUqnkm2++yVmzZvGtt97iqFGjhGMHLfTw4UOuXr2aDg4O9PT0ZFRUFA0Gg7Af3PIjgBKJhP+HvTePqrra//+f7zPDOYfhwAEOs0hAggcEEq7iQIpKTpCgUZpyc6gckDKLa9ng0iYrh1uZw/WaWTlfsywtzfKTpmZpZjnlgBMCyiTz8Pz+4T17eZLjvX1+n7VYv7X2Y62zlAPn/X699/Dcr/3a++zXxo0bXRoeFBRERVEYFBTEw4cPc8GCBZw8eTKjo6NpNpt5+PDh/1h5/y3/SYAdr+DgYD766KPiVKmYmBharVZGRkbSaDRy/fr1TtedP3++qDzHwTCBgYEiZYtarWZkxvvwMAAAIABJREFUZCS1Wi1fe+01Dh482OVZEH5+fjSbzeLcAz8/PxoMBn755ZftHj35nzh9+jSzsrLEAeGO79hbrVauXr2aAQEB7drh7e3NnJwczpgxg2+88QY9PT355JNPijM1HNdwnBA3c+ZMzpw5k8nJyXR3d+eUKVOEDd988w19fHyEHZMnT+Z3331Hf39/hoaGigNxXNWLY5CePXs233zzTZGJZcqUKeKMVovFwq1bt95WLxaLRaSw8vDwYGpqKhMSEsRhQY4zKxzZLFzVi06n46uvvkqVSsXjx49z4MCBnDBhgjhb1tPTkzabjb6+vuLA/kWLFglbli5dSg8PD8bExIhD7B977DEmJCRwz5499Pf351tvvcWjR4/e8XCiqKgorl+/XgyEPj4+Tgk+vb29GRQUREdqK8eZ3/379+fgwYNpNBoZEBDAhx9+2OkwpUuXLjEqKooeHh4MCwtjWlqaGKTas8Nx1q+jPTgSGTiS1wYHBzM1NZVBQUHcvHnzf2ynR44cYVRUFPPz89m3b19x9Gtqaqo4za89O0wmEy0WCzt37kw/Pz+RlHPnzp1cvHgxY2JiuGzZMubn53PXrl1O96yuruYnn3wiUhJptVoGBgYyMzOTGzZsuO2sHEf/ac8O8k+eBeHp6QlFUTB48GARj9RoNOjUqROMRiP+/ve/49SpU1i5ciV27tx52/acTz75BEuXLkViYiKqqqpw48YNqFQqfPbZZxgzZgymTZsGk8kELy8vbN++3eX2L5Ji72xCQgKmT5+O0tJSPPvssxg8eDCam5td7pl1RVtbm1gkKSsrQ3FxMSorK3Hx4kWX36gBIBaHHKEHAHjggQcQHByMpqYmnD9/Hmazud1pu0qlgre3N6KjoxEZGYno6Gg88cQTAG6uGOfm5sLDwwN/+ctf8Nlnn7l8ptLSUqSlpWH16tXQ6/VQq9UIDg7G0KFDsXfv3v+qLOrq6lBSUoKqqioUFxdjz5494rnb2tpw4cIF5OTkoLS01OVXopuamrBhwwbMnz8fL7/8MhISEuDj4yP2w77zzju4++67cePGDRw4cACvvfYadu/eDavVir/85S8iHOQom+bmZhw6dAhjx47F2rVrERgYiEmTJiEhIQH5+fkut/io1WrMnj0bKpUKL730Ep544gkYjUaMGzcOH3/8MbRaLdzc3MS1HFRVVaG1tRUhISFiCurn5weTySSSuQYEBGDnzp147LHHxJ5gV1uWrFYrTp48CZJYuHAhdu/ejWXLlsFqtSImJgb3338/CgoKMGrUKISGhsJqtTptTTOZTBg7dix+++03fPDBB1Cr1VCr1UhOTkbv3r2hKAoaGxsRHx+PvLw8l/3l5MmTePDBB0X4pqamxjFww8fHByNHjoTNZsOIESMwcuRIBAYGwsvLC6dOncK5c+eQnp6Op59+GpmZmU5ndAQGBuLgwYO4//77MXHiRGzcuBFDhw51WS91dXVwc3PDww8/DJPJhO3bt6O5uRnu7u4YNmwYNBoNfv31V1y9ehUlJSUuQ5ANDQ24cOECSOKZZ57B9OnTER0djXvuuQfAzS/R3H333U7rH7diNBpx/fp1/P777ygtLQVJdO3aFYsXL8bf/vY33HXXXVixYgX++te/On1pDAC2bt2KzZs3IyYmRux6GjBggFhk9fDwgKIoCA4ORkxMDHQ6ncuQHYD//WlojtOhYmNjOWnSJPbs2ZO5ubm02Ww0GAzctGmT08hRW1vLPXv2sKCggNHR0bTZbBw6dCi1Wi0TExO5cuVK+vn5cdy4cXzppZfECNKeHQCYlpYmpty3njf71FNPMTs7m3v37mVpaSkrKirY0NAg0pVUV1eztraWJSUlLCkp4dWrV1lSUsLjx49zwYIFLC4u5pAhQ2gwGDh16lSRY6o9OxRFEWlHevbsyXHjxjEtLY0BAQHi8O6YmBh+9913TmXR0NDAOXPm0MfHh5GRkQwLC+PQoUNFQkPl32ncvb29WVBQwDVr1ojEja7scKRTcqRasVgs7N69u/CES0pK7ugNb968mT179uTs2bP5wAMPsH///szLy6O3t7eYLoeHh9/xlClHGikAXLFiBT/77DNOmjSJ7u7uDA8Pp9Fo5IgRI9i3b1/abDY6Dsb28/Pj6NGjuW/fPmHPN998I7Kc4N8zAocn+Morr3D06NHUaDTt2nFrssWgoCCRBcLPz4/u7u4MDg6mu7s7P//8c6cymD17Nh999FEuXryYJpOJCQkJnD17NmNiYrhq1Sr6+PgwPT2db775pvCccnJyXE799Xo9Z86cyYCAAMbFxYk0UzqdjtnZ2czKyhLnLHfu3Jk+Pj63JQ1w8OOPPzItLU2kBzKZTJwzZw6HDRsmPHNX52drNBrGx8c7zVgdYZzCwkKmpaXR3d2da9euZXZ2tjjfd8mSJfznP//J/Px8jhgxggDYp08fJ7vq6ur4xhtv0GAwiFRRrupFURT26dOHly5d4tChQ9m9e3dxBvKPP/7IN998k1OnThXnNS9ZsoQlJSVOZ1g3NDRw3bp14nxmT09PGo1G9unTh2PGjGFgYCDfeustkd/OVXk4En4CEPnuHHoCoN2sMTU1NSJcGhgYyO7du9Pf35/p6en09vZmTEwMn376aZrNZh44cIBWq5UpKSkuz2km/2QIwpHvXq1Wc9OmTXzyySc5d+5cDhgwQCTIc+RKuvX4PPLm1C4+Pp7Dhw8X6U0cjcDT05Nz5szh2rVrhQDhDq47AO7fv58mk4kGg4E9e/akVqt1SuWTkJDAkJAQZmVlce3ateJA7EceeUQcGq7X60VKIcc0aOzYsSI7rONarirSkeXAarVy/vz5NBqN1Gg0XLp0qchtZzabuXHjRlZXV4s8bBs2bGCfPn2YnJzMwsJC5ubminxx+PcUcdmyZXz22We5Zs0aWiwWnj17lhEREe3aYbFY6OXlJdKo6PV6fvXVVzxz5oyIxQI3D+F2ZBioqalhfX096+rqWFVVxYMHD3Lu3LmcNGkSbTYbQ0JC6O/vz61bt3LQoEGiHBxZatuzw8fHR4SnIiMjeWvuLUVRmJSUxK1bt3LJkiXs0qUL7XY7w8PDGRsby5SUFBYUFJC8eSbxnj17OGbMGH7yySeizdntdk6YMIEWi0UcG9ieHY4yAG5mux0/fjz1ej3/8Y9/MDAwUIQxtm7d6pQ95aWXXnKKLzuSyPr7+zM4OJhPPfUUhw4dyuzsbJGeKTw83GV5OGwsLy9nQUEB9Xo9J06cyD59+vDhhx8WmZ/j4+NptVppNpv53nvv3dbxm5ubuW3bNqdsv46UQiqViiaTSbTX9uyw2+0ihg/czMbtSDrbr18/EZKIjIxkSkqKSIobFhYm3hs1ahTVajX79OnD6upqVldXkyRLS0vpiP1PmzaNDmegPTtuFbiwsDBxDrPZbOaoUaPo4+NDm81Gk8nEiRMnsm/fvtTpdJw1a5Zos+vWrWNwcDC1Wi19fHzo6+tLu90uQkOFhYVUFIUjR450KXwqlYozZszgyJEjRcLcdevW0dfXV/TBPwpwfX09Z86cKc5FdvT7KVOm8KeffuLrr79Os9nMESNGUFEUGo1GxsXFcfjw4UxMTPy/EWBHR0pJSRGNYdKkSTSZTExLS+MLL7zAuLi4dg/+nj9/vkhfEhcXx4SEBKfUznl5eU6LTo7ru7Lj1KlTfOaZZ/jcc8+Jzta3b1+RAFGlUnHLli387LPPOGHCBIaFhYm4cXh4OLt06cKRI0fSZrPx4Ycfpt1up7e3N+fNm8c9e/Y4jYauknI6OqmjQcXHxzMzM1PkGQNuLlaGhYUJ8Rg2bBiff/55nj9/nt9//z1zc3OFZ+WIMxqNRo4aNYparZYGg4G//fbbHWN8jgUItVrN3NxcTps2jUajkU8//TSnTJlCX19fAmBSUhKTkpKYnJzMfv36saCggOPHjxcLmF26dOGUKVOcBkCdTsfQ0FBqNBqRrvtOKYlqa2uFZ1FVVSXurdfrOWLECAYEBDjlGtPr9XzxxRd55MgR7tmzhw0NDdyyZQuBm7nfunTpIuzx9vZmdXU1PTw8WFNTc8f28dJLL1GlUolU7rfmJ+zbt6+IIa9Zs4YtLS1samril19+ySFDhtDNzU3kPps5cybDw8O5ceNG1tTU8Ndff2VWVpYQwtjYWIaGhrq041anwJGc9tYEnUOHDhVtc/HixSIjRVtbGxsbG9nY2CjSyFssFrq7uwsxdSz+bN68mWq12qXgJCYmigELAH/99Vf6+/szPDycWq2WgwYNEmdHO/KeOfpQREQEFy1axB07djAtLY3du3dnTEwM7Xa76NuVlZVUq9UcMmTIHRedHDYoiiJ0Iz8/n0ePHhWZZYCbCUFTU1M5bdo0sbaTkJDAvn37MjExkePHj2dqaiqLi4t59OhRkTwzKiqKX331FfV6vUj22p4dISEhTllrbp1BO7Jl/1GAp0+fLmZhSUlJVKvVzMzM5I4dOxgbG0uz2Uy73S4SrTrqx2AwMCws7P9GgB1T5pSUFKd8So7G41iR/COtra189dVXxSg5btw45ufn02q10mq18sknn+SgQYM4evRoERB3rES7atgHDhy4zVtJT0+np6cnDxw4wLS0NMbFxbFfv37ctm0bp0+fLoRfURQmJiZy3LhxYkRzpGKfNGmSmEY5ru1qSgWAGRkZopNpNBomJSVxx44dIn12Xl4eZ8yYwTlz5jA1NZVDhgwRi1sOu0NCQkSyR0d5+vj48J133nEaCFw1KJvNxvDwcI4aNYrTpk0Tiy23poCy2Wx84IEHGBUVxYiICJEiydfXl/369WO/fv0YERHBvLw8xsbG0mQyccCAAczJyaGbmxuzs7OdbHFVHmazmTNnzqROp2NFRYUQ4Pz8fAYEBNBgMLBPnz585plnRH62l19+WXiAPXr04FNPPUUADAgIYEFBgchMnJCQwHfeeYcVFRV3TH8OgI2NjSLr7iuvvMK4uDhqtVoeOnSImZmZYpAIDw/n2rVrOXnyZHp5eXHOnDlsbm7munXrGBoaypiYGCqKQo1GIzI339reNmzY4LKd/rHuHIPrHzv8sGHD6BDJN954g83NzTx69Cjz8/OZl5fHo0ePihT1MTExPHv2rPBgHTbdKTedt7e3SEXksH3hwoX09vamSqViUFAQX3zxRc6cOZPLly9ncXExt23bxr59+4rn/Oqrr7hmzRpRZps3b2ZzczObm5tZXl4uFtbu1D5u7VMAuHLlSqdceY5Q12OPPcbCwkJGRkZy+vTpfPHFF0XiXMcg5u3tTavVyry8PAYGBvL48eM8duwYZ82aJQ7id7WLys3NjUajkY888gjVarUIORmNRr755psMDw93EuCWlhbOnj2bI0aM4BtvvMG8vDyOGDGCH330ERMSErh7924uX76c3t7enDNnDsvKyqhSqYRD5ao8/rQAAzfjshqNhjExMaJz5+Xl0dPTk6mpqZwwYQKPHTvGtrY28ZozZ46TKDiMCgoK4vTp0+nl5cWioiL27t2b3bt3/686usMTd8QdFy9eTIvFwoMHD9Jut/O5554Tca+8vDy+8MILwnOYMGECP/zwQ4aFhfH5558XWQ0cLw8PD5EVWaVSudxWExcXxy1btvCFF16gRqNhRUWFiBk//fTTXLBggcjyqtfrOWXKFGZmZtJsNtPDw4MRERGcOnUqlyxZwsLCQjoELD8/n+Xl5UxLSxNZahVFYXR0dLt2OLJIZ2Zmsm/fvqJeHHHVLl26MDMzk9988w1XrFhBu93u9Lz5+flsbGzkJ598wtTUVB48eJCZmZn08vLi4sWLef36dapUKpHx9k4D44ABA8SAPH36dOENjBgxgi+++CIzMjI4evRoPvXUUxw0aJBIXlpYWMj33nvPyS6VSkWbzcYhQ4ZQq9UyOTlZvD9v3rw7tg/gZr6vjIwM5uXl8YEHHmD37t3FDAWAiAmHh4czKChIDMS3ioJjF0RycrK4jqIowru+UztNSkpyykbsqJfz588zKiqKUVFRXLp0KZOTk7l8+XKuX7+ePXr0YHh4OPPy8oQY7t27l/n5+Xz33XcZGBjI5cuX89FHHxXZR9RqtUio2p4dVqvVyWGaNWsWN23aRF9fX86YMYOTJ08WmZId/aV///7MyMjgvHnzmJOTw9jYWI4aNYoA6O/vLxJPBgYG0mg08vPPP/+PguMY9Nt7tTdYOcIiK1eu5IoVK/jHzzvCEMDNNY89e/aI93v06OEyNKTVajl69GixjvDHTD/x8fH86aefhH45ksf26NGDffr0IQB26dJF5CUEbs6A165dK57DoUv+/v4uU0X9aQHWarVsa2vj7t27qdPp+MorrzAzM5N6vZ6DBg2il5cX58+fz5kzZzI0NJTAzfCCh4eHSDvu2FeZlZVFlUrF0NBQETsymUyMi4vj/PnzGR0d7bIi/73Y4OSltleB3t7eTE9P5/vvv8/Ro0eL98eOHcuvvvqKffv2ZXx8PD/77DOR/dghlmq1mo899hh/++03l6leHIX8/PPPizhuXl4ejUYjn332WQYHB/P5558XA0FUVBQdW/kAMCUlRYzg/fr1Y3BwMGfMmOGykf4nwXGIjslkotlspqIo3LdvH3fs2MGYmBjm5eVxzZo1HDhwINPS0vjMM89w4MCBBG7GsxyJSB0vs9nMCRMmOHWGMWPG0Nvb+452WCwWzp07lyaTiYmJidTpdFy4cKHTNNMhSr179+awYcMYFhYmQjGOGYKbmxu7d+/O3NxcPvLII7d93lW93Cp8X3/9NW9dHMzOzhbhh2XLlon4tuP16KOP8vXXXxc/d+nShdu3b+fPP//Mvn37OrWRW9PhtGeHxWJxqu8/tlGLxcJly5Y5xWenTJnCsWPH0tfXl0FBQTSZTGJ7ppeXF/v16ydEJjU1lVqtlkuXLhVboNqzIyoqiqdOnRJCc+HCBV67do2BgYEcMmSI03Y6lUrF4cOH86233uJrr70mMlU7HImYmBjOmTOHMTExPHjwIEtLSxkaGsoXXniBWVlZVBTFEZ64zQ7HwDdv3jxqNBox47JYLDx58iQTEhI4d+5cJiYm8tlnn+XHH3/Mhx9+2ClnoOPl5+fHzMxM8Uzu7u589NFHCdycidbX17ssj06dOolZyNy5c4WTMG7cOEZERFCtVtNqtVKv1zMmJkYIv6NfOQaprVu30tPTkw888AB37drFLVu28B//+IdI4qlWq5mdne3SDpJQ/t1x/isURSkDcP6//sD/d8JI3vZ1GGmHtEPaIe34/7sdAP6cAEskEonk/44/9UUMiUQikfzfIQVYIpFIOggpwBKJRNJBSAGWSCSSDkIKsEQikXQQUoAlEomkg5ACLJFIJB2EFGCJRCLpIKQASyQSSQchBVgikUg6CCnAEolE0kFIAZZIJJIOQgqwRCKRdBBSgCUSiaSDkAIskUgkHYQUYIlEIukgpABLJBJJByEFWCKRSDoIKcASiUTSQUgBlkgkkg5CCrBEIpF0EFKAJRKJpIOQAiyRSCQdhBRgiUQi6SCkAEskEkkHIQVYIpFIOggpwBKJRNJBSAGWSCSSDkIKsEQikXQQUoAlEomkg5ACLJFIJB2EFGCJRCLpIKQASyQSSQchBVgikUg6CCnAEolE0kFIAZZIJJIOQgqwRCKRdBBSgCUSiaSDkAIskUgkHYQUYIlEIukgpABLJBJJByEFWCKRSDoIKcASiUTSQUgBlkgkkg5CCrBEIpF0EFKAJRKJpIOQAiyRSCQdhBRgiUQi6SCkAEskEkkHIQVYIpFIOggpwBKJRNJBSAGWSCSSDkIKsEQikXQQUoAlEomkg5ACLJFIJB2EFGCJRCLpIKQASyQSSQchBVgikUg6CCnAEolE0kFIAZZIJJIOQgqwRCKRdBBSgCUSiaSDkAIskUgkHYQUYIlEIukgpABLJBJJByEFWCKRSDoIKcASiUTSQUgBlkgkkg5CCrBEIpF0EFKAJRKJpIOQAiyRSCQdhBRgiUQi6SCkAEskEkkHIQVYIpFIOggpwBKJRNJBSAGWSCSSDkIKsEQikXQQUoAlEomkg5ACLJFIJB2EFGCJRCLpIKQASyQSSQchBVgikUg6CCnAEolE0kFIAZZIJJIOQgqwRCKRdBBSgCUSiaSDkAIskUgkHYQUYIlEIukoSP7XLwAEwJiYGCqKQsfPOp2OnTp1opubG4ODg+nn50etVktvb28GBwfTbDZTrVbT3d2dPj4+9PX1pUqlIgCqVCoajUZqNBp26dKFHh4e4roqlYqu7AgPD6darabBYKCvry+7detGrVZLjUYjPg+A3t7eNBqNDAwMZFxcHAMCAqhSqajRaKjX6xkaGsqIiAgqikK9Xk8ANBgMBEAPDw/x//bs0Ol01Gq1BMBu3bqJZzIYDE7lo9Vqxe+MRiPd3d2dbARARVGoVqupKAqNRiODg4PF+3fffTcDAwNd2uGwwWazEQDd3Nzo4eHhVBb+/v40GAz08vJi586dabFYnOovJCSEdrtd3Mdxb8fzK4pCPz8/8bv27PDw8KBWq2V0dDQVRWFoaKh47j+Whclkol6vp1qtpkajobu7O319fenl5UW1Wi3uqdPpxP+9vb3FdYKCglza4fgbu90u6iE+Pp5arZY+Pj6iXLy8vKjX6+nn50c/Pz96enrS39+fnp6eVBSFKpWKKpWKBoOBGo1GlIVKpWJMTIzoC3dqpzExMcJuR1n4+vrSZDLRy8uLJpOJZrOZGo2Gnp6eNBqN1Ol04rkBiPLRarX09/d3aueOsnK0O1d2GAwG0bfMZjN9fHwYEREh2u+tbeHWa7q5udHLy8vJHp1OJ9qn49p+fn6MiIigm5sbNRpNu3bcWn+39hcAvPvuu+nm5kaz2Sz6j7e3N8PCwtilSxdxPw8PD4aEhFClUgmbHGVx6zPeqZ1qNBrGx8fToSO39lV3d3cGBARQp9M5va9Wq0W5qNVq+vj4sFOnTgwMDGTXrl3p4+NDnU5Hu90ubPL09KSXl5dLO0j+7zzgTp06ISkpCTabDYqioKmpCSEhIXj88ccRGBiIGzduYMmSJRg+fDi6du2K+Ph4qFQqtLW1oaKiAq2trYiKikJQUBCefvpp3HXXXVi2bBnWrl0Lu90Oi8UCRVHQuXNnlzZcunQJd911F7y9vVFRUYGjR4+iubkZLS0tTn/n6ekJg8EAd3d31NfX4+rVq2hra4OiKNBoNFCpVHjooYcwY8YMREZGIj09HQ0NDXBzc8P169dx4MAB+Pv7t2tDU1MT+vTpA0VRkJqaCo1Gg/DwcDzzzDMYN24cVq5cifT0dGg0GkycOBGTJk2C1WpFXV2d03UMBgPsdjtGjBgBkqitrcXly5ehVquRkpKCRx99FFevXoXdbm/XDsczX716FTqdDseOHUNCQgKSk5NhMBjQuXNnuLm5YdWqVUhLS8PZs2dRUVEhPu/u7o62tjYcO3YMJSUlAACz2YxBgwahtbUVaWlpqKqqQllZGRRFQWJiYrt23LhxA83NzTh9+jRee+01lJeX49KlS8jLy4O7uzumTZuGoKAgUf4A0NraipaWFtTV1aGiogJ2ux2zZ8/GoEGDsHHjRhw6dAgARLnk5ubC398f3t7eLtuGgy+//BI6nQ4jRozA6dOncfLkSfj6+oryqqqqgkajQVVVFa5du4acnBzcf//9CAgIwDvvvIPNmzfDw8MDcXFxaGlpgVqtxvjx47Fp0yZRFvfccw/MZrNLG06dOoXKykqkpqbiu+++w3PPPYfm5mbU1taiqqoKkZGR6NmzJ3r27Im///3vWLZsGZKTk9HU1ITOnTvj4Ycfxr333guSCAgIwI0bN5CYmAgfHx+Ul5ejra0NISEhUBQFERERLu2YNm0aCgsLodFocOPGDVy/fh3nzp3D9u3bsWrVKqxcuRJ2ux1NTU2Ij49Hly5doNPp0NjYiG7dumHs2LGwWq3o378/evXqhczMTHh4eGDq1Knw9/dHWVkZzp8/j27dut3WBx1otVqMGDECu3fvRlpaGmbNmgV3d3d07twZer0eAHDXXXfBaDRiyZIleOCBB1BRUQEPDw+EhYUhICAAgYGBaGlpQUZGBhITEzFhwgScPXsW8fHxaG1tRXp6Onr37g0A0Gg07dqhUqmwbNkyqNVqFBcXw2w2o6SkBIMHDwYAlJaWwmw2Q6fTAQCSk5OxZs0aVFZW4rXXXkNrayuuX7+O8+fPo6SkBBcuXEBkZCSamppw7NgxADfbta+vLz788MM7NdH/nQfs8AzmzJlDo9FIg8FAnU7Hp59+mp07dxaeTlBQEAsKCrh48WIOGDCAeXl5XLRoEYODgzlo0CBarVaOGjWKiqJQq9VSr9dTp9Nx0qRJYqS9kx3p6elcvnw5DQYD+/Xrx549e9JgMDAkJIQLFixgYmKiGMV8fX1psViYkZHB0aNHU6fT0WQyiXtbrVaOHTuWGo2GNpuNWq2Wb7/9NqdNm+ZyRAfAZcuWUaVScerUqWLUzMjI4Icffsjo6GiqVCqmpqYyNDSUU6ZM4fjx4/nSSy9x5syZDAgIEJ6Uv78/Bw0a1K5nrNFoqCgKk5KS2rWjU6dOXLVqlfAIIiIixGcc15g1axbz8/M5atQoRkZGOs0yHB5EZmYmd+zYQZvNxqKiIlFO6enpol7xby/BVXm8++67HDBgAPV6PQcOHMgBAwYIj1+j0TAkJMRpBtC7d2+mpKQwPDycsbGxVKlUHDp0KHfs2EGz2cyoqCiq1WpGR0dz/PjxVKlUVKvVfPfdd/+jB9zU1ESDwSA8rU2bNjEkJIQA+NVXX3HAgAEcOXIkCwsLGR8fT41Gw7S0ND7zzDMcNGgQg4KC2L17dydPyGw2s6CgQPxsMplc2pGYmMg9e/aImU1kZCR1Oh1Xr17NoKAgjhkzhvPmzeOGDRu4ePFi5ubmcsCAAUxISGBqaip79+7N5ORk0R8wW5lhAAAgAElEQVQAMDAwkKtWrWJBQYHT7MTxb3t2hIeHs0uXLtTpdKyoqKDZbCYARkZGMiAggHl5eezbt6/wHhVF4csvv8x58+bR29ub48eP5+rVqwmAsbGxXLJkCbOzs4U3qNFomJCQQEVRmJeX57I83NzcqFarqdPp+Mwzz9BoNIr7xcbG0s3NjSaTiRqNhoGBgfT09BS/v/UZHW38scceo1qtpl6vp8lkYmFhIYOCgqhSqejj48OuXbu2a4fBYHBqg0lJSYyMjBSzgaVLlwpPdvDgwczKymJ2djbT0tKEzX986XQ6+vv7CxtDQkLo6+tLvV7PwMBAlx7wnxJgo9HIQYMGiUJwTNmPHj0qGldBQQGnTZvGfv368bnnnmNWVpaYwvn7+7OoqIhlZWU8ePAgp02bRj8/P44fP55RUVHUarXMy8ujTqcTwtSeHY7ps81m48KFC/nVV1/xypUrfO+992i1WhkREUGLxUKVSkWz2cy0tDQ+8sgj7N69O6dPn876+npWVVWxurqaVVVVPHz4MLOzs1lYWMgRI0aI5zMYDNTr9YyIiHA5ldm4cSPffvtt1tTUMDk5mVqtlu+88w4jIiJEx1er1XzhhRc4fPhwMdDY7Xamp6ezb9++fOSRRzhy5EgeO3aM/fv3p9ls5vjx46koilNIQK/Xt2uHVqvl4sWLWVBQwE2bNolGkpCQwPT0dJrNZo4aNYqdO3dmamoqw8PDmZeXx+HDhzMuLo5paWmMjIzkhAkT+NBDDzEnJ4d6vV40SA8PD27cuJGRkZFMTU11WS9xcXF0d3cXgq7RaLhp0yZarVbxDNnZ2Tx48CCzsrIYGRnJ0NBQpqSksEePHtTr9YyLi+PIkSOdQjD+/v7cvHmzaMyOurmTAPfv31904G3btvGtt96izWajoijs2bMn586dy6KiIk6ZMoVxcXHs168fBw0aJKaSarWamZmZ/Oabb5iamsqioiIxYAUHB9PNzY3l5eV3nOo6wmsOsXMMkG5ublQUhSkpKfTz82N2djbz8/P5xRdfsLa2ljU1NaypqeHbb7/NmJgYZmdnMyIigtOmTaOnpycfeOABWiwWvvTSSwwICBADUufOndu1w2g0MjIyksHBwUxJSeGECRNoMBi4YcMGrlu3jpGRkXSEj6xWqwgP+fj40Gw202q1MjMzk4MHD6afnx8zMzPFYFpSUsKXX36ZKpWKiqIwOTmZVqu1XTv8/f154cIFIcKzZs0SQqhSqdivXz/Onj2bMTExTEhIYExMDA0GAxMTE9m9e3cCYE5ODteuXUtFURgTE8O3336bkZGR3Lt3L/Pz80XfValULssDAF955RXOmTOHGo1GDJCOunQM2gUFBUxNTaVGoxF/5/i92WymXq9nbGwsR48ezdjYWIaGhrKqqorFxcW3DRr/JwKsUqlos9k4b9480bgB0GKxUK1Wi07u5uZGi8XC999/n1VVVSwtLWVpaSnLy8tZU1NDkmxpaWFNTQ3LyspYU1PDa9eu8fTp0/Tw8ODHH38sHrY9O259MEcsLSYmhj4+PsI7f/bZZ2m32zlw4EDa7XYaDAbOmjVL3P9WDh06xMjISO7YsYOPPvqok8fh6+vrUoADAwN54sQJbt++nUFBQZw9ezYfe+wxbtq06bbYktFoZGpqKjt37kwfHx8GBgYKD88hPunp6aysrGRZWRnPnTsnxElRFObm5rosD5PJRJPJxIkTJ9JisTApKUl4JWazWcS3VSoVQ0JC6OXlxYceeohZWVmMiIhgnz59GBYWxmHDhrGqqkrU2YULF/jRRx+xS5cu9PT0pFqtFqLsyg6bzSa8IQC0Wq1iIFq6dCkzMjJYUFDA+Ph4qtVqhoaGctq0aRw9ejTDw8OZmpoqhHjx4sViIPP09KROp+PEiRPZs2dPPv/883cU4IEDBwobUlNTmZycLNpHRkYGFyxYwNGjRzMoKEg81+DBg7lt2zbRXisrK9nc3Mzdu3cLodBqtUxJSaHBYOC5c+c4c+ZM+vn5tWuHr68vZ82aRZVKJWKkDg9/8ODBtFqtTEtLY3BwMFevXs2mpiandvn+++8zNjaWmZmZTEtLY2FhIVNTU2mxWJibm0sPDw8mJiZyxowZou+5Ko9HH32UUVFR7Nevn/ASPT09uWjRIs6bN4+hoaH08PDgmjVrWFZWxtLSUpaVlYlXZWUlq6qqWF5ezsrKSpaWlvLIkSMMDQ0VcXPHrNiV4FitVvr6+nLz5s188sknef36dfr4+FCv1zMlJYUWi4VPP/00582bxylTpvDzzz9nWVkZr1+/zrlz5wonJCcnh5s2baJarabZbBYOm81m44EDBzhy5Eghwu3ZoVaraTKZuG7dOo4ePZorVqzgyJEj6ebmRgCcPHkyFyxYwOTkZKd1lPz8fO7du5dFRUVMSEjglClTGBMTw/j4eP7zn//ktWvXSJKVlZXUaDQsKiriggUL7hgD/lMCHBcXR7VaTS8vLydxcby+/fZbXr58mcXFxbxw4QJra2tvE7s7UVlZSb1ezy+++EJ4Xq4a1K2v2NhYp0UCAPT09KRWq6XBYKBWq+Xzzz/Pqqqqdu/b1NTEy5cv8/fff6dDzG6doqvV6nbtCAsLo81m49q1a6nVannp0iVWVlYyOzubQ4YMYY8ePZyupdPpGBkZyfDwcIaHhzM6Opqenp5MTk5mWlqaI8RAkiwpKRGeAoA7zggcdWE0GllUVMSioiK6ublxxIgRXLJkCcPCwrhw4UJ2796d0dHRXLJkCZ988kn6+vpSrVYzPz+fly9fZllZmVO5XLhwQUyjYmNjGR4efkePz2q1csiQIWJmNGfOHPH8AQEBjIuLo9lsZmxsLNevX89Ro0bRarUyKSmJzz33HA8cOCA6mq+vLwcOHEgfHx+nejWZTHR3d+eiRYvuGBoyGAwcMWIEVSqVWLRyDAR6vZ5hYWGcNWsWk5OTGRwczMWLF7OsrIyNjY23tY9Dhw7Ry8uLWVlZnDt3Lj/++GMCYHBwMPft2+eyo6tUKu7cuZN+fn588sknmZaWJjpmcHAw//Wvf7FHjx5UqVRcv379bffdvn0709PTOXDgQE6dOpXh4eF86KGHhOMRFRXFvXv30sPDg1arlVFRUe3a4ebmxlmzZlGn01Gv19NsNnPu3LnctGkTJ0+ezP3794s+W19f324faY+Wlhb+/vvvfPPNN2k2m5mYmEgvLy+GhYW5LI+pU6dy9OjRHDNmDMPDw0Wd6HQ6hoWF0c/Pj3l5ebTb7Xz//ffFvaqrq/nKK68ID9TxvP/4xz+oKArd3d2pUqkYHh7Oq1evEri5yNqeHR4eHkxNTaWvr6/oMxcuXODQoUOFd+vt7X3bAuWTTz7JdevWMS0tTcwWAgMDmZiYyLq6OmFrZWUl1Wo1c3JyWFhY6LJe+GcFOCkp6bZYDAAuXryYR44cYUNDw39dee1V5s8//8zCwkIGBwfTw8ODZrO5XcMBcNGiRVSpVHRMewGwsLBQdNjMzEzOmjWLu3bt4okTJ1hRUfEfbWhubuZ3333nVOgO77E9Oxyj4/79+xkXFye86/79+9NisTA1NVWsgjoamWM1Ozk5mUVFRRwxYgRtNhsnT57M8+fPC1vKysoYFBRERVGYmJjI5OTkO3p8UVFRBMCsrCz6+/szMzOTw4cPZ3p6Ok0mEwMCAujm5kadTkebzUZvb2/m5uYyLy+Pnp6eLCwsvK08zp07J+Kvbm5uTE5O5vHjxxkQENCuHWq1WgwUycnJYso/fvx4Wq1WEZfWarVC9E6fPs2zZ8+yvLyczc3NvH79Ok+cOMHTp0+zuLiY27dvF/XgGPijoqLo7+/P6OjoO9bLyJEjqdVqGRMTw7CwMH7++ecMDQ0VA2tGRga3bNnCM2fOsLq62mW7aGho4K5duxgbG8vg4GBOmjSJR48eZffu3fnkk0/esV5CQkKoVqs5Z84c0Rbsdjv1ej1DQkJEKCU1NZW7du1yuu+uXbuYlpbGgIAADh06lDk5OezRowd37dpFRVHo4eEhdgOZzWZqtdp27VAUhRaLhatXr+agQYOo1WoZFhbGjRs3Mjo6mmFhYTx48OB/7B/tUVpaSpvNxtDQUI4dO5aTJ092aYdjZmw2m2k2m5mXlycGawCcMmUKw8LCaDKZqNVquWLFCqd7LV26lP7+/iI+q9PpGBgYSJ1Ox7i4ODFDefnll7l37947DowObxe4uVMqNTVV7H65tf/r9Xo+8sgj7NevH729vWmz2ejw5I1GI4OCgpwcp/LycrHGYjQaRUirPTv4ZwU4JCREbKkJCwujoigcMWIEx48fz5KSkv+60r799lsmJCSwT58+nDx5sohVnjhxgjabTXidd9reU1RURJVKJaZnjtimo/MVFhayb9++/Oyzz/5ru65cucLk5GRmZGSIykhPT3c5pQoICCAA9uvXjz///DNbW1tJkpGRkUxKSmJycjJXr14t4qaOuFC/fv2Yk5PDIUOGsKioiFlZWVy9erWw4/Lly4yLi6PFYuEXX3zhtHXNVXkYDAaGh4eLhRTH5xMSEuju7i6m4Iqi8IUXXuCkSZMYGhpKu93OGTNm8OzZs7eVx7lz55zire7u7uJ67dnh7e3NTz75RGwzS0hIoN1uF9P8Wxv2a6+9JqZsd+LChQscOHAggZsxZZPJxLCwMGGXq/JwlIFKpaLdbmdcXBzj4uL40UcfMTIykitWrODJkyfbDUm1x/Hjx6nT6ajRaJicnMz4+HiaTCYmJSW5rJe7775bdGibzcY9e/bw448/ps1mY+/evcViGAAuXLjwthnaqlWraDQaRagmMzOTbm5ujI2NFe1JpVIxLS2NcXFxLkMQDjsGDx7M3bt302g0UlEUhoeHMy4ujlarlXv37v2vyuGPVFVVMT8/nzt27ODgwYP54Ycf0s3NrV07DAYDhwwZQkVRuGfPHoaFhbFz586iz/r7+1On03H8+PG02+23CXB5eTnffPNNp/CnYzB1bG9Vq9XMzs4WgtyeHQDEOsuAAQPEQmJ2djaTk5M5ZcoUMTtTFIVvv/02z5w5wx9//JGFhYXi7x0L7MePHxc2Xr16VQyGr7/+uti+2Z4d5J/chlZSUoKamhoAN7eUKIqC7777Dps3b8awYcNw6tQpl5/95ptvsH79egA3t/9UVlaiT58+uHjxIq5cuYLvv/8e8+fPh9lsRltbG1pbW+Hm5ubyeqtWrUJbWxv279+P++67D3a7HWVlZUhNTUVUVBS2bt2KkSNHonv37v/185lMJowdOxaXL19GXFwccnJycPToUZfbjPR6PQIDAzF37lwUFBSgoaEBAODr64tz585Bo9Hgvffeg7e3NwIDA0ESU6ZMQffu3fH999+juLgYx48fx969e7Ft2zZxXcd2Fm9vb6SmpqK2tva2rWt/pKGhAVevXkV1dTUA4PTp07hx4wZOnz6NhoYGXLp0CXV1dfD19cWePXuwb98+XLlyBd26dUNhYSHCw8PFtU6cOIFevXohJydHXFulUqFr1644efIkQkND27WhtrYWL7zwgtgu9c477yAvL09saUpJScGaNWuwZ88e5Ofnw2Kx/Mc6qa+vx++//46ioiK0trZCo9HgypUrwi5XJCQkoGvXrlAUBadPn8aiRYuQnJwMvV6PMWPGID09HXfddRdMJtNtn926dStefPFFp/dIoqmpCQBQV1eHo0ePolevXvj1119d2lBTU+Po8Lhy5QqeeuopvPLKK/Dx8cGDDz4IHx8f8beBgYHw8PBw+vyAAQOwbt06LFiwAA8++CC0Wi3q6+tx7NgxaLVa3HPPPSCJkpISnD9/Xtj3R0pKSmCz2fD7779j9erVaGpqAkmcO3cOAQEByMnJuW2r5eLFi9GrVy+MGzcO27dvd/mMTU1N2LNnD0ji6tWryMjIQGNjY7t/29jYiIEDB0JRFPzwww+4evUqrly5gr/97W/IyspCXV0dDAYDBgwY0G4bO3v2LPbs2YPa2lqkpKRg+/btmDVrFtra2tDc3IwpU6agtbUV33zzzR3bRnR0NOrr67F9+3bce++90Gg0aGlpwb59+3D69Gl8+umnCAwMRFFRERYtWoR7770XnTp1Qrdu3RAUFASr1Yr8/HzMmTMHNpsN0dHRAG5uAx03bhw2bdqE1tZWbNmyBXa7HZWVlS5t+VMCfOs+28uXL4sHr6iowJgxY7B582YUFxfj2rVrqK+vF5/bsWMH3n33XSxcuBCjR4/Gli1bYLFY8O233+L7779HbW0trFYrNm7ciGvXruGll17C6NGjXVak0WjE5cuXMXToUHTt2hXHjx9Hc3MzunbtiuLiYpSUlODxxx9Hbm4ufH19XT7PoUOHkJubi6KiInHdYcOGITQ0FNOnT8eBAwdQXl7usjJJIjMzEy+++CKeeOIJsW+woqIC4eHhaGtrw6+//gpPT0+88sorGDBgAM6ePYsffvgBly9fxu+//45vv/0WpaWluHz5sriup6cnpk+fDrvdjsceewxpaWl3rBfHfevr65GWlib2D2s0Gmg0GiiKgqeeegrx8fG4ceMGYmNjERISAh8fH9jtdgQGBoprXb9+Hdu3b8e+fftw5swZADf3CXt6euLcuXOYO3euk6230tTUhC5duqBHjx5QqVSYO3cuzp07h7CwMMyZMweNjY2IjIxEWlqaU7188MEHuO+++/D3v/+93TK+fv06vvzyS8TGxmLWrFloaWlBUVERrFZru3a4u7tj7Nix2LdvH/z9/VFfX48bN25g4sSJ2LZtGz755BOnfdC3smnTJnz++ediL6kDLy8vPPLII8jNzcWsWbNAEj///LNTO/8jpaWlAIDhw4dj69atePDBB1FSUoJ58+Zh27ZtKC0tFXt616xZgx9++MHp8wEBAUhOTobVasX58+exf/9+8XyDBw/GmTNnQBIXL15EYmKiSwFua2tDZWUlsrOz8cknnyA9PR1BQUFQqVQ4c+YMvv76axw7dgz19fWYN28e7rvvPrz55pv4n//5H1RWVuKnn37Ca6+91u7zjRs3Dr169cKSJUtgNpsxZswYtLW1uayX5cuXo62tDQsXLoSPjw98fX3x6aefoqSkBD4+PnjllVfw9ddf4+DBg7d9/vjx49i1axdaW1tx5swZzJ8/H1u3bgVwc1DctGkTgJv9oLa21mW9XLlyBd27d8eFCxewefNmtLS0wMPDAyaTCePHj0djYyP27t2LjRs34vDhw7ddq7q6Gvv370dlZSVmz54t3lcUBc3NzVi8eDGWLFmC6upqFBcXu6wXAGh/p/IdSEhIwOHDh4VHVldXh7Fjx6JPnz5Yvnw5PvroI1RUVGDUqFHo1q0bgJubq8PCwnD06FHYbDZ4eXmhuroav/76KxoaGqAoChRFQWNjI/r164edO3eivLwcWq22XRscDzR27FjMnDkTx48fx+rVq3H27Fls2LABw4YNwwMPPHCb+K5fvx4bNmwAAPj4+KCtrQ1btmxBUlKSKEB3d3ccO3YMy5cvR1VVldP9/khLSwsiIyOxcuVKbNq0yWnj9+XLl9HQ0AC1Wo0ffvgBFRUVuH79OkpLS9HS0gJ3d3fcuHEDtbW1SExMxAsvvCA+29DQgMOHD2P06NEoKChARkYGFEVBp06d2rVDo9EIG8+dO4e2tjacPXsWjY2NsNlsqK+vx5kzZ1BdXY36+nrcfffdaG1thZ+fH4YPH+50rXPnzmH9+vUYOXIk9u3bB4vFgnPnzkFRFGRlZd1RcGw2G86dOyd+/uKLLxAQEIDu3bvj0KFDePzxx2/7ssDq1auxYsUKKIpy20zj6tWrWLlyJWpqanDkyBF4enri008/BQAEBwe77GQkRT3X1NTA19cXb775JrKysrB7925ERkbC3d3d6TPffPMNtm7dip07d+LatWsICwtDenq6+L2Hhwfuu+8+LFu2DBs2bABJXLp0CQCgVqvbtcNRJ8eOHcOqVatQVVWFsWPH4ttvv8WePXtQV1eHy5cvw9fXF4cOHUJhYSGKiopw3333iWucOHECa9euRWVlJa5evSqu+8svv+DatWuwWCzQarU4c+aMS0+rvr4eTU1N+O6773Dt2jWcPHkS3bp1Q3BwMC5cuIDi4mK89NJL2LlzJ7788kuUlZUJ2w8fPozQ0FA8+OCD7V53586dGD58OHbv3o3U1NR2739reRw5cgS5ubnYuHEj3NzcoFarUVVVJb7MdOTIEXzxxRdobW29bXYSHh6OnJwcXL9+Hb/++it27twJk8mEhIQETJs2DStWrEBWVhY+/fRTlJeXu6wXtVqNjIwMnD9/Hj/99BNCQkJw9epVXLt2DXv27MHgwYOxe/dulJSUYPjw4U7eeHR0NLp06QKj0YicnBwkJCQ4Pd+FCxcwbNgwNDc3w2QyuRzoHfwpAbbZbCgvL4ePjw+CgoKg1Wpx5MgR/Pbbb3j++edx4sQJNDU14aGHHnKaXnXq1Anh4eGorKzE6dOnUVJSgtraWphMJpBEWFgYTp48Kb6ZUl9fj4qKCuHZ/ZHm5mYAwPfffw+VSoX77rsParUaX331FSZOnIhBgwa1O7Wsr6/H+fPn4efnB61Wi127dqG5uVlcD7gZhnjqqacwbdo0hIaGorGx0aXo1NfXY/PmzbeN+N7e3vj999+h0WgQGxuLs2fPgiRiY2NhMBjg6emJ5uZm7N27FwEBAXjooYfQt29f8fnGxkbs27cPPj4+qKurw6FDh0BShH/+SFNTE8aOHYv3338fKSkp2LJlC4qLixEdHY2IiAhUVlZi06ZNojF88cUXSE9PR58+fZy+bfjzzz9j+fLluOeee3D06FEMHjwY69atE98Ce/DBBzFlyhSXA1JtbS1+/PFHFBQUoFevXtDpdDhx4gR++OEHFBQUIDc3F15eXuLv161bh4ULFyImJgaTJk1CZGSk+N2JEyfw3HPPYe/evVAUBXa7HUeOHIGvry9CQ0OxcuXKO4ZlvvvuO/j4+KCmpgZtbW04ffo0tm7disuXL0Oj0aC+vh4LFizAzz//jMDAQDQ2NuLgwYM4e/YsqqurceTIEafrXb9+HR999BHCw8OxceNGp9+pVO1PJP39/TFmzBi88cYbmDFjBrZs2YJDhw4hOzsber0e8fHxOHPmDBobG9GnTx8cP34cJ0+edBLgGzdu4LfffkNlZSWMRiO6desGq9WKLVu2wN/fHzExMVAUBV9//bXLsnDMWh0hH0d4rLm5GVVVVdDpdEhKSsKNGzdQXV0Ng8EA4KbnPmzYMERFRbUrrmq1Gp6enqisrIS/vz+OHj0KLy8vBAUFtWuH2WxG165dERERAZVKhdraWnh4eCAkJASnT5+Gm5sbvvzyS1y8eBEGgwFLly6Ft7c3MjIyANycWR47dgwNDQ0ivKNWq9G5c2eUlJSgqqoKjY2NePXVV/HRRx+Jb1H+kcbGRvj5+SEpKQl33XUXPvzwQxiNRly/fh379+9HS0sLysrK8Nhjj2HUqFFiplVcXIwff/wRpaWl6Nmzpwhvnj9/Hi+++CJqa2vFDHf79u2iHP442N/Kn/aAL168iISEBFRXV6NXr144duwYysvL0dLSgnPnziEqKgq///475s+fj4ceeggpKSkAgJ49e4qp/qlTp3Do0CHo9Xro9XrhybS2tuLYsWMICAjAPffcg8OHD7u0Y/jw4fj6669RW1uLS5cu4dVXX0Xnzp3Ru3dvJ/F1hEQOHDiAL7/8UohYY2MjiouLb7tuS0sLiouLQRKVlZVobW11acONGzdw4MAB8dXpW99va2uDwWDAtWvX0NDQAIPBICq5qqoKvXr1wv333w9vb2+nOHVVVRW2bduG/v3747PPPkNbWxsuXLiAsWPH3tbxb7X50qVL8PLyQklJCeLi4nD06FG0tLTg/PnzaGhowEMPPYQdO3bg+PHjaGpqQlJSktPo7bj3zz//DLPZjP3798NsNiM6OhoJCQnYvXs3jh8/jsrKSpdhnerqalitVvzyyy947rnn8MUXX+Dq1av461//itGjRwvxXbp0KX777TecPn0a99xzD8aNGyfaiYOysjJ89tlnMBgMuO+++/DLL79AURR4eXmhrq4OP/zwAwICAlyWx/Dhw1FWVoby8nIoioKoqCjcuHEDGRkZ2L17N15++WUcPHgQ586dg8ViwcSJE/HEE0+gtLQUdXV1iIqKcrpmXV0d9u3bhwEDBqCpqQl9+/bF7t274eXl5XKm1tjYiPj4eCiKgl9++QVlZWX46aefYLVaUV9fj6qqKlgsFiQmJkKj0bTr0Tc2NqK0tBQajQb9+/fH+fPncdddd2Hq1Kk4fPgwrly5AqPRKGam7aFWq9HW1oaLFy8iPT0dZWVlqK6uxsWLF9Ha2orevXsjPz8fiqKgZ8+eaGtrQ0tLCxISEm6rFwelpaWYM2cOSOLEiRNoaWnBxYsXcfHiRSfn64/1UlVVhRMnTginxc/PDw8++CAOHDiAqKgo/Pjjj2htbYXZbMaxY8dw+vRpIcAlJSXYv38/VCqVWMSqra3FL7/8goMHD6KsrAz+/v6inTri73+krq4OVVVVOHPmDNLT0xESEoJTp06hT58++PXXX/HTTz9h8uTJePzxx4X3+69//Qvr1q3DwYMHERgYiDFjxojrlZeXY/Xq1bBYLGhqasLXX38NRVFEfd5pLetPCfC1a9cQGxuLyspKnD9/HhaLBS0tLbj33nvxyy+/IDAwEAMHDsSWLVvESOjt7Y2oqCjY7XZxlsGpU6dQU1ODVatWQafT4caNG3jqqafw+uuvw2QywW6349q1a06e6a14eXnh6aefxuOPP47Gxkbs2LEDAJCbm+vkjX7wwQfYvXs36urqcOHCBZSWlqKqqgqXLl2Cp6cnWltboVKpxPfQgZvT/w8++AAARAjCVQE67Gtra3Oq7KqqKqjVajQ1NSEjIwPBwcHw8vKC1WpFRUUF6gk3qR4AACAASURBVOvr8Ze//AU9evRot4w/+OADTJ06VUy3gZvepcMz+SNmsxm9evXCDz/8AHd3d8TExKCsrAyVlZWoqKhAWloaQkJC8Ne//hX//Oc/kZaWhtjYWKdrXL9+HZcvX4bBYMB3330Hm82Go0ePws/PD926dcPu3btRU1MDT09Pl564oyyqq6tx5v+1991BUZ5r+9f2CgvLAsvSYYWVpR1AIUAUQlGCoCQ2jkbhQ0WiOepIYowlEqPnqElMYgvGbjRqii1oosagxt5L0KhYICK9ShPk/v7wvM+wgSUn3+/MZH4z3DPvjGX3fe/3Kfdzt72ue/fwxRdf4NVXX8WUKVNMCkxKpRLW1tYYMmQIoqKiWBGDk/Lycpw+fZqrnMPPzw/nzp1De3s7jh07Bj8/P1RVVbEUQHfz0tzcjNbWVuh0OlhYWLDx4PP5sLa2hlgsxtixY5nxHDRokFljU1JSghUrVkAkEuHFF18EANy8eZO9r7liYm1tLSsYrly5ErNnz0ZwcDCuXr2Kjo4OPHz4ENnZ2WhubsZ3332H3377rcs9mpubUV1dDRsbGzQ2NuLJkycAAKPRiLKyMvTr1w9r1qzBxIkTzRpgbk80NTVBLBbj6dOnUCgU8PPzQ2FhIV577TUYDAaoVKo/LFo/ePAAGzduREVFBT7//HPIZDJWBwKe1y/M1UyePHmChIQEnD59Gi+88ALOnDkDd3d33Lt3Dw0NDXj69Ck6OjpgYWHBvGuVSsW+r9Vq4e3tjdu3b7N34ozwb7/9xtKYW7du7bEIx6VsfvnlF3z55ZcIDQ1FY2Mj2traMGXKFPB4PIwbN84k9cDtK4PBgKCgICQkJAB4nib74YcfYG1tDaVSifLycvB4PGi1WggEAvB4PLP7FviTBlgmk2HWrFlYunQpgoODceXKFbS3t+Px48e4fv06/Pz8UFNTw8BsCgoKUFtbi6FDh5qEMH369MHcuXPR1taGffv2YdCgQRgwYACWL1+O0NBQiEQiXLx40azrzufzce7cOVRXV4PH48HNzQ1FRUWwtrY2yfsUFBRg//79CAoKglKpRGVlJfh8PjQaDfr164eKigoUFBSYGGAiQltbGwYOHIhjx44BgIl321kEAgGePXuG9PR0k/zvpEmT0NjYiPb2drz++utmc7e/l4qKCuzatQtyuRynT5820enu3btmc1q2trZQKBSor69HQEAA8/ra29vRt29fuLi4YMeOHYiIiEBKSgpiY2NNwGxu3bqFAwcO4MaNG2hsbMSwYcMgFApx6dIlXLhwgQEp3bx5E1VVVaiqqupWD+4gUKlU2LRpEwoLC2Ftbd0lRO8un8hJe3s7rl+/jm+++QY6nQ7l5eW4evUq06GoqAjLly9HTk4O6/joTo4cOYKYmBhIJBI8evQIkZGR2LhxI8LCwhAXF4fk5GTo9Xqz3mtn6ejoYOA5FRUVkEqlDLSovr7e7DoVCASws7Njf+cAocRiMct5azQa7Ny5E8XFxd3eQ6fTITg4GDdu3MDdu3ehUqnA5/PxxRdfoKmpCQEBAWhqauqxA4nH4yEtLQ3+/v4oKCjA5s2b0bdvX7Zf+vXrZ2Lo/mgsioqKsHfvXtYZolAo4OHhgXv37kEoFJodUyJCeno6du/eDQcHB/Tp0wc1NTU4fvw4/Pz8cPLkSdTU1EAoFLKourMT5uPjg5SUFGzbts0keuXWl0QigY+PD06ePAmBQGDWE+fxeGwNnzt3Dq+//jpOnTqFM2fOYP369V2iH+B5R0p8fLzJv927dw+rV6/Gd999B1dXVxiNRla4FggE0Ol0rFBqTv6UAea8OCsrK6hUKuj1ety5cwe3b99mFfa8vDw4Ozvj9u3bqKurQ1FREZ49ewaVSoW+ffuye3l4eGD27NkoLS3FmDFjsG/fPri6urK8oZOTE6qrq7vVo7q6Gu+88w70ej1DX+PxeJg7d65JZdza2hpSqRRKpRKlpaXslBQKhVAoFAgLC0NUVJRJyoLP50Or1cLOzg62trZobW01m/O0tbVFaWkpvLy8sHHjRqSlpUEkEmHu3Lndfv748eNobGwEn88Hn8+HhYUFpFIpSwX89ttv+PjjjxEUFIStW7dCLpejqakJPB4PdnZ2uH//frf3bWlpwf79+0FEOHHiBBoaGvD48WPY29sjKioKx44dw+3bt3Hz5k1s3769i5dz7tw5fPLJJ3BxcYG7uzuUSiVKSkrQ2NiIp0+f4vTp0wgNDcWdO3fMGl/guWeRlpaG7OxsFBYWwtbWFu+//z5sbGwwcuTIbvPyvxcu/3rz5k14eHigpKSEVeNdXFxQWlqKyspKXLhwwWx7IFfUnTNnDubOnYvr168jIyMDIpEIb731lsk65KSsrAwCgaDb9IqTkxMWL16MlpYWLFq0CD4+PqzV0M3NzeyYKBQK3LhxA0QEtVqNjz/+GNXV1ZDJZGhuboaVlRUqKyuZzt1tVHd3dwwePBiPHz+GpaUla+E8ceIELC0tsX79+j+MSpRKJdzd3WFpaQl3d3fw+Xy4u7vj/PnzkMvl2LdvH+zt7c2mdDqLh4cHZsyYgYMHDwJ4bpBlMhn+8Y9/YPr06aitrTUbEQiFQqxfvx6//vorbt++jaSkJBw8eBBtbW1wcnLC8ePHoVKpUF9fj7a2NpSVlZkUserr6yEWi9GvXz+UlpbC0tIStbW1KCoqgkgkglQqhU6nQ3JyMr799tseD5Xi4mL87W9/w+XLl1FYWIibN29ixIgR//FBBDyPSuvr6xEaGorq6mpYW1sjNDQUly9fxqNHj8Dn8xlapDn5Uwa4ubkZK1euxMmTJ9lma29vh06ng5+fH3bs2AELCwuEhoayCn9ERAQeP36MVatWYfDgwdBoNMwbtrOzY7mUq1evwtraGnfu3MGIESPw448/svCtO2lvb0dqaio6Ojrw3nvvwdvb26Rod+LECezduxclJSXMU9fr9ejo6MCtW7dQUlKCrKws1gEBPC9mcZ0aXPtYTwtbJpNBrVZj1apVsLKygkwmg4WFBZqbm0FEGDx4MC5dugRra2tUVFTgyJEjaGtrQ0tLCxobG2FjY4OQkBBmgJVKJYsiFAoFG3Nvb2+cPHnSbFhVWVmJpKQkHD9+nLUqKZVKqNVquLu744cffkB7ezsSExNN+n2B597v5cuXWUjn6uqKjRs3oqqqChqNBkKhkBUriAhnz541e6I3NjYiPz8fDx8+BBHBz88PGo0Gd+7cQWtrKzo6OnDq1CnU1NQgKCgIHh4eXbyl+/fv4+jRoxCLxXB0dGTQiXw+H7a2tqitrWUHnLnx4PF4ePbsGSIiInDhwgW0tbVh/vz5GDhwoEkRkJPr16/j8ePH6NOnj9n8tlarxcqVK6HVahEREYFDhw7h4sWL+Nvf/oaTJ0+a1YNLkyiVSrz88svYsWMH+Hw+WlpaUFNTgz59+rCe4o6Oji7PVyqVcHJyglKphFKpRFVVFZ49e4aXX34Zbm5uOHLkCAuNz5w5060eAPDuu+8iODgYOp0Oz549Y5GqnZ0dVqxYAQCsCOrt7c1ymGq1GjqdzmTcOOMIPG+BjImJgY+PD4DndRxzdRM7Ozt20JeVlcHOzo7lc/Pz8/Hs2TOEhobi6tWr6Nu3L3sGJzU1NSgsLER9fT10Oh0z4EKhEA8fPkRrayt27NiBjz76CLt27TLbgSAUCmFjYwO5XA69Xo9PPvkETU1NWLJkCeuHLi8vR2NjIyorK+Hi4tItJK2fnx/ef/997NixA3v27MGdO3dgMBggEonw888/My+9pzrSnzLA9fX1cHNzg7+/P4qLi1FXVwej0YgffvgBBQUFUCqV8PX1xcWLF1FcXAxPT0+0tbVBJBIx77hzRVWj0SA2NhbZ2dmwsrLCTz/9hOHDhyM5ORmnT5/uUfG2tjbU1taioKAAz549g9FoNNnMR44cQW1tLVQqFfz8/NDY2MjyMVKpFNHR0cz41tXV4cSJE3j69CnOnz+PX375BZmZmXj33Xd7LG7cv38fsbGx0Gq1cHZ2xrvvvgupVMo23fvvv4/c3FzmycXExMDJyYk1wZeVlcHV1ZXdT6VSYfDgwSgsLMS5c+fYAu3bty9u3bpl9nSWSCT45Zdf8MILL7DcqYWFBe7fv4+qqiqIxWLw+XxMnDgRAQEBJt89fPgwdu/eDZ1OxzyrkJAQ/PTTTxAIBIiMjERwcDCOHDnC/YzTbJ+nRCJhh5FOp4ODgwNmz56N6upqXLx4Effu3cOyZcvw7NkzjBw5EiEhIejXr59JiubevXs4e/Ys+vXrh0GDBqGiogLA86jn0qVLrP2Mz+ebza3x+Xzo9Xpcv34dwcHBOHPmDGJiYrBmzZou3tmVK1dw6NAh9O/f30SP0tJS9kMVTtRqNT766CMAQGJiIo4ePYrHjx+bjdSePHmClStXYteuXRCJRBg5ciQOHjyI5uZm8Hg86HQ6HD9+HA8ePMDw4cNhZ2dn0gnCPdPLywtyuRy2trY4deoUmpqakJOTA3t7e3z++efQaDTIy8vr8kMOTtrb28Hn83Hz5k1MmjSJeWhSqRR37tyBk5MTtm3bBqlUirKyMrzwwgusQ8nJyQlhYWFd2hU7j7VUKsXcuXPh5OQEoVBo0orYWRQKBWQyGRwdHVnLJ5fe44q7NTU1aGtrQ//+/XHt2jWT77/00ksoKytDeno6wsPDERgYiEuXLkGr1bK2SwCYM2cO0607aW1txZw5cxAdHY0vvvgCP/74IwIDA5kD9/PPP6OgoADl5eU4f/48JkyYgKSkpG7vdfPmTbzzzjt48cUXERwcjA8++ADp6emwsLDAwYMH4e3tjV9//bXb7wJ/0gC3tLTA09MTcrkct2/fxtGjR+Ho6MgMpdFoRGlpKXg8HkQiERobG3Hz5k2kp6dj/Pjx3d7v0qVLuHjxIrZs2YJXXnmFncz9+/dHfn5+t3oolUq4urris88+Y4Wyr776Ch9++CHbYDExMTh27BgaGxtx+fJlBhb+0ksv4dNPP2UFl7q6Ouzfvx/Lli1DcXExZs2aBQ8PDzQ3N4PP5yMpKQnXr183OyYqlQo7duyAt7c3nJycoFar0draCpVKhX/+858oLy9HTU0NnJyccOXKFdy7dw8ODg5oaGjAvXv3kJeXxwo7VVVV2Lp1Kx4+fAipVAq9Xo/KykocP34cAwcOZGDPv5eGhga89tprePr0KW7dugULCwsUFxfDxsYGX375JUpKSroFyX7w4AFKSkrw5MkTPHnyBL6+vlCr1Xjy5AkUCgWsra0xcuRIbN++He+//z6uXLmCa9eumS1K1tfXw8LCApaWlhg0aBBKSkqQm5vLmtkLCgogEokQFxeH+/fvY/v27UhJSUF2djY7iMRiMdzc3KDVarFp0yZcunSJ3V8ikaC8vByZmZnIzc01+0MMsViMiIgIzJs3D6NGjcLFixexatWqbkPjAwcOICQkxKQNsKGhAV9//TX27NmDRYsWQa1Wo7a2Fv369WOf4fP5CAsLw7Zt28waPrFYjJ9//hlExNazjY0NxGIxXFxcEBsbC4FAgCdPnmDz5s1wcHDotl5gaWkJGxsbXLhwAUKhEDU1NZg+fTq0Wi2EQiFOnTqF2NjYLj/k4ESj0bDWx9dffx0RERGIiorC2bNnGamBXq9Hnz59cPToURw+fBguLi7o378/fvvtty75ZT6fD4VCAbFYjJqaGnawjRw5EhUVFWYN8P379zF48GB899136NevH7Zv324SxVRWVoKIoNfrsW/fPpSUlMDf3x8PHz5k60OhUMDT0xMCgQAbNmzAuXPn4OjoCLlczjo36urqYGFhYTb0F4lELIefmpoKBwcHTJo0CTdu3EBTUxPeeOMNODs7s1RI5/pOVVUVHjx4wIqh165dg6enJ9LS0jB16lRIJBJ89dVXOHToEK5cuYIXX3wRd+/e7VYP4P/ACZebm4tly5ZBp9PB29sbLS0tICKEh4fj5MmTeOWVVzBz5kzExcVh6dKlyM/P79b4PnnyBEePHsWyZcsgFouxbNkyREREMO90+/btZhHtvb298corr8Dd3R02NjaQyWSssMHJ1q1b4erqCm9vbyQmJrJC39///neTavf169exZMkSTJ48GXV1dfjwww/x9OlTfPHFF9DpdFizZg1LB/xeeDwevvnmGxgMBqSlpeHWrVusFayurg6+vr7Q6XT4n//5H/B4PISEhCAtLQ2lpaUoLCyEvb29STHB2toar7zyClJSUqBUKvHzzz9DLpfDz88Pa9euZfnC7iQzMxMrV66Eg4MD3njjDSiVStZuxLE6/N6D3rhxIz7++GO0tLTA0dERQUFBePToEX788UcWVaxduxYSiQQ2NjbYuXMnAJj1xHk8HhoaGlBYWIilS5eiuLgY+/btQ1hYGIYMGQJXV1d4eHjg6NGjCAkJwauvvorTp0/j8OHD7B5NTU24cuUKCgsLMX36dAiFQuh0OkgkEsTGxgJ4XsRTKpWs4PF7aWtrw8aNG5GXl4cNGzb0mIN75513TIorZWVlyMvLw4EDB3D06FGkpKRg1qxZ+Pvf/46ioiJcvHgRly5dwu3bt3Ho0CHweDyzEUFzczN27NgBIsLOnTtx7NgxjBgxArdu3UJ2djaSk5MZ+4KHhweKiopw48YNtrk5aWlpQUdHB1566SVYW1tDo9FAoVDg5ZdfRlNTE2xsbHD48GGzP8R4+PAhhg8fjgULFkAul6OyshJ6vR67d+9Gfn4+du/eDVtbW5SUlLCuIHd3d7S1tSE4OBjZ2dkm9xMKhdBqtTAajbCxsWGpNKlUigcPHpj1PDv/kra+vh7e3t4mhWUbGxukpqZiwYIFMBqNUCgUWLduHTZv3sw+YzQaERMTg+rqakyZMgWOjo6Ijo5GTU0NJBIJ3nrrLQwbNgyxsbFmi4FcCpCzFxKJBCdOnEBGRgYWLVqER48eQavVoqKiAjNnzmQdDwBw/vx5TJs2DYmJiRgxYgSWLFkCd3d3VFRUwNfXF15eXtBoNJgzZw6Cg4Oxbt26ntlbzIFEmAOx4PBI0QktCHgOOG0wGGjjxo3/EfLYqVOnSKvVko+PD61YsYJsbW3J09OTUlJSSCgU0tq1a8nJyalbEAuj0cjQsRISEshgMFBkZCSVl5f/4XM5aWpqoocPHzJEKPwbbGb06NE0fPhwhh88e/Zss+Az3LsvXbqUAYokJSWRTCYjV1dXxgiRl5dHiYmJJBAISKFQkIuLCwUFBZFer6fXXnuN6XThwgXS6/V07do1CgwMZNxjBoOBoqKiegQXAUDbt28nkUhEmzZtYkhkiYmJ9NZbb9GDBw+6jMHixYvJ19eXQkJCyMfHh+Li4kggEJBKpSKNRkO+vr7k6urKQE6455hD+FcoFAwoBwBt3bqVFi9eTBcvXmRza29vTwKBgOzt7enzzz/vAlm6b98+CggIoMjISAJANjY2NHXqVLK1taXCwkKGeAWYx3t1dXVlgC39+/cngUDQBWqzO2lpaaEVK1ZQWFgYA1DiLpVKRampqWRjY0MKhYLGjBlDY8aM6RGek/s/Dpg9ODiY8eDNmjWLMjMzyc7OjmxtbcnJyYlCQkIoKyuLTpw40UW3M2fOkJubGwUGBtK0adPI3t6eramRI0f2iJ/Nrc0VK1YwuMXKykqT+3NrffDgwWQwGGjz5s1mx+nChQsMRSwzM5MGDRpEb775JoWEhFB9fT0pFIpu9ejMAafRaEx4GJVKJQ0fPpyOHz9OU6dOpdDQUIqPjydPT0/Kyclhz66rq6Pc3FwSi8VkZWVFsbGxbEx5/+aR4/F4NHToULP7ViAQUE5ODvH5fFKr1TRjxgzSaDQ0cuRIsrS0ZIhtbm5u9P3337NnV1RU0Pfff0/Tp0+nhIQESk9Pp4yMDJoyZQpD2RMIBCawrbx/cyN2pwfRn0RDw78xXjMyMtiEe3t7k5ubGyUkJNC9e/fMThrRc7jHqqoqKi8vp/z8fKakhYUFZWVlkV6vpxUrVpBCoaDLly+T0WjsVnFuQ4wdO5YZvoMHD3aL5fp7aWhooLKyMjp8+DDFxcUxEkAbGxuSyWQ0dOhQGjVqFNnY2JC/vz/dv3//Dxc2N/Cdoezy8vJo0aJF5OvrS19//TUlJSWRvb09OTs7k5eXFzk7O3eBgbx27RoZDAaaNGkSLVu2jFQqFT148IDu3r3LWBrMLWxuojnD2xmR/9KlSyZjUFVVRcXFxbRx40YKCQkhNzc3ioyMJBcXF1IqlWQwGEiv1zN2jZkzZzLmBW6+utOjT58+9PXXXxOfz2cbKyQkhEEyZmRkkLOzM4lEIlqwYAEtXLiQfvjhBxPdSkpK6LPPPqOIiAi2sQQCAXl5edGUKVNM3tOcHv7+/jR06FATJg5zBpgDv3/w4AEdOHCAtm3bRkuWLCGDwWACJcq995QpU8hgMFBOTg4BYLCG5uaFI9LkgMTt7e3J2tqarKysKC4ujh1ycrmcfH19adq0aXT69Okuel67do2SkpJoxowZ9Nlnn5GdnR2bD26szaH2cQSm3GcTEhLMYmP3JC0tLVRaWkrHjh1jJLr/+te/yMLCgi5cuED+/v60adOmHnF4OT0WLlxIEomEkaUmJCSQlZUVGY1GCg8Pp7CwMHJyciJbW1v6+OOPmQ67d+8mV1dXZuQ4dhkOBpX3bwqv+/fv94imOHXqVOLxeDRp0iQSiUQUGRlJVlZWpFAoKCsrixITE03moaGhgWbOnEmOjo60cOFC2rBhAxkMBkpLS2PrpDsC2p4ORvq/GOD+/fuTSCQikUhEdnZ2ZG9vT48ePepx4hoaGqi0tJROnTpF48ePp7i4ONqzZ4+JwdLpdHTr1i2qqqoiAMwr6E4PHo9Hc+fOZZRDUqmUZDJZt5CKnaW1tZVWr15NUVFRlJWVRTNmzKDMzEzy9PSkTz/9lEQiEQMNF4vFtG7dOjbJ5sajs/6WlpbsnYYMGUJqtZpUKhUFBQWRo6MjOTg4kEqloujoaEpISCCBQECpqalMv6dPn9JPP/1EVlZWjAKns8dnzvPUaDRkaWlJjx49ovz8fLKysiILCwtSKpXE5/O7GOBp06aRk5MTGY1GCgsLI29vb9JqteTt7U0ODg4UFxdHXl5eDBbU0tKSgbmLxWJSq9V/OB5hYWEM7D0vL4/c3NzotddeI2dnZwoICKDQ0FAKCQmhdevWmYBZ79y5kzEVR0ZGkqurK/n6+tL169cZJKdAIGCYuN3pwfEUchEBAPrll1+otLSUSktLqbKyksrKyqi0tJSSkpIYiwoA5gFFRkaazMHv4TSB54DfMTExZjcYF0WMHz+eEhISaPDgwQxadNasWeTi4kJ5eXlUV1dHw4YNIx8fH3J2djbx+Dg5duwYY/EeO3YsffPNN+Th4UHW1tYUHh5OIpGIlEplt3pwsKlchCYUCv+jiICTpqYmKisro2+//ZbCwsJo8uTJzANeuHAhyeVyMhgMJJfLSSqVkqOj4x+uj/Xr15NKpaL4+HiSSqUUEhJC1tbWFB0dzbCrw8PDacuWLSYRdUVFBS1dupQZWwBkZ2dH27ZtY1Esdyiag4Hk8Xi0YcMGE4MpFospKSmJRo0axfbP4cOH2XM//fRTcnFxIYFAQCEhITRhwgRKSkqi9957j95++20Si8XswOecMaPRSB4eHv9dAyyXy8loNFJcXNwfMl5w3GuzZ89mYNx8Pp/kcrlJGsPS0pKF7h999BFJJBIqKCgwq7hSqaTw8HBmHDtzalVXV5vwvTU3N1NDQwPV1tbS3r17aeDAgSQSicjHx4fefvttWrx4MQmFQpLJZCSVSmnYsGG0YsUKxh4AwCwAeecJlMlkpNfr2XfUajULZcLCwmj8+PEMn9bJyYkiIyPJYDCYGGAutOt8yeVy9q7mxkMkElF8fDwJhUKSSCQUFRXF8IzlcjldvXrVZF5yc3PJ29ub4ZpylOsuLi6UkpJCRqORgXYbDAaaN28eVVVV0YQJE5jn0dMG644txcfHh6ysrBgJpJeXF33yySc0f/58WrVqFdPtzJkzNHbsWEpOTmYYz3v27CG5XE5Dhw412TDmQl2RSESjR4+mf/7zn2xuO3uxHFWTQqEgkUhEbm5ujFCxu0sulzNSUo6hRCKR0Pr16/+jFERwcDAdOHCAURBNnTqVVq9eTYMHDyZvb29KT08ng8HAxm327NnU3NxMra2tVFtbS7W1tXT06FGm5/Lly2nChAk0e/Zsys3NJZVKRWFhYT16fBwQ/p07d0ipVNLdu3epo6Ojx73LPZtLXXCsE1zEpdVqKTc3l4YMGUJSqZSmTZtGycnJPa5TbkyEQiFFR0czD/iNN95gOL5yuZwmTpxI48ePp8DAQJP1sWvXLpLJZKRQKMjOzo5kMhnZ2NhQQkICu296ejqbq+70UCgUVFhYyPCYRSIRxcXF0fDhwxmDS2xsLJ09e5aampqooaGBsrKyCHhOfJCVlUWpqalkMBho8uTJJsDwAoGAYaWLRCKKjo7+7xrgTZs2UWpqKmMwaGpqoubmZmppaaGmpiZqbW1li2f27NmkVqvJx8eH/P39WX5k3rx5tGfPHrK0tCSFQkHx8fHk4+PD3Hgu//dHC1sikVBOTg6NGTOGZDIZ+fr6UlpaGmk0GjIajaRWq2n+/PmUlJREYrHY5MTjyA5lMhlFR0cz7iZbW1vavXs3ZWRkmHhA3enh6urKgOMBMLZnqVRK+fn5jNfsgw8+oJCQEIqJiSEfHx/i8/mk1WrJ2dmZRo8ezcbx9OnTJJVK2QnKUc9wRt1cTosjVASeA9FHRETQl19+SQ4ODgSAzp8/T21tbWyeJk6cSAKBgCIjIykuLo4GDRpEOTk5DMSae25KSgplZGSQUqk0YRwwRwXE6ckZs85EhgBo+vTpNGTIEMrKymLzDaCLx3fgwAHSarUUHh5OTk5ONGPGDPLx8WFhN5/Pp3HjxpmdF5VKh9tY6wAACUJJREFURUlJScygcZQ5AOjq1at08OBB8vHxoX379tHo0aMpKCioS96Om1O5XM6A/i0sLCgjI4MdTp3fzdw65Wh4RCIRM/5yuZzefPNN0ul0jFTAw8ODkWDyeDyaPn06rVmzhqXYONYGmUxGsbGxtGTJEtq1a5eJzua4CzknYvTo0ey9eDwe3b17l+1b7mpubqb29naaPn06CYVCdkD179+flEoleXp6kqurK9na2tKrr77KDnHOcx07duwf7luOydjBwYG8vLyYExAXF0fp6emMuVooFNLAgQNp//79zK5s2rSJPDw8aPz48RQZGUlr1qyhdevWkb+/P4lEIuYBc3PfnR7cAbJu3ToSiUSUkpJCPj4+bKw5EPz4+HjKyMigoKAgdohrtVq2Vjo7Gmq1mnFJcms0Pj6e7eP/igFWqVQkFApp79691NjYSKtXryalUklGo5EpmZiYSE5OTjRu3Djatm0bo8jR6/WMuM/X15fS09NNFkPnF+FogaytrbtVnDtpuImxtLSkwMBAEovFdPDgQVaUi4+Pp8zMTDp06BDLd/7eu+lMsaTRaOi9995jf/fx8SEej2c218idcFzxipsg7s8ODg6Me2rRokW0a9cuiomJofDwcBo1ahS5urpSXFwcffDBByQWiykhIYFKS0spPT2dsc5KJBL68ccfWXrE3ML+17/+xSjmjUYjcYUGe3t7WrhwIc2fP59tpK+++ooxUS9YsICys7OZZ6NUKsnb25s0Gg1NnTqVvQt37+XLl5tlPNBqtSQSiUzm9dVXX2Wpmb1791JISAipVCrG2QeAFixY0MUbO3PmDIWFhdH69espPz+f3NzcqK6ujsaPH29y6JkbDy4nyF16vZ6EQiFZWlqyzRUVFUVff/01o8bh9PHx8aH4+HiWy3ZxcaE5c+aQSCQiT09PmjZtGlVUVNDQoUPZuJjTIzIykng8Hvn7+zOd3377bbKzsyMAjNGFKxpKJBLS6/UsdOU43Li0CDdPnbnyOo+5OT1EIlGXyCQqKorCw8NNKNpdXV1py5YtNGnSJIqOjqa0tDTy8PAgZ2dnmjBhAk2aNMkkmkhPTyedTkeTJ0+m9957j0UpPRngiRMn0vnz58nS0pImTJhgQik/Z84cRuo6c+ZMSklJIYPBQImJiWQ0GsnCwoI5Fpyx43Th1nhraysrBpvTw8XFhdauXcvePSUlheRyOVlbW9OoUaPYgfXRRx9RQkICI7NNT09nbNRBQUFdxlQul1NGRgY7xH19fTnKov+OB7x//35atWoViUQiGjZsGEVGRpJYLKYjR46YeDXcKcMVeLgQTigUklqtpvj4ePrwww9JpVJRSkpKt6GfOW4pgUBAmzdvJj6fTx4eHmyDczpx/Fdjx44lT09PsrW1Ja1WyxY513Fhb2/PuKBSU1Npy5Yt9Pbbb7MwqLW1tUdqE7VazfifuGp7VFQUSaVS4vF4NGzYMEpJSaHs7Gx2H+4KCgqimJgYEovFxKVUiouLWbidmJjIPisWi+nXX3/t8US/cuUKSSQSlhbSaDQEgBUU3dzcSKlUklqtNmH4FYvF5O/vT6tXr6ZVq1aRVColg8FAwcHBZG9vTzt37iSxWExTpkyhyMhIRqXenR42NjbU0dFBy5cvJxcXFxo5ciSdOXOG7t69S9HR0cyT554ZERFBgYGBNHz4cPr+++9NjPDZs2cZeWl+fj5LR3BcX2+++WaPG4zbGFwe9vc53IEDB1JcXBzZ2trS6NGjacCAASSRSMjW1tYkJcHlKTMzM+mNN94gS0tLys3NZWucC03/SI/OB/7SpUspKyuLAgMDKTU1ldLT01kVvfNlaWlJo0ePZgU/Z2dn1pWTmJhIAQEBJJfLqa2tjbKyskilUpnVo729ndLT003GYdmyZTR58mSTdCDnRHDUY9xlbW1N8fHxNHr0aGZoOneKcAY+MjLSLFUUlxYLCwsjHo9H7777ronxnz59OoWEhNCePXsoOTmZZs+eTbNmzaIxY8bQpk2b6JNPPqGAgACmm7+/P/3000+kUCiYXmKxmDFmm5sXPz8/dmA1NDQwGq/fz5VarWYEoJztkEqlpFKpSCKRkJeXF6Wnp5NMJiMPDw+SSCQ0aNAg2rNnD40YMYLGjRtHcrmcNBrNf88AA6C4uDhSKBRsIXPVTK7CzX0uKSmJ9u/fT0VFRbR48WLy8vIyMUQSiYQ8PDxYoUUgEFBERITJIHSnh0KhoG+//ZYN4pdffkmWlpYUGhpKRUVFtHXrVvL19WUhjF6vp7Fjx1JKSgrLayYnJ9OGDRvIzc2NhQ5cblggEFBgYCABz3nFzC0onU5HHh4epNfrKTk5mQYMGMDo4Hfv3s0M0cKFCykzM5O0Wi0NHz6cEhISKCwsjBnZ8PBwWrFiBXtnpVJJaWlpBDzPLbe1tf0hx1Xna968eYw7LSQkhORyOU2ePJkmTZrEwjvO08vOzqZx48ZRUlISff7559S/f386cOAAOTk5kaurK5WWlnYxDOY8Pq5TgTsMAgMDSSgUstSMSqWi+fPnU0BAAAGggIAASk5OJm9vb5YSmjdvHm3ZsoVEIlEXwyCTyZgX2tP64PP5dOjQIWZstm3bRp0r8NwllUpJKpWSl5cX86r8/f1pwIAB7DMKhYKioqLoH//4BzNE+/fvp4EDBxKPx6OMjIwe5+XYsWOUk5PDwtv9+/dTWlqaCVv25s2bTQ7c7i6udpCVlUU5OTmM5XjKlCnEdTn0tE6B59yFSUlJlJCQ0IXxFwBFR0fTwIEDTcbdYDAwyni9Xs9SMp2/x6Wz/mheAFBGRgaLHDrfY8CAAd3O0eTJk1m7n9FopJkzZ9KCBQsoPT2dHBwc2D4VCoUUGhpKa9eupRkzZjD7ZE6PQ4cO0YIFC0xo5w8cOEBz5sxhzktqairFxsaSSqWi2NhYlhpNT0+nTZs2Mft37do1E6/89weum5ubWQPM+7dC/5HweLwKAA//4y/8v4srEXX5uVOvHr169OrRq8f/73oA+HMGuFd6pVd6pVf+e/Knf4rcK73SK73SK/8d6TXAvdIrvdIrf5H0GuBe6ZVe6ZW/SHoNcK/0Sq/0yl8kvQa4V3qlV3rlL5JeA9wrvdIrvfIXSa8B7pVe6ZVe+Yuk1wD3Sq/0Sq/8RdJrgHulV3qlV/4i+V8z0wlNdbpbzQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EvH0Sl6mRjxh",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "0e204558-29a3-42b8-9500-7b0aeefc2723"
      },
      "source": [
        "main(noise_factor_train = 0.7, noise_factor_test = 0.7)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_11\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_54 (Conv2D)           (None, 28, 28, 16)        160       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_33 (MaxPooling (None, 14, 14, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_55 (Conv2D)           (None, 14, 14, 8)         1160      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_34 (MaxPooling (None, 7, 7, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_56 (Conv2D)           (None, 7, 7, 8)           584       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_35 (MaxPooling (None, 4, 4, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_57 (Conv2D)           (None, 4, 4, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_15 (UpSampling (None, 8, 8, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_58 (Conv2D)           (None, 8, 8, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_16 (UpSampling (None, 16, 16, 8)         0         \n",
            "_________________________________________________________________\n",
            "conv2d_59 (Conv2D)           (None, 16, 16, 16)        1168      \n",
            "_________________________________________________________________\n",
            "up_sampling2d_17 (UpSampling (None, 32, 32, 16)        0         \n",
            "_________________________________________________________________\n",
            "flatten_5 (Flatten)          (None, 16384)             0         \n",
            "_________________________________________________________________\n",
            "dense_10 (Dense)             (None, 64)                1048640   \n",
            "_________________________________________________________________\n",
            "dense_11 (Dense)             (None, 10)                650       \n",
            "=================================================================\n",
            "Total params: 1,053,530\n",
            "Trainable params: 1,053,530\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 1.5990 - accuracy: 0.4747 - val_loss: 0.8791 - val_accuracy: 0.7194\n",
            "Epoch 2/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.6501 - accuracy: 0.7941 - val_loss: 0.5014 - val_accuracy: 0.8382\n",
            "Epoch 3/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.4077 - accuracy: 0.8725 - val_loss: 0.3594 - val_accuracy: 0.8847\n",
            "Epoch 4/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.3182 - accuracy: 0.9000 - val_loss: 0.2747 - val_accuracy: 0.9129\n",
            "Epoch 5/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.2614 - accuracy: 0.9189 - val_loss: 0.2456 - val_accuracy: 0.9223\n",
            "Epoch 6/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.2286 - accuracy: 0.9278 - val_loss: 0.2061 - val_accuracy: 0.9349\n",
            "Epoch 7/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.2050 - accuracy: 0.9363 - val_loss: 0.2025 - val_accuracy: 0.9356\n",
            "Epoch 8/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1861 - accuracy: 0.9417 - val_loss: 0.1719 - val_accuracy: 0.9456\n",
            "Epoch 9/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1705 - accuracy: 0.9469 - val_loss: 0.1607 - val_accuracy: 0.9498\n",
            "Epoch 10/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1576 - accuracy: 0.9505 - val_loss: 0.1525 - val_accuracy: 0.9511\n",
            "Epoch 11/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1485 - accuracy: 0.9531 - val_loss: 0.1444 - val_accuracy: 0.9534\n",
            "Epoch 12/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1407 - accuracy: 0.9549 - val_loss: 0.1386 - val_accuracy: 0.9557\n",
            "Epoch 13/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1309 - accuracy: 0.9580 - val_loss: 0.1190 - val_accuracy: 0.9615\n",
            "Epoch 14/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1233 - accuracy: 0.9607 - val_loss: 0.1275 - val_accuracy: 0.9576\n",
            "Epoch 15/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1170 - accuracy: 0.9624 - val_loss: 0.1127 - val_accuracy: 0.9639\n",
            "Epoch 16/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1132 - accuracy: 0.9635 - val_loss: 0.1080 - val_accuracy: 0.9655\n",
            "Epoch 17/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1102 - accuracy: 0.9644 - val_loss: 0.1025 - val_accuracy: 0.9667\n",
            "Epoch 18/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.1061 - accuracy: 0.9657 - val_loss: 0.0956 - val_accuracy: 0.9697\n",
            "Epoch 19/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0954 - accuracy: 0.9694 - val_loss: 0.0936 - val_accuracy: 0.9698\n",
            "Epoch 20/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0922 - accuracy: 0.9701 - val_loss: 0.0893 - val_accuracy: 0.9708\n",
            "Epoch 21/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0886 - accuracy: 0.9709 - val_loss: 0.0831 - val_accuracy: 0.9732\n",
            "Epoch 22/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0863 - accuracy: 0.9721 - val_loss: 0.0864 - val_accuracy: 0.9712\n",
            "Epoch 23/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0855 - accuracy: 0.9722 - val_loss: 0.0810 - val_accuracy: 0.9740\n",
            "Epoch 24/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0809 - accuracy: 0.9737 - val_loss: 0.0849 - val_accuracy: 0.9721\n",
            "Epoch 25/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0794 - accuracy: 0.9742 - val_loss: 0.0734 - val_accuracy: 0.9764\n",
            "Epoch 26/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0738 - accuracy: 0.9761 - val_loss: 0.0751 - val_accuracy: 0.9758\n",
            "Epoch 27/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0727 - accuracy: 0.9765 - val_loss: 0.0652 - val_accuracy: 0.9795\n",
            "Epoch 28/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0679 - accuracy: 0.9780 - val_loss: 0.0706 - val_accuracy: 0.9762\n",
            "Epoch 29/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0693 - accuracy: 0.9772 - val_loss: 0.0649 - val_accuracy: 0.9788\n",
            "Epoch 30/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0637 - accuracy: 0.9790 - val_loss: 0.0596 - val_accuracy: 0.9806\n",
            "Epoch 31/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0625 - accuracy: 0.9795 - val_loss: 0.0565 - val_accuracy: 0.9820\n",
            "Epoch 32/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0583 - accuracy: 0.9812 - val_loss: 0.0583 - val_accuracy: 0.9805\n",
            "Epoch 33/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0618 - accuracy: 0.9794 - val_loss: 0.0586 - val_accuracy: 0.9806\n",
            "Epoch 34/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0571 - accuracy: 0.9814 - val_loss: 0.0512 - val_accuracy: 0.9837\n",
            "Epoch 35/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0554 - accuracy: 0.9816 - val_loss: 0.0525 - val_accuracy: 0.9827\n",
            "Epoch 36/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0522 - accuracy: 0.9828 - val_loss: 0.0513 - val_accuracy: 0.9831\n",
            "Epoch 37/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0530 - accuracy: 0.9827 - val_loss: 0.0565 - val_accuracy: 0.9807\n",
            "Epoch 38/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0506 - accuracy: 0.9833 - val_loss: 0.0455 - val_accuracy: 0.9855\n",
            "Epoch 39/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0460 - accuracy: 0.9848 - val_loss: 0.0418 - val_accuracy: 0.9869\n",
            "Epoch 40/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0443 - accuracy: 0.9856 - val_loss: 0.0400 - val_accuracy: 0.9875\n",
            "Epoch 41/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0455 - accuracy: 0.9848 - val_loss: 0.0407 - val_accuracy: 0.9864\n",
            "Epoch 42/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0450 - accuracy: 0.9848 - val_loss: 0.0487 - val_accuracy: 0.9832\n",
            "Epoch 43/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0444 - accuracy: 0.9852 - val_loss: 0.0380 - val_accuracy: 0.9877\n",
            "Epoch 44/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0408 - accuracy: 0.9865 - val_loss: 0.0347 - val_accuracy: 0.9889\n",
            "Epoch 45/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0379 - accuracy: 0.9876 - val_loss: 0.0366 - val_accuracy: 0.9883\n",
            "Epoch 46/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0401 - accuracy: 0.9870 - val_loss: 0.0383 - val_accuracy: 0.9872\n",
            "Epoch 47/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0362 - accuracy: 0.9881 - val_loss: 0.0329 - val_accuracy: 0.9894\n",
            "Epoch 48/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0351 - accuracy: 0.9884 - val_loss: 0.0325 - val_accuracy: 0.9897\n",
            "Epoch 49/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0335 - accuracy: 0.9890 - val_loss: 0.0313 - val_accuracy: 0.9898\n",
            "Epoch 50/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 0.0313 - accuracy: 0.9899 - val_loss: 0.0345 - val_accuracy: 0.9883\n",
            "313/313 [==============================] - 4s 14ms/step - loss: 0.0995 - accuracy: 0.9721\n",
            "Test loss : 0.09951125830411911\n",
            "Test accuracy: 97.2100019454956\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aVlj1BQkRl2R",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "9b8591b8-601a-4eb4-b3c7-3a175a2bafe7"
      },
      "source": [
        "main(noise_factor_train = 0.9, noise_factor_test = 0.9)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_12\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_60 (Conv2D)           (None, 28, 28, 16)        160       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_36 (MaxPooling (None, 14, 14, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_61 (Conv2D)           (None, 14, 14, 8)         1160      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_37 (MaxPooling (None, 7, 7, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_62 (Conv2D)           (None, 7, 7, 8)           584       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_38 (MaxPooling (None, 4, 4, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_63 (Conv2D)           (None, 4, 4, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_18 (UpSampling (None, 8, 8, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_64 (Conv2D)           (None, 8, 8, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_19 (UpSampling (None, 16, 16, 8)         0         \n",
            "_________________________________________________________________\n",
            "conv2d_65 (Conv2D)           (None, 16, 16, 16)        1168      \n",
            "_________________________________________________________________\n",
            "up_sampling2d_20 (UpSampling (None, 32, 32, 16)        0         \n",
            "_________________________________________________________________\n",
            "flatten_6 (Flatten)          (None, 16384)             0         \n",
            "_________________________________________________________________\n",
            "dense_12 (Dense)             (None, 64)                1048640   \n",
            "_________________________________________________________________\n",
            "dense_13 (Dense)             (None, 10)                650       \n",
            "=================================================================\n",
            "Total params: 1,053,530\n",
            "Trainable params: 1,053,530\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 1.8983 - accuracy: 0.3277 - val_loss: 1.1886 - val_accuracy: 0.5989\n",
            "Epoch 2/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.8469 - accuracy: 0.7173 - val_loss: 0.5993 - val_accuracy: 0.8044\n",
            "Epoch 3/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.5226 - accuracy: 0.8290 - val_loss: 0.4463 - val_accuracy: 0.8566\n",
            "Epoch 4/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.4150 - accuracy: 0.8648 - val_loss: 0.3766 - val_accuracy: 0.8776\n",
            "Epoch 5/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.3448 - accuracy: 0.8884 - val_loss: 0.3226 - val_accuracy: 0.8965\n",
            "Epoch 6/50\n",
            "30/30 [==============================] - 70s 2s/step - loss: 0.2958 - accuracy: 0.9048 - val_loss: 0.2749 - val_accuracy: 0.9120\n",
            "Epoch 7/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.2717 - accuracy: 0.9130 - val_loss: 0.2782 - val_accuracy: 0.9091\n",
            "Epoch 8/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.2443 - accuracy: 0.9225 - val_loss: 0.2228 - val_accuracy: 0.9284\n",
            "Epoch 9/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.2193 - accuracy: 0.9305 - val_loss: 0.2091 - val_accuracy: 0.9340\n",
            "Epoch 10/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.2017 - accuracy: 0.9348 - val_loss: 0.1930 - val_accuracy: 0.9386\n",
            "Epoch 11/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1918 - accuracy: 0.9380 - val_loss: 0.1841 - val_accuracy: 0.9415\n",
            "Epoch 12/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1797 - accuracy: 0.9422 - val_loss: 0.1854 - val_accuracy: 0.9394\n",
            "Epoch 13/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1684 - accuracy: 0.9460 - val_loss: 0.1654 - val_accuracy: 0.9463\n",
            "Epoch 14/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1598 - accuracy: 0.9489 - val_loss: 0.1586 - val_accuracy: 0.9497\n",
            "Epoch 15/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1532 - accuracy: 0.9514 - val_loss: 0.1470 - val_accuracy: 0.9531\n",
            "Epoch 16/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1451 - accuracy: 0.9539 - val_loss: 0.1439 - val_accuracy: 0.9535\n",
            "Epoch 17/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1390 - accuracy: 0.9560 - val_loss: 0.1354 - val_accuracy: 0.9566\n",
            "Epoch 18/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1345 - accuracy: 0.9570 - val_loss: 0.1282 - val_accuracy: 0.9594\n",
            "Epoch 19/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1302 - accuracy: 0.9584 - val_loss: 0.1280 - val_accuracy: 0.9587\n",
            "Epoch 20/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1250 - accuracy: 0.9603 - val_loss: 0.1183 - val_accuracy: 0.9625\n",
            "Epoch 21/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1200 - accuracy: 0.9618 - val_loss: 0.1176 - val_accuracy: 0.9627\n",
            "Epoch 22/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1172 - accuracy: 0.9623 - val_loss: 0.1107 - val_accuracy: 0.9646\n",
            "Epoch 23/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1145 - accuracy: 0.9642 - val_loss: 0.1052 - val_accuracy: 0.9667\n",
            "Epoch 24/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1085 - accuracy: 0.9649 - val_loss: 0.1030 - val_accuracy: 0.9667\n",
            "Epoch 25/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1059 - accuracy: 0.9662 - val_loss: 0.1014 - val_accuracy: 0.9678\n",
            "Epoch 26/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1026 - accuracy: 0.9667 - val_loss: 0.1017 - val_accuracy: 0.9673\n",
            "Epoch 27/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1042 - accuracy: 0.9660 - val_loss: 0.1069 - val_accuracy: 0.9650\n",
            "Epoch 28/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.1017 - accuracy: 0.9675 - val_loss: 0.1022 - val_accuracy: 0.9669\n",
            "Epoch 29/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0985 - accuracy: 0.9679 - val_loss: 0.0998 - val_accuracy: 0.9670\n",
            "Epoch 30/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0935 - accuracy: 0.9700 - val_loss: 0.0945 - val_accuracy: 0.9694\n",
            "Epoch 31/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0886 - accuracy: 0.9716 - val_loss: 0.0823 - val_accuracy: 0.9738\n",
            "Epoch 32/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0852 - accuracy: 0.9728 - val_loss: 0.0832 - val_accuracy: 0.9737\n",
            "Epoch 33/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0840 - accuracy: 0.9726 - val_loss: 0.0868 - val_accuracy: 0.9718\n",
            "Epoch 34/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0819 - accuracy: 0.9735 - val_loss: 0.0767 - val_accuracy: 0.9761\n",
            "Epoch 35/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0798 - accuracy: 0.9743 - val_loss: 0.0758 - val_accuracy: 0.9753\n",
            "Epoch 36/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0789 - accuracy: 0.9748 - val_loss: 0.0732 - val_accuracy: 0.9768\n",
            "Epoch 37/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0776 - accuracy: 0.9750 - val_loss: 0.0866 - val_accuracy: 0.9719\n",
            "Epoch 38/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0764 - accuracy: 0.9748 - val_loss: 0.0765 - val_accuracy: 0.9746\n",
            "Epoch 39/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0761 - accuracy: 0.9753 - val_loss: 0.0713 - val_accuracy: 0.9768\n",
            "Epoch 40/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0704 - accuracy: 0.9770 - val_loss: 0.0686 - val_accuracy: 0.9780\n",
            "Epoch 41/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0696 - accuracy: 0.9773 - val_loss: 0.0648 - val_accuracy: 0.9795\n",
            "Epoch 42/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0675 - accuracy: 0.9783 - val_loss: 0.0651 - val_accuracy: 0.9787\n",
            "Epoch 43/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0651 - accuracy: 0.9790 - val_loss: 0.0615 - val_accuracy: 0.9803\n",
            "Epoch 44/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0664 - accuracy: 0.9782 - val_loss: 0.0640 - val_accuracy: 0.9793\n",
            "Epoch 45/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0628 - accuracy: 0.9796 - val_loss: 0.0632 - val_accuracy: 0.9798\n",
            "Epoch 46/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0612 - accuracy: 0.9804 - val_loss: 0.0550 - val_accuracy: 0.9826\n",
            "Epoch 47/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0597 - accuracy: 0.9807 - val_loss: 0.0594 - val_accuracy: 0.9811\n",
            "Epoch 48/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0610 - accuracy: 0.9797 - val_loss: 0.0515 - val_accuracy: 0.9836\n",
            "Epoch 49/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0578 - accuracy: 0.9812 - val_loss: 0.0567 - val_accuracy: 0.9815\n",
            "Epoch 50/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0554 - accuracy: 0.9816 - val_loss: 0.0528 - val_accuracy: 0.9827\n",
            "313/313 [==============================] - 5s 15ms/step - loss: 0.0953 - accuracy: 0.9704\n",
            "Test loss : 0.09526178240776062\n",
            "Test accuracy: 97.03999757766724\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K3imxZ7xRny3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "b1998fcc-a7ed-4a68-ec2a-ad7e743371b1"
      },
      "source": [
        "main(noise_factor_train = 0.4, noise_factor_test = 0.7)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_13\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_66 (Conv2D)           (None, 28, 28, 16)        160       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_39 (MaxPooling (None, 14, 14, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_67 (Conv2D)           (None, 14, 14, 8)         1160      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_40 (MaxPooling (None, 7, 7, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_68 (Conv2D)           (None, 7, 7, 8)           584       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_41 (MaxPooling (None, 4, 4, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_69 (Conv2D)           (None, 4, 4, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_21 (UpSampling (None, 8, 8, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_70 (Conv2D)           (None, 8, 8, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_22 (UpSampling (None, 16, 16, 8)         0         \n",
            "_________________________________________________________________\n",
            "conv2d_71 (Conv2D)           (None, 16, 16, 16)        1168      \n",
            "_________________________________________________________________\n",
            "up_sampling2d_23 (UpSampling (None, 32, 32, 16)        0         \n",
            "_________________________________________________________________\n",
            "flatten_7 (Flatten)          (None, 16384)             0         \n",
            "_________________________________________________________________\n",
            "dense_14 (Dense)             (None, 64)                1048640   \n",
            "_________________________________________________________________\n",
            "dense_15 (Dense)             (None, 10)                650       \n",
            "=================================================================\n",
            "Total params: 1,053,530\n",
            "Trainable params: 1,053,530\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/50\n",
            "30/30 [==============================] - 65s 2s/step - loss: 1.5689 - accuracy: 0.4914 - val_loss: 0.8076 - val_accuracy: 0.7294\n",
            "Epoch 2/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.6368 - accuracy: 0.7857 - val_loss: 0.5000 - val_accuracy: 0.8310\n",
            "Epoch 3/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.4293 - accuracy: 0.8584 - val_loss: 0.3696 - val_accuracy: 0.8802\n",
            "Epoch 4/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.3273 - accuracy: 0.8943 - val_loss: 0.3159 - val_accuracy: 0.8985\n",
            "Epoch 5/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2761 - accuracy: 0.9120 - val_loss: 0.2560 - val_accuracy: 0.9186\n",
            "Epoch 6/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2339 - accuracy: 0.9265 - val_loss: 0.2267 - val_accuracy: 0.9281\n",
            "Epoch 7/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2104 - accuracy: 0.9330 - val_loss: 0.2048 - val_accuracy: 0.9365\n",
            "Epoch 8/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1946 - accuracy: 0.9390 - val_loss: 0.1994 - val_accuracy: 0.9366\n",
            "Epoch 9/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1842 - accuracy: 0.9422 - val_loss: 0.1634 - val_accuracy: 0.9485\n",
            "Epoch 10/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1600 - accuracy: 0.9499 - val_loss: 0.1516 - val_accuracy: 0.9527\n",
            "Epoch 11/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1531 - accuracy: 0.9519 - val_loss: 0.1529 - val_accuracy: 0.9519\n",
            "Epoch 12/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1440 - accuracy: 0.9547 - val_loss: 0.1401 - val_accuracy: 0.9563\n",
            "Epoch 13/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1343 - accuracy: 0.9578 - val_loss: 0.1249 - val_accuracy: 0.9606\n",
            "Epoch 14/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1254 - accuracy: 0.9606 - val_loss: 0.1254 - val_accuracy: 0.9607\n",
            "Epoch 15/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.1241 - accuracy: 0.9612 - val_loss: 0.1183 - val_accuracy: 0.9631\n",
            "Epoch 16/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1131 - accuracy: 0.9645 - val_loss: 0.1200 - val_accuracy: 0.9612\n",
            "Epoch 17/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1151 - accuracy: 0.9633 - val_loss: 0.1032 - val_accuracy: 0.9677\n",
            "Epoch 18/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1041 - accuracy: 0.9669 - val_loss: 0.0980 - val_accuracy: 0.9691\n",
            "Epoch 19/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1024 - accuracy: 0.9673 - val_loss: 0.0933 - val_accuracy: 0.9706\n",
            "Epoch 20/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0956 - accuracy: 0.9703 - val_loss: 0.0878 - val_accuracy: 0.9729\n",
            "Epoch 21/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0938 - accuracy: 0.9707 - val_loss: 0.0898 - val_accuracy: 0.9716\n",
            "Epoch 22/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0899 - accuracy: 0.9715 - val_loss: 0.0818 - val_accuracy: 0.9743\n",
            "Epoch 23/50\n",
            "30/30 [==============================] - 67s 2s/step - loss: 0.0851 - accuracy: 0.9732 - val_loss: 0.0791 - val_accuracy: 0.9758\n",
            "Epoch 24/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0820 - accuracy: 0.9742 - val_loss: 0.0843 - val_accuracy: 0.9730\n",
            "Epoch 25/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0812 - accuracy: 0.9748 - val_loss: 0.0815 - val_accuracy: 0.9736\n",
            "Epoch 26/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0769 - accuracy: 0.9757 - val_loss: 0.0765 - val_accuracy: 0.9753\n",
            "Epoch 27/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0745 - accuracy: 0.9763 - val_loss: 0.0688 - val_accuracy: 0.9780\n",
            "Epoch 28/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0720 - accuracy: 0.9773 - val_loss: 0.0665 - val_accuracy: 0.9799\n",
            "Epoch 29/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0715 - accuracy: 0.9772 - val_loss: 0.0720 - val_accuracy: 0.9767\n",
            "Epoch 30/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0686 - accuracy: 0.9785 - val_loss: 0.0645 - val_accuracy: 0.9798\n",
            "Epoch 31/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0683 - accuracy: 0.9776 - val_loss: 0.0659 - val_accuracy: 0.9791\n",
            "Epoch 32/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0679 - accuracy: 0.9780 - val_loss: 0.0682 - val_accuracy: 0.9775\n",
            "Epoch 33/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0653 - accuracy: 0.9791 - val_loss: 0.0599 - val_accuracy: 0.9807\n",
            "Epoch 34/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0605 - accuracy: 0.9807 - val_loss: 0.0599 - val_accuracy: 0.9808\n",
            "Epoch 35/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0593 - accuracy: 0.9807 - val_loss: 0.0578 - val_accuracy: 0.9811\n",
            "Epoch 36/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0575 - accuracy: 0.9816 - val_loss: 0.0539 - val_accuracy: 0.9830\n",
            "Epoch 37/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0543 - accuracy: 0.9829 - val_loss: 0.0521 - val_accuracy: 0.9838\n",
            "Epoch 38/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0524 - accuracy: 0.9833 - val_loss: 0.0524 - val_accuracy: 0.9829\n",
            "Epoch 39/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0530 - accuracy: 0.9828 - val_loss: 0.0489 - val_accuracy: 0.9844\n",
            "Epoch 40/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0499 - accuracy: 0.9840 - val_loss: 0.0461 - val_accuracy: 0.9844\n",
            "Epoch 41/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0498 - accuracy: 0.9835 - val_loss: 0.0447 - val_accuracy: 0.9861\n",
            "Epoch 42/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0483 - accuracy: 0.9841 - val_loss: 0.0422 - val_accuracy: 0.9869\n",
            "Epoch 43/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0476 - accuracy: 0.9845 - val_loss: 0.0589 - val_accuracy: 0.9800\n",
            "Epoch 44/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0491 - accuracy: 0.9839 - val_loss: 0.0423 - val_accuracy: 0.9865\n",
            "Epoch 45/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0437 - accuracy: 0.9856 - val_loss: 0.0411 - val_accuracy: 0.9868\n",
            "Epoch 46/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0433 - accuracy: 0.9863 - val_loss: 0.0375 - val_accuracy: 0.9880\n",
            "Epoch 47/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0397 - accuracy: 0.9874 - val_loss: 0.0371 - val_accuracy: 0.9885\n",
            "Epoch 48/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0390 - accuracy: 0.9876 - val_loss: 0.0364 - val_accuracy: 0.9882\n",
            "Epoch 49/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0382 - accuracy: 0.9877 - val_loss: 0.0352 - val_accuracy: 0.9887\n",
            "Epoch 50/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0381 - accuracy: 0.9879 - val_loss: 0.0392 - val_accuracy: 0.9869\n",
            "313/313 [==============================] - 5s 14ms/step - loss: 0.1558 - accuracy: 0.9543\n",
            "Test loss : 0.15583421289920807\n",
            "Test accuracy: 95.42999863624573\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "92PJFwXvRqja",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "a86fa042-afe4-4260-f919-e5d1b24eedf4"
      },
      "source": [
        "main(noise_factor_train = 0.7, noise_factor_test = 0.4)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_14\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_72 (Conv2D)           (None, 28, 28, 16)        160       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_42 (MaxPooling (None, 14, 14, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_73 (Conv2D)           (None, 14, 14, 8)         1160      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_43 (MaxPooling (None, 7, 7, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_74 (Conv2D)           (None, 7, 7, 8)           584       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_44 (MaxPooling (None, 4, 4, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_75 (Conv2D)           (None, 4, 4, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_24 (UpSampling (None, 8, 8, 8)           0         \n",
            "_________________________________________________________________\n",
            "conv2d_76 (Conv2D)           (None, 8, 8, 8)           584       \n",
            "_________________________________________________________________\n",
            "up_sampling2d_25 (UpSampling (None, 16, 16, 8)         0         \n",
            "_________________________________________________________________\n",
            "conv2d_77 (Conv2D)           (None, 16, 16, 16)        1168      \n",
            "_________________________________________________________________\n",
            "up_sampling2d_26 (UpSampling (None, 32, 32, 16)        0         \n",
            "_________________________________________________________________\n",
            "flatten_8 (Flatten)          (None, 16384)             0         \n",
            "_________________________________________________________________\n",
            "dense_16 (Dense)             (None, 64)                1048640   \n",
            "_________________________________________________________________\n",
            "dense_17 (Dense)             (None, 10)                650       \n",
            "=================================================================\n",
            "Total params: 1,053,530\n",
            "Trainable params: 1,053,530\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 1.5683 - accuracy: 0.4887 - val_loss: 0.8611 - val_accuracy: 0.7130\n",
            "Epoch 2/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.6383 - accuracy: 0.7933 - val_loss: 0.4917 - val_accuracy: 0.8419\n",
            "Epoch 3/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.4258 - accuracy: 0.8634 - val_loss: 0.4111 - val_accuracy: 0.8645\n",
            "Epoch 4/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.3398 - accuracy: 0.8918 - val_loss: 0.3024 - val_accuracy: 0.9046\n",
            "Epoch 5/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2825 - accuracy: 0.9105 - val_loss: 0.2558 - val_accuracy: 0.9196\n",
            "Epoch 6/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2469 - accuracy: 0.9216 - val_loss: 0.2306 - val_accuracy: 0.9261\n",
            "Epoch 7/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2224 - accuracy: 0.9295 - val_loss: 0.2165 - val_accuracy: 0.9309\n",
            "Epoch 8/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.2025 - accuracy: 0.9355 - val_loss: 0.1983 - val_accuracy: 0.9366\n",
            "Epoch 9/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1920 - accuracy: 0.9395 - val_loss: 0.1993 - val_accuracy: 0.9361\n",
            "Epoch 10/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1824 - accuracy: 0.9416 - val_loss: 0.1743 - val_accuracy: 0.9448\n",
            "Epoch 11/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1693 - accuracy: 0.9461 - val_loss: 0.1683 - val_accuracy: 0.9459\n",
            "Epoch 12/50\n",
            "30/30 [==============================] - 67s 2s/step - loss: 0.1592 - accuracy: 0.9493 - val_loss: 0.1481 - val_accuracy: 0.9532\n",
            "Epoch 13/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1489 - accuracy: 0.9531 - val_loss: 0.1458 - val_accuracy: 0.9539\n",
            "Epoch 14/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1449 - accuracy: 0.9538 - val_loss: 0.1360 - val_accuracy: 0.9566\n",
            "Epoch 15/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1399 - accuracy: 0.9549 - val_loss: 0.1340 - val_accuracy: 0.9581\n",
            "Epoch 16/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1317 - accuracy: 0.9582 - val_loss: 0.1191 - val_accuracy: 0.9625\n",
            "Epoch 17/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1234 - accuracy: 0.9599 - val_loss: 0.1210 - val_accuracy: 0.9608\n",
            "Epoch 18/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1187 - accuracy: 0.9621 - val_loss: 0.1119 - val_accuracy: 0.9650\n",
            "Epoch 19/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1130 - accuracy: 0.9639 - val_loss: 0.1075 - val_accuracy: 0.9654\n",
            "Epoch 20/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1123 - accuracy: 0.9644 - val_loss: 0.1094 - val_accuracy: 0.9648\n",
            "Epoch 21/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1113 - accuracy: 0.9645 - val_loss: 0.1107 - val_accuracy: 0.9644\n",
            "Epoch 22/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.1041 - accuracy: 0.9671 - val_loss: 0.0964 - val_accuracy: 0.9697\n",
            "Epoch 23/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0982 - accuracy: 0.9687 - val_loss: 0.0963 - val_accuracy: 0.9689\n",
            "Epoch 24/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0962 - accuracy: 0.9697 - val_loss: 0.0892 - val_accuracy: 0.9715\n",
            "Epoch 25/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0948 - accuracy: 0.9700 - val_loss: 0.0905 - val_accuracy: 0.9710\n",
            "Epoch 26/50\n",
            "30/30 [==============================] - 64s 2s/step - loss: 0.0929 - accuracy: 0.9707 - val_loss: 0.0861 - val_accuracy: 0.9727\n",
            "Epoch 27/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0922 - accuracy: 0.9703 - val_loss: 0.0879 - val_accuracy: 0.9715\n",
            "Epoch 28/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0862 - accuracy: 0.9727 - val_loss: 0.0846 - val_accuracy: 0.9732\n",
            "Epoch 29/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0847 - accuracy: 0.9729 - val_loss: 0.0797 - val_accuracy: 0.9747\n",
            "Epoch 30/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0814 - accuracy: 0.9741 - val_loss: 0.0921 - val_accuracy: 0.9697\n",
            "Epoch 31/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0830 - accuracy: 0.9730 - val_loss: 0.0747 - val_accuracy: 0.9763\n",
            "Epoch 32/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0780 - accuracy: 0.9748 - val_loss: 0.0753 - val_accuracy: 0.9756\n",
            "Epoch 33/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0744 - accuracy: 0.9762 - val_loss: 0.0698 - val_accuracy: 0.9780\n",
            "Epoch 34/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0729 - accuracy: 0.9764 - val_loss: 0.0692 - val_accuracy: 0.9776\n",
            "Epoch 35/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0714 - accuracy: 0.9772 - val_loss: 0.0642 - val_accuracy: 0.9798\n",
            "Epoch 36/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0712 - accuracy: 0.9771 - val_loss: 0.0624 - val_accuracy: 0.9801\n",
            "Epoch 37/50\n",
            "30/30 [==============================] - 66s 2s/step - loss: 0.0667 - accuracy: 0.9786 - val_loss: 0.0649 - val_accuracy: 0.9785\n",
            "Epoch 38/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0643 - accuracy: 0.9793 - val_loss: 0.0609 - val_accuracy: 0.9798\n",
            "Epoch 39/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0677 - accuracy: 0.9778 - val_loss: 0.0664 - val_accuracy: 0.9783\n",
            "Epoch 40/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0627 - accuracy: 0.9794 - val_loss: 0.0607 - val_accuracy: 0.9802\n",
            "Epoch 41/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0579 - accuracy: 0.9816 - val_loss: 0.0572 - val_accuracy: 0.9815\n",
            "Epoch 42/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0581 - accuracy: 0.9813 - val_loss: 0.0589 - val_accuracy: 0.9804\n",
            "Epoch 43/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0597 - accuracy: 0.9805 - val_loss: 0.0565 - val_accuracy: 0.9814\n",
            "Epoch 44/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0569 - accuracy: 0.9812 - val_loss: 0.0527 - val_accuracy: 0.9831\n",
            "Epoch 45/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0534 - accuracy: 0.9829 - val_loss: 0.0539 - val_accuracy: 0.9822\n",
            "Epoch 46/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0535 - accuracy: 0.9826 - val_loss: 0.0521 - val_accuracy: 0.9822\n",
            "Epoch 47/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0556 - accuracy: 0.9807 - val_loss: 0.0495 - val_accuracy: 0.9838\n",
            "Epoch 48/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0529 - accuracy: 0.9819 - val_loss: 0.0481 - val_accuracy: 0.9847\n",
            "Epoch 49/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0497 - accuracy: 0.9835 - val_loss: 0.0455 - val_accuracy: 0.9850\n",
            "Epoch 50/50\n",
            "30/30 [==============================] - 63s 2s/step - loss: 0.0477 - accuracy: 0.9839 - val_loss: 0.0568 - val_accuracy: 0.9797\n",
            "313/313 [==============================] - 5s 14ms/step - loss: 0.0908 - accuracy: 0.9744\n",
            "Test loss : 0.09084675461053848\n",
            "Test accuracy: 97.4399983882904\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXxc5X3v8c9vFs2MFkuy5H3BBswesME4UKCBAClbWNpewuI09DZx0iQtaZM0kJuV3qTpfeWmaVLShFJuyAKEkkAccMIOacJqDElYjG3AtuRV1mZptIxm5nf/OEf2WBZYYI2Edb7v12teM2eZOb8jj89vnuc5z/OYuyMiItEVG+8ARERkfCkRiIhEnBKBiEjEKRGIiEScEoGISMQpEYiIRJwSgUSKmX3fzP73CPddb2ZnlTsmkfGmRCAiEnFKBCIHIDNLjHcMMnEoEcjbTlgl82kz+72ZZc3sP81smpn90sy6zOwBM6sv2f9CM3vBzDrM7BEzO7Jk2yIzWxW+7ydAesixLjCz58L3PmZmx44wxvPN7Fkz22lmTWb2pSHbTw0/ryPcflW4PmNm/9fMNphZp5n9Jlx3upk1D/N3OCt8/SUzu8PMfmRmO4GrzGyJmT0eHmOLmf2bmVWUvP9oM7vfzNrMbJuZfdbMpptZj5k1lOx3vJm1mFlyJOcuE48Sgbxd/RlwNnAY8F7gl8BngSkE39u/BTCzw4BbgU+E21YAvzCzivCieBfwQ2Ay8F/h5xK+dxFwE/BhoAH4HrDczFIjiC8L/AVQB5wP/LWZXRx+7kFhvN8OY1oIPBe+7+vACcAfhTH9A1Ac4d/kIuCO8Jg/BgrA3wGNwMnAmcBHwxhqgAeAXwEzgUOBB919K/AIcGnJ574fuM3dB0YYh0wwSgTydvVtd9/m7puA/waedPdn3b0PuBNYFO73PuAed78/vJB9HcgQXGhPApLAN919wN3vAJ4uOcYy4Hvu/qS7F9z9ZqA/fN8bcvdH3P0P7l50998TJKN3hZuvAB5w91vD47a6+3NmFgP+J3C1u28Kj/mYu/eP8G/yuLvfFR6z192fcfcn3D3v7usJEtlgDBcAW939/7p7n7t3ufuT4babgaUAZhYHLidIlhJRSgTydrWt5HXvMMvV4euZwIbBDe5eBJqAWeG2Tb7nyIobSl4fBHwyrFrpMLMOYE74vjdkZu80s4fDKpVO4CMEv8wJP+OVYd7WSFA1Ndy2kWgaEsNhZna3mW0Nq4u+OoIYAH4OHGVm8wlKXZ3u/tRbjEkmACUCOdBtJrigA2BmRnAR3ARsAWaF6wbNLXndBHzF3etKHpXufusIjnsLsByY4+61wHeBweM0AYcM854dQN/rbMsClSXnESeoVio1dKjgfwdWAwvcfRJB1VlpDAcPF3hYqrqdoFTwflQaiDwlAjnQ3Q6cb2Znho2dnySo3nkMeBzIA39rZkkz+1NgScl7/wP4SPjr3sysKmwErhnBcWuANnfvM7MlBNVBg34MnGVml5pZwswazGxhWFq5CfiGmc00s7iZnRy2SawB0uHxk8DngH21VdQAO4FuMzsC+OuSbXcDM8zsE2aWMrMaM3tnyfYfAFcBF6JEEHlKBHJAc/eXCX7ZfpvgF/d7gfe6e87dc8CfElzw2gjaE35W8t6VwIeAfwPagXXhviPxUeA6M+sCvkCQkAY/dyNwHkFSaiNoKD4u3Pwp4A8EbRVtwD8DMXfvDD/zRoLSTBbY4y6iYXyKIAF1ESS1n5TE0EVQ7fNeYCuwFjijZPtvCRqpV7l7aXWZRJBpYhqRaDKzh4Bb3P3G8Y5FxpcSgUgEmdmJwP0EbRxd4x2PjC9VDYlEjJndTNDH4BNKAgIqEYiIRJ5KBCIiEXfADVzV2Njo8+bNG+8wREQOKM8888wOdx/aNwUoYyIws5sIurlvd/djhtluwL8S3GbXA1zl7qv29bnz5s1j5cqVox2uiMiEZmave5twOauGvg+c8wbbzwUWhI9lBL0kRURkjJUtEbj7rwk6zLyei4AfeOAJoM7MZpQrHhERGd54NhbPYs9BtJrDdXsxs2VmttLMVra0tIxJcCIiUXFANBa7+w3ADQCLFy/e637XgYEBmpub6evrG/PYxlI6nWb27Nkkk5o/RERGz3gmgk0Eo0QOmh2ue9Oam5upqalh3rx57DnQ5MTh7rS2ttLc3Mz8+fPHOxwRmUDGs2poOfAX4aiPJxGMib7lrXxQX18fDQ0NEzYJAJgZDQ0NE77UIyJjr5y3j94KnA40hnOxfpFgtijc/bsEUwqeRzDiYw/wl/t5vP15+wEhCucoImOvbInA3S/fx3YHPlau44uIjLtiEfo7oW/oYyfkslDMlzwK4AWIJeDQM/Hpx1FwyBWKDOSd/kKB6lSCyorRv2wfEI3Fb3cdHR3ccsstfPSjH31T7zvvvPO45ZZbqKurK1NkIhHgTr6vi21bNjMQS2CVk4kn0yTjMeIxIxEz0sk4qURsd6naHQZ6oacVOpth56aS500Usy30xzL0xqro8gydxQyt+TSt+QqKxImbEYsZsViMmBnxmDE53sNU2qgvtFGV20GydxvWvT24yL9ZD/0jG30q9xRO4p7CO3nB5wHGVy45hivfedC+3v2mKRGMgo6ODr7zne/slQjy+TyJxOv/iVesWFHu0ERGT64H2l6FtlegownyvVDIQyEXPgagOACZyTBpBtTM3P1cNYWiO707d9DfuZ3czu3ku7ZT6G4h5gUqJ8+kpnE2FXUzoHoaVFQFxywMQGcz3rGRbMt6eravJ9++kcLObcR6dpDqb6O60E6a3B73nvd4inaqafVq2j2Y3nqS9VAb66GGXqrpIcneF+guKtnsDewoTqLS2qihh2rr5RB6ONb69/knavNqtno9272e7Sygp+Jkuirq2JpLsS2XZqdX0UWGnVSR9RR5EiQSCRpqMjROqqRxUiVzMjmO6fpvjm5/kI903MNHE8vZmZnDhhnvoaZ2FiUzs44aJYJRcM011/DKK6+wcOFCkskk6XSa+vp6Vq9ezZo1a7j44otpamqir6+Pq6++mmXLlgG7h8vo7u7m3HPP5dRTT+Wxxx5j1qxZ/PznPyeTyYzzmckBKZ8Di0H8df5753qgdR20roUda/Edayn2tOPxCorxCjxWEbyOVeCFHNb2KomO16jo2TrsxxUsQTGWpBhL4hanYmAnMS/ssc8AccydKitSNYJT6LVK+mMZJhXaiOEYUB0+tnsdbV5Hh9WSSx1FcdIUEjVTqaybRsoGSPR3kujvIJHroCbXQUOug6JDT2wGHVZFExl2ehU7ixk6rIbu1DSy6en0ZWYSz9RQWRGnKpVg+qQ002vTzKhNU1WbhqRBrisoTbizawrp8HVPrJK2btje0cvmjl42tfeyqaOXgUKRhqoKjq6qoKGqgslVKSZXVdBQXcG0mjSTMolh2v/+CPgMZFth9d1MeuFO3vHa9+GYhcDCEfwF35wDbhjqxYsX+9Cxhl566SWOPPJIAL78ixd4cfPOUT3mUTMn8cX3Hv2629evX88FF1zA888/zyOPPML555/P888/v+s2z7a2NiZPnkxvby8nnngijz76KA0NDXskgkMPPZSVK1eycOFCLr30Ui688EKWLl2617FKz1UmIHfobYfu7ZDdHjzHEnDIGZCuBaBvoEB7T47uvjzV6QS1mSSZRAzbvAqeugFeuBMKOTyewpNVDCQy5CxDLylS/a3U5nZf0IsYm7yRVq+hggIVDAQPy1PBAAVibPRprPfpvFacHjz7dJp8Kj2kyBMH9ryIxSgyt6KLwyu7OSS9k4OSncyItVORTFLINFDMNEBlI1Y9hVh1IwVidO/YRH/HFoqdWyC7nYre7cTzPfSkp5GrngW1c0k2zKV6yjym1E/ioIZKGqoqonUDRbYVEilIVb+lt5vZM+6+eLhtKhGUwZIlS/a41/9b3/oWd955JwBNTU2sXbuWhoaGPd4zf/58Fi4MMv0JJ5zA+vXrxyxe2YfedsjuCP4TJjLBczITXKDdg3rmrs2wc3NQx7wzuJiRrILKBqicHD4aIFMP/d14ZzPZlo10bd9AvqOZePdmKvt3UJ1vJzFMlcUACZ6JvYNf5U/gntzxtBC0K6XI8d744/xF/H6Ojb1Klgy/Tp/F5tgkin3dpPr7qbR+Kumjkn522qFsq3g3bZUH0VM9n3zdwdROqqE6nSAe1nXHzEjEg+dk3KhKJZhckWBOKsG5qTjVqQSZijiG4e44UHQPfigDdZkkVak3e2k5ar//mSa8qoZ97/MWTbhE8Ea/3MdKVdXuwu8jjzzCAw88wOOPP05lZSWnn376sH0BUqnUrtfxeJze3t4xiTVyurfDmnsh3wepmuBRUQ2pavKJKrId28lteRHf/hLJ1jVkOteS7ht+WJMiMdyM+JBqkAIxOqyWtPdRxfD/joNVHUXPsNUns41GOhPvoD1RR0esno5YHZ2xenbG66mP9fAuf5oTe3/Ll4o38sW0saP+OLJ1hzOj+V5SAx20ZObz8/q/59HMmbTkkjRUVTCjLsPMugyT69LMrMswozbDpPRw1RASdRMuEYyHmpoaurqGn/Gvs7OT+vp6KisrWb16NU888cQYRyf0tMFLyyn84afENvwG8+KwuyWA2vB11lOs9Vms8yNZUzyLbV5PheVJMUCaHBkbIGMDJGJOZ7yBzuQUulNT6U1NYyDTQCaVolB08gN9pAY6SOc6SA90ksl3UlE5icrGuUyeOZ/Z06dx8JQqTqnNEIu90QV6WVD62P4Stvpupry0nCnr74AjzoclH2LKvNO4yIyLRvtvJ5GgRDAKGhoaOOWUUzjmmGPIZDJMmzZt17ZzzjmH7373uxx55JEcfvjhnHTSSeMY6dtYYQCaV8Jrj0LXlmA5vBvF8/3kB3IMFJxCPM1APMNALEUuliFnKXKxNEVL7m60tAQFS+IDWSY3PcC8zqdIUGBDcTp3Fy/knsJJtHotsyrzzK4sMLsqz/RUnmmpHBVVdRSmHEFy8lxqK1MszCT540ySqooEibjtqj4Zl1/VZjDtqODxrn8I7juPxcc+DplwJlxj8UR3wJxrbwe8+jBsWhXUj0+aBZNmBo+amUE9e9ur8MpD8MpD+Gu/xnLdODF6KiYz4HH6PU5fMU5PIU6fJzCKZMiRoZ+M9VNJPxlyxOz1v8NNPoXH0n/MhunnUDl3IQumT2LB1Gpm1WdIJXQRlehQY7GUX7EIW38H6x6AtQ9A89O7e0kO06GmP15FqpAFYBNTeCT/Tv67+A4eKx7Fzr5q6iqTzKzNMKs+w6y64NFYU0E6ESeVjJFKxEknY6TiMVI2QMwHsGKeeHGAWDFPzHPEYjFmzDmC9+mCL/KGlAhkZIpF6NwI7Rugext0bd3zuWU1ZING1e7Jx7B6zlU8kD+WFa2z6OjqotFbmW5tzKCN6dbGtEI729MHs6XhJFJTD2VeYxUXT67i6oZK5kyupPpN33UiIm+V/rdJwD0Y+ySXhVw3tL0G21+EltX49peg5WVsILvHW/LxND0VjXQlJrMlsZD700fz047D2bE5aHI9uLGKhfNqmV0/h+m1aabWBB10pk9K01hdQSI+noPfisggJYIoGuiF534Mz9wc3E6Z68ZzWYy969pbqGNNcTZriqeyxmez3qez3evY7vV0kYGsYQaN1SmOm13LB06s47g5dRw7u5a6yopxODkRebOUCKKktx2evhF/8ntYtoWO+mN4Nb2EpryxsRCnq5iilzR19fVQN5fOqkOJVU2mKhWnsiLBEakEi1MJ6quS1FVWUF9ZQV0myaRMkvgb3vooIm9nSgQR0L5lPV2PfIvp626lotDDY7aIb/V/hCe3HEEqEWfR3DpOPH4yp86bzPEH1at+XiRi9D9+FLzVYagBvvnNb7Js2TIqKyv3L4imp+DlX0JPK8WeNrrbt9PX1Uqst426Ygc1wN3Fk/jlpPdRfdAizptdyzWzazl6Zi0VCdXVi0SZEsEoeL1hqEfim9/8JkuXLn3riaD1FQr3f5H46l9QtATdsRpaClXsKFbTSR2JqkOomTqXxPFLOeuIY7hYv/ZFZAhdFUZB6TDUZ599NlOnTuX222+nv7+fSy65hC9/+ctks1kuvfRSmpubKRQKfP7zn2fbtm1s3ryZM844g8bGRh5++OE3PE6+WKR3oMC3H1xLy/YtvLP5Jt6T/QU5j/O9/J/zH4XzqKqexLuOmsq7j5jKqQsaqc0kx+ivICIHqomXCH55DWz9w+h+5vR3wLlfe93NX/va13j++ed57rnnuO+++7jjjjt46qmncHcuvPBCfv3rX9PS0sLMmTO55557gGAMotraWr7xjW/w8MMP09jYuOeHDvRRLBbozRfpyRXJ5or0DhTp6O6j86Hv8KnkXVTTy5N15/GHwz7GvOlzuX1qDUfPnLSPMWtERPY08RLBOLvvvvu47777WLRoEQDd3d2sXbuW0047jU9+8pN85jOf4YILLuC0007b+82FAYo9bRSzbSSKfcSAqvAxBSAGWCufS/4YDj0bzr6Ok6cdxcljdnYiMhFNvETwBr/cx4K7c+211/LhD394r22rVq1ixYoVfO5zn+PMM8/kC1/4QvCennbyLe3EB7qJAX2eojXWSLIiTSYZIxW3YDwdL0JmAD5wN8wfJpGIiLwFEy8RjIPSYaj/5E/+hM9//vNceeWVVFdXs2nTJpLJJPl8nsmTJ7N06VLqamq48cbvUWx9lZpMku7NL1M75yDaqKOQrqemupqpFfHhR7hM7YD5S8b4DEVkIlMiGAWlw1Cfe+65XHHFFZx8clBhU11dzY9+9CPWrX6RT3/mM8RwkokY//5Pn6XQ382VV1zOWUs/wcyZs3jkkUfUMUtExpyGoS6nYhH6wmkOB3oAGIil6Shm6ChmSKarmTopRWXFyPPx2/ZcReRtTcNQj7V8P/TsCCab9gIeT5FNTWNLf4refJyadJJZbzIBiIiUi65EoynXE8yu1b8zWE7X0p9qYENXjL7eAjXpJIfUpN7CxN4iIuUzYa5I7j6+k3L3dwUzblkMqqfjlQ209jpbOvqImzOvsYpJ6f3r3HWgVeOJyIFhQiSCdDpNa2srDQ0N45MM+nYG4/cnktBwKAMkaG7vpatvgEnpJLPqMyT3c+x9d6e1tZV0Oj1KQYuIBCZEIpg9ezbNzc20tLSM/cEHeoPG4HgSqqbQt3UN7dkcRaA2k6Q3lWDdKIWVTqeZPXv26HyYiEhoQiSCZDLJ/Pnzx/7AL9wJd30wGIJi6c/4tyda+fp9azhieg3funwRh02rGfuYRETepAmRCMbF726Du/4aZi+BK2/n56u7+fp9a7h44Uy+9mfHkk5qwnQROTCUdSB6MzvHzF42s3Vmds0w2w8yswfN7Pdm9oiZHRj1Hs/+CO78CMw7FZb+lGe3F/n0Hb9nyfzJ/J8/P05JQEQOKGVLBGYWB64HzgWOAi43s6OG7PZ14AfufixwHfBP5Ypn1OR64FfXBkngitvZ3Btn2Q+fYdqkFN9deoImeRGRA045r1pLgHXu/qq754DbgIuG7HMU8FD4+uFhtr/9vPjzoJ/A6dfS40k+ePNKenMF/vMDJzK5SpO1i8iBp5yJYBbQVLLcHK4r9TvgT8PXlwA1ZtYw9IPMbJmZrTSzleNyZ1CpZ38Ikw+hOOdk/v4nv2P11p18+wo1DIvIgWu86zE+BbzLzJ4F3gVsAgpDd3L3G9x9sbsvnjJlyljHuNuOdbDht3D8+/nGA2v51Qtb+ex5R3LG4VPHLyYRkf1UzruGNgFzSpZnh+t2cffNhCUCM6sG/szdO8oY0/559odgcX6VOIN/e3gd71s8h786dRxuWxURGUXlLBE8DSwws/lmVgFcBiwv3cHMGs1sMIZrgZvKGM/+KQzAc7eQO+Q9fPqX2zhxXj3/ePEx4zushYjIKChbInD3PPBx4F7gJeB2d3/BzK4zswvD3U4HXjazNcA04Cvlime/rb0Pstu5o3gG2Vyer17yDt0hJCITQlk7lLn7CmDFkHVfKHl9B3BHOWMYNat+QL5yGl9ePZP3nTiHBWocFpEJQj9pR2LnZlh7Hw+kziKeSPJ3Zx023hGJiIwaJYKReO4W8CJf3bqYD512MFMnaQRQEZk4NNbQvhSL+LM/5IWK4+hJzGXZHx883hGJiIwqlQj2ZcNvsPb1/Ef3Kfzd2Qs0u5iITDhKBPtQfOZmuqhizeTTed/iOft+g4jIAUaJ4I30tuMvLudn+T/i789bSGI/ZxkTEXk70pXtDfSvuo14Mcfz0y7mrCM1jISITEyq8H497nQ+dhPbivO48uIL1INYRCYslQheR+7pm5maXcPvZvwPFs6pG+9wRETKRolgOG2vEb/vs/y2cDQz3vVX4x2NiEhZKREMVSzAXR8l78anBz7MooP2mh5BRGRCUSIY6vHrYeNj/Gjyx0k1HqRZx0RkwlMiKLXtBXjoH/Ej38u/t53AorlqGxCRiU+JYFA+Bz/7MKRr2XTKV9mRHeD4ufXjHZWISNnp9tFBj34Ntv0BLruVlS1xACUCEYkElQgANj4Jv/kXWLQUjjiPVRvbqaqIc/h0zTkgIhOfEkGuB+78MNTOhj/5JwBWbWznuDl1xGPqRCYiE58SwYbHoP01OOefIT2Jnlyel7Z0qVpIRCJDiaBjQ/A84zgAft/cSaHoHH+Q7hgSkWhQIuhsglgCaqYDQbUQwKI5KhGISDQoEXQ0waRZEAvuFFq1oYODG6uoV0cyEYkIJYLOJqibC4C78+zGdhapfUBEIkSJoKMJaoOZxza29dCazalHsYhESrQTQT4HXVugLkgEg+0DumNIRKIk2olg5ybAd5UIVm3oUEcyEYmcaCeCzqbguaREoI5kIhI10U4EHWEiqJ1DTy7P6q3qSCYi0RPtRDBYIqidze+a1JFMRKIp2omgowmqp0MipY5kIhJZ0U4EnRt3tQ88u7FdHclEJJKinQjCPgTuzqqNHepIJiKRVNZEYGbnmNnLZrbOzK4ZZvtcM3vYzJ41s9+b2XnljGcPxWJw+2jdHDa29dCWzal9QEQiqWyJwMziwPXAucBRwOVmdtSQ3T4H3O7ui4DLgO+UK569dG+DQg5q56gjmYhEWjlLBEuAde7+qrvngNuAi4bs48Ck8HUtsLmM8expVx+Cuaza0EF1KsFh09SRTESip5yJYBbQVLLcHK4r9SVgqZk1AyuAvxnug8xsmZmtNLOVLS0toxNdx8bgOSwRHDenVh3JRCSSxrux+HLg++4+GzgP+KGZ7RWTu9/g7ovdffGUKVNG58hhiaCncoY6kolIpJUzEWwC5pQszw7Xlfor4HYAd38cSAONZYxpt44mSNfxUptTKDoL56ihWESiqZyJ4GlggZnNN7MKgsbg5UP22QicCWBmRxIkglGq+9mHziaom0N7dgCAxurUmBxWROTtpmyJwN3zwMeBe4GXCO4OesHMrjOzC8PdPgl8yMx+B9wKXOXuXq6Y9tDRBLVzyebyAFSlEmNyWBGRt5sRXf3M7GfAfwK/dPfiSD/c3VcQNAKXrvtCyesXgVNG+nmjxj0oERz8Lrr7g0RQrUQgIhE10hLBd4ArgLVm9jUzO7yMMZVfbzvkuqF2Dtn+wRJBfJyDEhEZHyNKBO7+gLtfCRwPrAceMLPHzOwvzSxZzgDLomQegu7+AgBVFSoRiEg0jbiNwMwagKuADwLPAv9KkBjuL0tk5VQyD0G2P09lRZyY+hCISESNtI3gTuBw4IfAe919S7jpJ2a2slzBlU1Jr+Js/2a1D4hIpI30Cvgtd394uA3uvngU4xkbHU2QyEBlA939G5UIRCTSRlo1dJSZ7epxZWb1ZvbRMsVUfoPzEJiR7c/r1lERibSRJoIPuXvH4IK7twMfKk9IY6CjCWpnA9Ddn9cdQyISaSNNBHEz29WaGg4xfeBO5dUZTEgD0N1fUNWQiETaSK+AvyJoGP5euPzhcN2BJ5eFntZdU1SqakhEom6kV8DPEFz8/zpcvh+4sSwRlVtnc/BcOxdQIhARGdEVMBxW4t/Dx4GtY3dnMgjaCFQ1JCJRNtJ+BAuAfyKYcjI9uN7dDy5TXOXTubszWb5QpD9fVK9iEYm0kTYW/z+C0kAeOAP4AfCjcgVVVp1NYHGomUF2cHgJ3TUkIhE20kSQcfcHAXP3De7+JeD88oVVRh1NMGkWxBN0h0NQ16RVIhCR6BrpFbA/nEJyrZl9nGCmseryhVVG4YQ0QMnIo0oEIhJdIy0RXA1UAn8LnAAsBT5QrqDKqmN3H4KuPiUCEZF9XgHDzmPvc/dPAd3AX5Y9qnIpDEDX5r1KBLprSESibJ8lAncvAKeOQSzlt3MzeHFXiWBX1ZDuGhKRCBvpFfBZM1sO/BeQHVzp7j8rS1Tl0rl3HwJQiUBEom2kV8A00Aq8u2SdAwdWItg1Ic3uXsWg20dFJNpG2rP4wG0XKLWrM1kw8mg2N9iPQCUCEYmukfYs/n8EJYA9uPv/HPWIyqljI1RNhWTQObq7P08iZqQSI56xU0RkwhnpT+G7S16ngUuAzaMfTpmV9CGA3QPOlYywLSISOSOtGvpp6bKZ3Qr8piwRlVNHE0x/x65FDTgnIjLyDmVDLQCmjmYgZVcsBkNQDykRKBGISNSNtI2giz3bCLYSzFFw4Mi2QKF/1x1DoGkqRURg5FVDNeUOpOyG9CGAYJrKSRpwTkQibkRVQ2Z2iZnVlizXmdnF5QurDDo2Bs+1qhoSESk10jaCL7p75+CCu3cAXyxPSGUyTIlA01SKiIz89tHhEsaBdQU94oJgHoL0roKN7hoSEWHkF/OVZvYN4Ppw+WPAM+UJqUwaDgkeIXcPSwRqLBaRaBtp1dDfADngJ8BtQB9BMnhDZnaOmb1sZuvM7Jphtv+LmT0XPtaYWcebCX5/9A0UKbqGlxARGeldQ1lgrwv5GwnnMbgeOBtoBp42s+Xu/mLJ5/5dyf5/Ayx6M8fYH4Mjj9YoEYhIxI30rqH7zayuZLnezO7dx9uWAOvc/VV3zxGUJC56g/0vB24dSTyjQdNUiogERlo11BjeKQSAu7ez757Fs4CmkuXmcN1ezOwgYD7w0NwpP1MAAAv+SURBVOtsX2ZmK81sZUtLywhDfmPdSgQiIsDIE0HRzHZ1yTWzeQwzGul+uAy4I5wNbS/ufoO7L3b3xVOmTBmVA2pSGhGRwEivgv8L+I2ZPQoYcBqwbB/v2QTMKVmeHa4bzmWMoPF5NKlqSEQkMKISgbv/ClgMvExQj/9JoHcfb3saWGBm882sguBiv3zoTmZ2BFAPPP4m4t5vu0sEun1URKJtpIPOfRC4muBX/XPASQQX7ne/3nvcPW9mHwfuBeLATe7+gpldB6x098GkcBlwm7uPZlXTPmX7NTuZiAiMvGroauBE4Al3PyP8Ff/Vfb3J3VcAK4as+8KQ5S+NMIZRpaohEZHASBuL+9y9D8DMUu6+Gji8fGGV3667hiqUCEQk2kZ6FWwO+xHcBdxvZu3AhvKFVX7Z/jyZZJx4TNNUiki0jbRn8SXhyy+Z2cNALfCrskU1BrK5PNWai0BE5M2PIOruj5YjkLHW3V9QHwIREd76nMUHvO6+AY08KiJChBNBtr+ghmIRESKcCDQpjYhIILKJIJvTNJUiIhDlRKD5ikVEgAgngqBqSI3FIiKRTAT5QpG+gaJKBCIiRDQRZHPBgHNqLBYRiWoi0KQ0IiK7RDoRqGpIRCSiiaBLJQIRkV0imQhUIhAR2S3iiUC3j4qIRDIRdPfrriERkUGRTASqGhIR2S2SiaBbjcUiIrtEMhFk+/PEY0YqEcnTFxHZQySvhNlwCGozzVcsIhLJRKBpKkVEdotoItA0lSIigyKZCLL9Bd0xJCISimQi0DSVIiK7RTIRZPvzmrheRCQU3USgEoGICBDRRKBpKkVEdotcInB3sjk1FouIDCprIjCzc8zsZTNbZ2bXvM4+l5rZi2b2gpndUs54APrzRQpFVyIQEQmV7WpoZnHgeuBsoBl42syWu/uLJfssAK4FTnH3djObWq54Bg2OM1STViIQEYHylgiWAOvc/VV3zwG3ARcN2edDwPXu3g7g7tvLGA9QMvKo7hoSEQHKmwhmAU0ly83hulKHAYeZ2W/N7AkzO6eM8QDQ1achqEVESo331TABLABOB2YDvzazd7h7R+lOZrYMWAYwd+7c/TpgVkNQi4jsoZwlgk3AnJLl2eG6Us3AcncfcPfXgDUEiWEP7n6Duy9298VTpkzZr6CyOU1TKSJSqpyJ4GlggZnNN7MK4DJg+ZB97iIoDWBmjQRVRa+WMSZNUykiMkTZEoG754GPA/cCLwG3u/sLZnadmV0Y7nYv0GpmLwIPA59299ZyxQSaplJEZKiyXg3dfQWwYsi6L5S8duDvw8eYUCIQEdlT5HoWd++6fVRtBCIiEMFEkO3Pk07GSMQjd+oiIsOK3NUwmKYyOd5hiIi8bUQwEWjkURGRUpFLBJqLQERkT5FLBN1KBCIie4hcIshqvmIRkT1EMhGoRCAislvkEkFw15Aai0VEBkUuEWT785qLQESkRKQSQaHo9A5ovmIRkVKRSgSDQ1BrmkoRkd2ilQg04JyIyF4ilQi6NU2liMheopUIdk1TqbuGREQGRSoRZMPZyXTXkIjIbpFKBN1qIxAR2UukEkF2V9WQEoGIyKBoJYKcSgQiIkNFKhF0q0QgIrKXSCWCbH+emEE6GanTFhF5Q5G6Imb7C1SnEpjZeIciIvK2EalE0K25CERE9hKtRNCnuQhERIaKVCLI5pQIRESGilQiUNWQiMjeIpUIgmkqNc6QiEipiCUCTUojIjJUpBKBqoZERPYWmUTg7mHVkBKBiEipyCSC/nyRfNFVIhARGaKsicDMzjGzl81snZldM8z2q8ysxcyeCx8fLFcsGnlURGR4ZbsqmlkcuB44G2gGnjaz5e7+4pBdf+LuHy9XHIM0F4GIyPDKWSJYAqxz91fdPQfcBlxUxuO9IU1TKSIyvHImgllAU8lyc7huqD8zs9+b2R1mNme4DzKzZWa20sxWtrS0vKVgdk1TqRKBiMgexrux+BfAPHc/FrgfuHm4ndz9Bndf7O6Lp0yZ8pYOlFXVkIjIsMqZCDYBpb/wZ4frdnH3VnfvDxdvBE4oVzCalEZEZHjlTARPAwvMbL6ZVQCXActLdzCzGSWLFwIvlSsYlQhERIZXtquiu+fN7OPAvUAcuMndXzCz64CV7r4c+FszuxDIA23AVeWKZ1eJoEKJQESkVFmviu6+AlgxZN0XSl5fC1xbzhgGzZ1cyTlHT9egcyIiQ0Tm5/F7jp7Oe46ePt5hiIi87Yz3XUMiIjLOlAhERCJOiUBEJOKUCEREIk6JQEQk4pQIREQiTolARCTilAhERCLO3H28Y3hTzKwF2PAW394I7BjFcA4UUT1viO6567yjZSTnfZC7Dzt88wGXCPaHma1098XjHcdYi+p5Q3TPXecdLft73qoaEhGJOCUCEZGIi1oiuGG8AxgnUT1viO6567yjZb/OO1JtBCIisreolQhERGQIJQIRkYiLTCIws3PM7GUzW2dm14x3POViZjeZ2XYze75k3WQzu9/M1obP9eMZYzmY2Rwze9jMXjSzF8zs6nD9hD53M0ub2VNm9rvwvL8crp9vZk+G3/efhPOGTzhmFjezZ83s7nB5wp+3ma03sz+Y2XNmtjJct1/f80gkAjOLA9cD5wJHAZeb2VHjG1XZfB84Z8i6a4AH3X0B8GC4PNHkgU+6+1HAScDHwn/jiX7u/cC73f04YCFwjpmdBPwz8C/ufijQDvzVOMZYTlcDL5UsR+W8z3D3hSV9B/brex6JRAAsAda5+6vungNuAy4a55jKwt1/DbQNWX0RcHP4+mbg4jENagy4+xZ3XxW+7iK4OMxigp+7B7rDxWT4cODdwB3h+gl33gBmNhs4H7gxXDYicN6vY7++51FJBLOAppLl5nBdVExz9y3h663AtPEMptzMbB6wCHiSCJx7WD3yHLAduB94Behw93y4y0T9vn8T+AegGC43EI3zduA+M3vGzJaF6/brex6Zyesl4O5uZhP2nmEzqwZ+CnzC3XcGPxIDE/Xc3b0ALDSzOuBO4IhxDqnszOwCYLu7P2Nmp493PGPsVHffZGZTgfvNbHXpxrfyPY9KiWATMKdkeXa4Liq2mdkMgPB5+zjHUxZmliRIAj9295+FqyNx7gDu3gE8DJwM1JnZ4A+9ifh9PwW40MzWE1T1vhv4Vyb+eePum8Ln7QSJfwn7+T2PSiJ4GlgQ3lFQAVwGLB/nmMbScuAD4esPAD8fx1jKIqwf/k/gJXf/RsmmCX3uZjYlLAlgZhngbIL2kYeBPw93m3Dn7e7Xuvtsd59H8P/5IXe/kgl+3mZWZWY1g6+B9wDPs5/f88j0LDaz8wjqFOPATe7+lXEOqSzM7FbgdIJhabcBXwTuAm4H5hIM4X2puw9tUD6gmdmpwH8Df2B3nfFnCdoJJuy5m9mxBI2DcYIfdre7+3VmdjDBL+XJwLPAUnfvH79IyyesGvqUu18w0c87PL87w8UEcIu7f8XMGtiP73lkEoGIiAwvKlVDIiLyOpQIREQiTolARCTilAhERCJOiUBEJOKUCETGkJmdPjhSpsjbhRKBiEjEKRGIDMPMlobj/D9nZt8LB3brNrN/Ccf9f9DMpoT7LjSzJ8zs92Z25+BY8GZ2qJk9EM4VsMrMDgk/vtrM7jCz1Wb2YysdEElkHCgRiAxhZkcC7wNOcfeFQAG4EqgCVrr70cCjBL22AX4AfMbdjyXo2Ty4/sfA9eFcAX8EDI4OuQj4BMHcGAcTjJsjMm40+qjI3s4ETgCeDn+sZwgG8SoCPwn3+RHwMzOrBerc/dFw/c3Af4Xjwcxy9zsB3L0PIPy8p9y9OVx+DpgH/Kb8pyUyPCUCkb0ZcLO7X7vHSrPPD9nvrY7PUjr2TQH9P5Rxpqohkb09CPx5ON774HywBxH8fxkc2fIK4Dfu3gm0m9lp4fr3A4+Gs6Q1m9nF4WekzKxyTM9CZIT0S0RkCHd/0cw+RzALVAwYAD4GZIEl4bbtBO0IEAz7+93wQv8q8Jfh+vcD3zOz68LP+B9jeBoiI6bRR0VGyMy63b16vOMQGW2qGhIRiTiVCEREIk4lAhGRiFMiEBGJOCUCEZGIUyIQEYk4JQIRkYj7/9RZXiSLR7HWAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yK5QKNScFby7"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}