CNN-MNIST.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "CNN\\Convulated Neural network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OF-aSlVcmTge"
      },
      "source": [
        "#**The MNIST Dataset**\n",
        "\n",
        "\n",
        "MNIST: *Modified National Institute of Standards and Technology*"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "VSF51YZw9g0T"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From c:\\Users\\Asus\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\n",
            "\n"
          ]
        }
      ],
      "source": [
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
        "import matplotlib.pyplot as plt # plotting library\n",
        "%matplotlib inline\n",
        "from keras.models import Sequential\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.optimizers import Adam # - Works ,RMSprop\n",
        "from tensorflow.keras.utils import to_categorical, plot_model\n",
        "from keras import  backend as K\n",
        "from keras.layers import Dense\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Activation,Flatten\n",
        "import random\n",
        "random.seed(0)\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "gaJHcSTN9lSe"
      },
      "outputs": [],
      "source": [
        "# import dataset\n",
        "from keras.datasets import mnist\n",
        "\n",
        "\n",
        "# load dataset\n",
        "(x_train, y_train),(x_test, y_test) = mnist.load_data()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KuygmbsdkP-1",
        "outputId": "83be697a-36c8-496c-b8b2-1c7cf1caf272"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array([5, 0, 4, ..., 5, 6, 8], dtype=uint8)"
            ]
          },
          "execution_count": 3,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "y_train"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "weWo-QuTkLri",
        "outputId": "d5c04767-0451-4e1d-9748-e1f6b6ab3e97"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train labels:  {0: 5923, 1: 6742, 2: 5958, 3: 6131, 4: 5842, 5: 5421, 6: 5918, 7: 6265, 8: 5851, 9: 5949}\n",
            "\n",
            "Test labels:  {0: 980, 1: 1135, 2: 1032, 3: 1010, 4: 982, 5: 892, 6: 958, 7: 1028, 8: 974, 9: 1009}\n"
          ]
        }
      ],
      "source": [
        "# count the number of unique train labels\n",
        "unique, counts = np.unique(y_train, return_counts=True)\n",
        "print(\"Train labels: \", dict(zip(unique, counts)))\n",
        "\n",
        "\n",
        "\n",
        "# count the number of unique test labels\n",
        "unique, counts = np.unique(y_test, return_counts=True)\n",
        "print(\"\\nTest labels: \", dict(zip(unique, counts)))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303
        },
        "id": "2s1sHopT-Gpu",
        "outputId": "6c73981d-164c-4dd5-df47-2e97ca2aece1"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 500x500 with 25 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# sample 25 mnist digits from train dataset\n",
        "indexes = np.random.randint(0, x_train.shape[0], size=25)\n",
        "images = x_train[indexes]\n",
        "labels = y_train[indexes]\n",
        "\n",
        "\n",
        "# plot the 25 mnist digits\n",
        "plt.figure(figsize=(5,5))\n",
        "for i in range(len(indexes)):\n",
        "    plt.subplot(5, 5, i + 1)\n",
        "    image = images[i]\n",
        "    plt.imshow(image, cmap='gray')\n",
        "    plt.axis('off')\n",
        "\n",
        "plt.show()\n",
        "# plt.savefig(\"mnist-samples.png\")\n",
        "# plt.close('all')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "AOYQNIXx-Kko"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "10"
            ]
          },
          "execution_count": 6,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# compute the number of labels\n",
        "num_labels = len(np.unique(y_train))\n",
        "num_labels"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "KYNphycF-q0e"
      },
      "outputs": [],
      "source": [
        "# convert to one-hot vector\n",
        "y_train = to_categorical(y_train)\n",
        "y_test = to_categorical(y_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "0cj3Hiyl_PPN"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(60000, 28, 28)"
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "x_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "FwqjLzGz-yx6"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "784"
            ]
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# image dimensions (assumed square)\n",
        "image_size = x_train.shape[1]\n",
        "input_size = image_size * image_size\n",
        "input_size"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "mBG_aYRa_vbk"
      },
      "outputs": [],
      "source": [
        "# resize and normalize\n",
        "x_train = np.reshape(x_train, [-1, input_size]) #add -1 to match new reshape with original train data.\n",
        "x_train = x_train.astype('float32') / 255\n",
        "x_test = np.reshape(x_test, [-1, input_size])\n",
        "x_test = x_test.astype('float32') / 255"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "LxAzfNznpTNF"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(60000, 784)"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "x_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "W5o8GxNl_8KF"
      },
      "outputs": [],
      "source": [
        "# network parameters\n",
        "batch_size = 128\n",
        "hidden_units = 256\n",
        "dropout = 0.45"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "w4KHIk-PDX6-"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From c:\\Users\\Asus\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\keras\\src\\backend.py:873: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
            "\n"
          ]
        }
      ],
      "source": [
        "model = Sequential()\n",
        "model.add(Dense(hidden_units, input_dim=input_size,activation = 'relu'))\n",
        "model.add(Dense(hidden_units,activation = 'relu'))\n",
        "model.add(Dense(num_labels, activation='softmax'))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "rz4LDhrHEEdG"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense (Dense)               (None, 256)               200960    \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 256)               65792     \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 10)                2570      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 269322 (1.03 MB)\n",
            "Trainable params: 269322 (1.03 MB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "model.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "9mW-MgL5EMiN"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "execution_count": 15,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "plot_model(model, to_file='mlp-mnist.png', show_shapes=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "hffHa8XSETB-"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From c:\\Users\\Asus\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\keras\\src\\optimizers\\__init__.py:309: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
            "\n"
          ]
        }
      ],
      "source": [
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer='adam',\n",
        "              metrics=['accuracy'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "Wx9pbG0QGpqM"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "WARNING:tensorflow:From c:\\Users\\Asus\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\keras\\src\\utils\\tf_utils.py:492: The name tf.ragged.RaggedTensorValue is deprecated. Please use tf.compat.v1.ragged.RaggedTensorValue instead.\n",
            "\n",
            "WARNING:tensorflow:From c:\\Users\\Asus\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\keras\\src\\engine\\base_layer_utils.py:384: The name tf.executing_eagerly_outside_functions is deprecated. Please use tf.compat.v1.executing_eagerly_outside_functions instead.\n",
            "\n",
            "329/329 [==============================] - 3s 6ms/step - loss: 0.3113 - accuracy: 0.9098 - val_loss: 0.1547 - val_accuracy: 0.9539\n",
            "Epoch 2/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.1165 - accuracy: 0.9650 - val_loss: 0.1212 - val_accuracy: 0.9634\n",
            "Epoch 3/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0736 - accuracy: 0.9779 - val_loss: 0.0978 - val_accuracy: 0.9707\n",
            "Epoch 4/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0514 - accuracy: 0.9843 - val_loss: 0.0896 - val_accuracy: 0.9742\n",
            "Epoch 5/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0390 - accuracy: 0.9882 - val_loss: 0.1024 - val_accuracy: 0.9709\n",
            "Epoch 6/20\n",
            "329/329 [==============================] - 2s 5ms/step - loss: 0.0266 - accuracy: 0.9923 - val_loss: 0.0906 - val_accuracy: 0.9746\n",
            "Epoch 7/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0206 - accuracy: 0.9937 - val_loss: 0.0990 - val_accuracy: 0.9741\n",
            "Epoch 8/20\n",
            "329/329 [==============================] - 2s 5ms/step - loss: 0.0198 - accuracy: 0.9936 - val_loss: 0.1156 - val_accuracy: 0.9717\n",
            "Epoch 9/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0160 - accuracy: 0.9946 - val_loss: 0.1126 - val_accuracy: 0.9718\n",
            "Epoch 10/20\n",
            "329/329 [==============================] - 2s 5ms/step - loss: 0.0171 - accuracy: 0.9942 - val_loss: 0.1137 - val_accuracy: 0.9736\n",
            "Epoch 11/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0164 - accuracy: 0.9950 - val_loss: 0.1031 - val_accuracy: 0.9755\n",
            "Epoch 12/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0046 - accuracy: 0.9989 - val_loss: 0.1144 - val_accuracy: 0.9769\n",
            "Epoch 13/20\n",
            "329/329 [==============================] - 2s 5ms/step - loss: 0.0067 - accuracy: 0.9978 - val_loss: 0.1310 - val_accuracy: 0.9737\n",
            "Epoch 14/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0146 - accuracy: 0.9947 - val_loss: 0.1219 - val_accuracy: 0.9744\n",
            "Epoch 15/20\n",
            "329/329 [==============================] - 2s 5ms/step - loss: 0.0087 - accuracy: 0.9969 - val_loss: 0.1299 - val_accuracy: 0.9739\n",
            "Epoch 16/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0086 - accuracy: 0.9973 - val_loss: 0.1222 - val_accuracy: 0.9760\n",
            "Epoch 17/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0071 - accuracy: 0.9977 - val_loss: 0.1327 - val_accuracy: 0.9736\n",
            "Epoch 18/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0066 - accuracy: 0.9976 - val_loss: 0.1371 - val_accuracy: 0.9758\n",
            "Epoch 19/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0081 - accuracy: 0.9972 - val_loss: 0.1431 - val_accuracy: 0.9747\n",
            "Epoch 20/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0108 - accuracy: 0.9963 - val_loss: 0.1259 - val_accuracy: 0.9776\n"
          ]
        }
      ],
      "source": [
        "H = model.fit(x_train, y_train, epochs=20, batch_size=batch_size, validation_split=0.3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "5eUKrGCQGryE"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 0s 1ms/step - loss: 0.1009 - accuracy: 0.9802\n",
            "\n",
            "Test accuracy: 98.0%\n"
          ]
        }
      ],
      "source": [
        "loss, acc = model.evaluate(x_test, y_test)\n",
        "print(\"\\nTest accuracy: %.1f%%\" % (100.0 * acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "eVclDsNtHDxt"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1875/1875 [==============================] - 3s 2ms/step - loss: 0.0396 - accuracy: 0.9929\n",
            "\n",
            "Test accuracy: 99.3%\n"
          ]
        }
      ],
      "source": [
        "loss, acc = model.evaluate(x_train, y_train)\n",
        "print(\"\\nTest accuracy: %.1f%%\" % (100.0 * acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "PUGEEu10shI3"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x25e3dce73d0>"
            ]
          },
          "execution_count": 20,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x800 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "epochs = np.arange(0, len(H.history[\"loss\"]))\n",
        "plt.figure(figsize=(10,8))\n",
        "plt.style.use(\"ggplot\")\n",
        "\n",
        "\n",
        "# plot training and validation loss\n",
        "plt.style.use(\"ggplot\")\n",
        "plt.plot(epochs, H.history[\"loss\"], label=\"train_loss\")\n",
        "plt.plot(epochs, H.history[\"val_loss\"], label=\"val_loss\")\n",
        "plt.title(\"Training Vs Validation Loss\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DwrxSgv5vKmb"
      },
      "source": [
        "# Adding L2 Regualarisation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "id": "36ALHOscsUa9"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "329/329 [==============================] - 3s 8ms/step - loss: 0.4932 - accuracy: 0.9105 - val_loss: 0.3029 - val_accuracy: 0.9489\n",
            "Epoch 2/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2396 - accuracy: 0.9608 - val_loss: 0.2196 - val_accuracy: 0.9605\n",
            "Epoch 3/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1780 - accuracy: 0.9719 - val_loss: 0.1834 - val_accuracy: 0.9693\n",
            "Epoch 4/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1487 - accuracy: 0.9762 - val_loss: 0.1784 - val_accuracy: 0.9653\n",
            "Epoch 5/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1325 - accuracy: 0.9787 - val_loss: 0.1622 - val_accuracy: 0.9703\n",
            "Epoch 6/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1220 - accuracy: 0.9806 - val_loss: 0.1705 - val_accuracy: 0.9670\n",
            "Epoch 7/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1157 - accuracy: 0.9818 - val_loss: 0.1607 - val_accuracy: 0.9706\n",
            "Epoch 8/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1098 - accuracy: 0.9838 - val_loss: 0.1562 - val_accuracy: 0.9701\n",
            "Epoch 9/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1031 - accuracy: 0.9847 - val_loss: 0.1619 - val_accuracy: 0.9688\n",
            "Epoch 10/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0983 - accuracy: 0.9861 - val_loss: 0.1483 - val_accuracy: 0.9714\n",
            "Epoch 11/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0994 - accuracy: 0.9847 - val_loss: 0.1640 - val_accuracy: 0.9691\n",
            "Epoch 12/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0932 - accuracy: 0.9867 - val_loss: 0.1612 - val_accuracy: 0.9694\n",
            "Epoch 13/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0954 - accuracy: 0.9859 - val_loss: 0.1507 - val_accuracy: 0.9734\n",
            "Epoch 14/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0843 - accuracy: 0.9889 - val_loss: 0.1482 - val_accuracy: 0.9717\n",
            "Epoch 15/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0895 - accuracy: 0.9866 - val_loss: 0.1693 - val_accuracy: 0.9681\n",
            "Epoch 16/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0774 - accuracy: 0.9909 - val_loss: 0.1372 - val_accuracy: 0.9735\n",
            "Epoch 17/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0853 - accuracy: 0.9876 - val_loss: 0.1469 - val_accuracy: 0.9720\n",
            "Epoch 18/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0805 - accuracy: 0.9889 - val_loss: 0.1532 - val_accuracy: 0.9698\n",
            "Epoch 19/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0777 - accuracy: 0.9895 - val_loss: 0.1656 - val_accuracy: 0.9658\n",
            "Epoch 20/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.0786 - accuracy: 0.9895 - val_loss: 0.1457 - val_accuracy: 0.9722\n"
          ]
        }
      ],
      "source": [
        "from keras.layers import Dense , Activation, Dropout\n",
        "import keras\n",
        "\n",
        "\n",
        "model = Sequential()\n",
        "model.add(Dense(hidden_units, input_dim=input_size, kernel_regularizer=keras.regularizers.l2(0.001)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dense(hidden_units))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dense(num_labels))\n",
        "model.add(Activation('softmax'))\n",
        "\n",
        "\n",
        "\n",
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer='adam',\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "\n",
        "\n",
        "H = model.fit(x_train, y_train, epochs=20, batch_size=batch_size,validation_split=0.3) #you can specify a validation set or not (it's optional). You can add \"validation_split=0.3\" in case you want to add validation set."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "id": "Uep67Vh3uGaY"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 1s 2ms/step - loss: 0.1253 - accuracy: 0.9767\n",
            "\n",
            "Test accuracy: 97.7%\n"
          ]
        }
      ],
      "source": [
        "loss, acc = model.evaluate(x_test, y_test)\n",
        "print(\"\\nTest accuracy: %.1f%%\" % (100.0 * acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "Rx6q316Iu1Vy"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1875/1875 [==============================] - 3s 1ms/step - loss: 0.0933 - accuracy: 0.9869\n",
            "\n",
            "Test accuracy: 98.7%\n"
          ]
        }
      ],
      "source": [
        "loss, acc = model.evaluate(x_train, y_train)\n",
        "print(\"\\nTest accuracy: %.1f%%\" % (100.0 * acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "id": "XKLp5KUQu36v"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x25e3ffb73d0>"
            ]
          },
          "execution_count": 24,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1IAAALCCAYAAADUN+LwAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAACmhElEQVR4nOzdd3jV5f3/8dfn5JzsTRISEkgCYQmIA1DZS0XEvbXfXx3F2eGo2jrqqla0Vdtqh9pqtVZxsRyIgyUoAgoyZY+EhITsnZycz++PkxyIJJB55vNxXV7k7Pe5c4S8ct/3+zZM0zQFAAAAAGgzi6cLAAAAAABfQ5ACAAAAgHYiSAEAAABAOxGkAAAAAKCdCFIAAAAA0E4EKQAAAABoJ4IUAAAAALQTQQoAAAAA2okgBQAAAADtRJACAA8xDEMTJ07s9PNMnDhRhmF0viC0ycMPPyzDMLRkyZJm17f3+7lkyRIZhqGHH364S+v7sdbqBQB0DkEKQMAyDKNd/7366queLtnrORwO9enTR4ZhaPPmzce8b1VVlWJjYxUcHKz8/PxOv/b9998vwzB0zz33HPe+N954owzD0LPPPtvp1/W0V1991ec+n03hrrtDJAB0J6unCwAAT3nooYeOuu65555TaWmpfvWrXyk2NrbZbSeddFKXvv6WLVsUHh7e6ed57bXXVFVV1QUVdZ7FYtH111+vRx55RC+//LKeeeaZVu/7zjvvqLS0VJdeeqmSkpI6/do/+9nP9Ic//EGvvfaaHn/8cdlsthbvV1lZqbfeekshISH66U9/2unXbdJV38+u9vOf/1xXXnml+vTp4+lSAMCvEKQABKyWfhv+6quvqrS0VLfffrsyMjK69fUHDRrUJc/jbT8g33DDDfr973+v119/XU8++aSCg4NbvN/LL78syTk71BUyMzM1depUffrpp1qwYIEuvvjiFu/31ltvqby8XFdffbXi4+O75LWlrvt+drWEhAQlJCR4ugwA8Dss7QOANmjah1RXV6dHH31UAwcOVEhIiK699lpJUmlpqZ5++mlNnjxZaWlpCg4OVmJios4//3x99dVXLT5nS3tqjtzP8u6772rUqFEKDw9XfHy8rrzySuXk5LRa25GO3H+zbt06nXvuuYqNjVV4eLgmTJiglStXtlhTbm6urrvuOiUlJSksLEwnnXSS/vOf/7RrP0/v3r01bdo0HTp0SHPmzGnxPlu3btWXX36pvn37aurUqa7rd+3apRtvvFFZWVkKCwtTfHy8hg0bpptvvlmFhYXHfe2mUPbSSy+1ep+m25ruu3jxYt1444064YQTFB0drbCwMA0dOlSPPPKIampqjvuaTVrbI3Xw4EHdcMMN6tmzZ7Mxbc3atWv1q1/9SsOHD1d8fLxCQ0PVv39/3XXXXSouLm5234kTJ+q6666TJF133XXNlqLu2bNH0rH3SH3++eeaNm2a4uPjFRISogEDBug3v/mNSktLj7pv0+fMbrfriSeeUP/+/RUSEqLevXvr3nvvVV1dXZvHqr3Wrl2rSy65RElJSQoJCVF6erpuvfVW5ebmHnXfgwcP6te//rUGDhyoiIgIxcbGauDAgbr22mu1a9cu1/1M09R//vMfjR49WomJiQoNDVXv3r119tlna/bs2d32XgD4D2akAKAdLrnkEq1evVrnnHOOLrzwQteStC1btuj+++/X+PHjde655youLk779u3T/Pnz9fHHH2vBggWaNm1am1/nb3/7m+bPn6/zzz9fEyZM0KpVqzR79mytX79e69atU0hISJueZ82aNXrqqad0xhln6Gc/+5n27dun9957T1OmTNG6des0cOBA133z8/N1xhlnaO/evRo/frxGjx6tvLw83XrrrTrrrLPaNU4zZ87Uhx9+qJdffllXXHHFUbc3zUbdcMMNrhCYm5urkSNHqqysTNOnT9cll1yimpoa7d69W6+//rp+/vOfq0ePHsd83QsuuEBJSUlatGiR9u3bd9Rs3caNG7Vq1SoNGDBAEyZMkCTNmjVLW7du1ejRo3XuueeqpqZGK1as0MMPP6wlS5bos88+U1BQULvef5NDhw5p9OjR2rVrl8aOHauxY8cqNzdXN998c6tj+tJLL2nOnDmaMGGCpk6dKofDobVr1+qZZ57Rxx9/rFWrVikqKkqSdO211yo2Nlbz5s3TBRdc0Gz56Y+Xpv7YP//5T91yyy2KiIjQZZddpqSkJC1ZskSzZs3SggULtGLFihaf4+qrr9by5ct1zjnnKDo6Wh999JGeeuop5efn65VXXunQOB3LBx98oEsuuUSmaerSSy9Venq61q5dq7///e+aN2+evvzyS2VmZkpy7rsbM2aMdu7cqTPPPFPnnXeeTNPU3r17NW/ePF166aXq27evJOeeuj/84Q/KzMzU5ZdfrpiYGOXm5mr16tV65513WvzcAkAzJgDAJT093ZRk7t69u9n1EyZMMCWZw4YNMwsKCo56XElJSYvX79+/30xJSTEHDRp01G2SzAkTJjS77qGHHjIlmVFRUeb333/f7LarrrrKlGTOnj27xdqOtHjxYlOSKcl85ZVXmt32j3/8w5Rk3nLLLc2uv/76601J5j333NPs+nXr1pnBwcGmJPOhhx466n20xG63m7169TINwzB37drV7Lba2lozMTHRtFqtZm5uruv6v/zlL6Yk87nnnjvq+SoqKsyqqqo2vfY999zTaq2//OUvTUnm008/7bpu586dpsPhOOq+DzzwgCnJfOutt5pd3/Q9Wrx4cbPrW/p+zpw505Rk3n777c2uX716tWm1Wlusc8+ePabdbj+qnpdfftmUZD755JPNrn/llVda/D4fq949e/aYwcHBZlRUlLlly5Zm97/llltMSebMmTObXd/0OTvllFPMwsJC1/UVFRVmv379TIvF0uz7eSxNNR3v81ReXm7Gx8ebFovFXLZsWbPbnnzySVOSeeaZZ7qumz9/fovjbZrOz11ZWZnrcnx8vJmammpWVlYedd+W/l8GgB9jaR8AtMNjjz3W4n6TmJiYFq9PS0vTpZdeqq1bt2rfvn1tfp1f/vKXGjZsWLPrZs6cKUn65ptv2vw8Y8aMcS0/bHL99dfLarU2e566ujq9+eabiomJ0QMPPNDs/sOHD9f/+3//r82vKUlBQUG6/vrrZZqm/vWvfzW7bd68eSooKNB5552n5OTkox4bFhZ21HUREREtXt+SmTNnyjAMvfLKK3I4HK7ra2tr9d///lfBwcHNxqRv374tto+/4447JEmffPJJm173x+rr6/XGG28oKirqqCWRI0aM0DXXXNPi49LT01ucAbv++usVHR3d4XqO9N///ld1dXX6+c9/ftTerscff1xRUVF6/fXXVVtbe9RjZ82a1WxvWUREhK655ho5HA6tWbOm07Udad68eSoqKtIVV1yhcePGNbvtrrvuUkZGhj799NOj/t9q6bMSHBzsmslrYrPZWhxr9pQBaAuCFAC0w6hRo1q9bcWKFbr88svVu3dvhYSEuPaq/PWvf5WkFvc3tWbEiBFHXde7d29JOmqfTHufx2azqWfPns2e54cfflB1dbVOPPHEo37YlKSxY8e2+TWb/OxnP5PFYtErr7yihoYG1/VNe5SagmGT888/X5GRkbrtttt0ySWX6MUXX9SmTZtkmma7XjcrK0uTJk3Svn37moWO9957T0VFRbrwwgub/aBcWVmpJ554QiNHjlRMTIwsFosMw3AtI2zP9+1IW7duVVVVlU466STFxMQcdXtrZ07V19fr+eef19ixYxUfH6+goCAZhiGLxaKysrIO13Okb7/9VpI0efLko26Li4vTySefrJqaGm3duvWo27vqs9nZOq1Wq8aPHy9J+u677yRJEyZMUGpqqp588klNmzZNf/nLX7R27dpmn78m11xzjfbs2aMTTjhBv/3tb7Vw4cIW94YBQGvYIwUA7dDSDIokzZkzR5deeqlCQ0N15plnql+/foqIiJDFYtGSJUu0dOnSFn+735qW9qZYrc6/slv6obA9z9P0XEc+T9MPkD179mzx/q1dfyzp6ek688wz9cknn+ijjz7Seeedpz179uizzz5Tenq6zj777KPu/8033+jhhx/WwoUL9f7770ty/pD+61//Wr/85S/b/NozZ87UF198oZdfflnnnHOOpJa7BNbX12vy5Mn65ptvNHToUF1xxRVKTEx0tU5/5JFH2vV9O9LxxrS1z9IVV1yhOXPmqG/fvrrggguUnJzs2hP33HPPdbielmpLSUlp8fam60tKSo66ras+m23R3jqjo6P19ddf66GHHtL8+fNdQTohIUG33nqrHnjgAdf39tlnn1Xfvn31yiuv6Mknn9STTz4pq9Wq6dOn609/+pOysrK69L0A8D8EKQBoh5aWgEnSgw8+qODgYK1Zs0aDBw9udttNN92kpUuXuqO8DouOjpbk7HjWktauP54bb7xRn3zyiV566SWdd955+te//iXTNHXDDTfIYjl6UcTgwYM1e/Zs2e12rV+/Xp999pn++te/6le/+pUiIiJ0ww03tOl1L774YiUkJGjBggU6ePCgysvLtWTJEmVlZTWb3Zg3b56++eYbXXvttUc1SsjNzdUjjzzSofctyTUL1drY5eXlHXXdmjVrNGfOHE2dOlUff/yxK6BIzsOOn3rqqQ7X01JteXl5GjJkyFG3N3XDa2kmzZ2OrLMlLdWZlpbm+pxt3rxZX3zxhV544QU9+uijcjgceuyxxyQ5l5/efvvtuv3225Wfn68vv/xSb731lt555x1t2rRJmzZtanNTFwCBiaV9ANAFduzYoRNOOOGoEOVwOPTll196qKq2GzRokMLCwvT999+rvLz8qNs7+h7OP/98JScn66OPPtL+/fv1yiuvuPZPHYvVatWpp56qe++9V2+++aYkae7cuW1+3eDgYP30pz9VfX29/vOf/7h+sP7Zz37WLAzv2LFDklo8c6qz4XfQoEEKDw/XunXrWlwy1lI78qZ6zj///GYhSnLujauurj7qMU17fNozG3TyySe3WkNJSYnWrVun0NDQoz7P7nasOu12u5YvXy5JOuWUU4663TAMDRkyRL/4xS/06aefSmr9M5SUlKSLL75Yb7/9tiZPnqydO3dq48aNXfMmAPgtghQAdIGMjAxt375dBw4ccF1nmqYefvhhbd682YOVtU1wcLCuuOIKlZaW6ve//32z29avX6/XXnutQ89rtVp17bXXqqGhQddcc41ycnI0ffp0paamHnXftWvXthg4mmZ0wsPD2/XaTXuwXnrpJb366quy2WxHNd5oOnT5xz+o79q1S/fee2+7Xu/HbDabrrnmGpWXlx/VbGLNmjV64403jnpMa/Xk5+frtttua/F1mvZytaeZyU9+8hPZbDb99a9/dYW3Jg8++KDKysr0k5/8xOMzMhdeeKHi4+P15ptv6uuvv25223PPPafdu3dr6tSprjb3mzZtanEG8MefodraWq1YseKo+9XX16uoqKjZfQGgNSztA4AucMcdd+jmm2/WySefrEsuuUQ2m00rVqzQ5s2bdd5552nBggWeLvG4nnzySX3xxRd66qmntGrVKo0ePVq5ubl6++23NX36dM2dO7fF5XjHM3PmTM2aNcs1e3DkHqUjvf766/rnP/+psWPHql+/foqLi9POnTu1YMEChYSE6Pbbb2/X6w4cOFDjx4/XsmXLJDnPAPvxfqXzzjtPWVlZeuaZZ7RhwwadfPLJ2rdvnz744AOde+657QonLXniiSf0+eef67nnntOaNWtc50jNnj1b06dP1/z585vdf+TIkRozZozef/99jR49WmPHjtXBgwf18ccfa+DAgerVq9dRr3HGGWcoPDxczz33nAoLC117r37xi1+0ujQvIyNDzz33nG677Tadcsopuvzyy5WYmKilS5fqq6++0qBBgzRr1qxOvfe2mDt3ruvg4B8766yzdPXVV+vf//63LrvsMk2YMEGXXXaZ+vTpo7Vr12rRokVKTk7WP//5T9djPv30U919990644wzNGDAACUlJSk7O1vz5s2TxWLR3XffLUmqrq7W2LFjlZWVpVNPPVXp6emqqanRp59+qi1btuj888/3+GwcAO9HkAKALnDTTTcpJCREzz33nP7zn/8oLCxM48aN0yuvvKL33nvPJ4JUz549tXLlSt1333366KOPtGrVKg0cOFB/+9vfFBERoblz57r2UrVH3759NWXKFH322WdKS0tzNX/4sauuukq1tbVauXKl1q5dq+rqaqWmpurKK6/UXXfdpaFDh7b7tW+88UZXkGopwEVEROiLL77Qb37zGy1ZskTLly9X37599eCDD+rOO+/U7Nmz2/2aR0pISNCKFSt03333acGCBVqzZo0GDhyov//978rIyDgqSAUFBWn+/Pl64IEH9NFHH+kvf/mLUlNT9bOf/UwPPPCATjjhhKNeIy4uTu+9954eeeQRvfrqq6qsrJTknHU61h6nW2+9VVlZWfrjH/+o9957T1VVVerdu7fuvvtu3Xfffcc90LcrrF+/XuvXr2/xttjYWF199dW64IILtGLFCj3xxBP65JNPVFpaquTkZN1888168MEHm4XLs88+W/v27dOyZcs0b948lZWVKSUlRWeeeabuvPNOjR49WpLz+z5r1iwtXrxYK1eu1Ny5cxUVFaV+/frp73//+3GXngKAJBlme/vKAgACzv33368nnnhCCxcuPKrbHgAAgYggBQBwOXDgwFHLxzZs2KDRo0crODhYOTk5Cg0N9VB1AAB4D5b2AQBcRowYoaysLA0dOlQRERHavn27PvzwQzkcDv3zn/8kRAEA0IgZKQCAyyOPPOJqAFBeXq7Y2Fidfvrp+vWvf62JEyd6ujwAALwGQQoAAAAA2olzpAAAAACgnQhSAAAAANBOBCkAAAAAaCeCFAAAAAC0E+3PGxUXF8tut3u6DElSYmKiCgoKPF1GwGC83Y8xdz/G3L0Yb/djzN2PMXcvxts9rFar4uLi2nbfbq7FZ9jtdtXX13u6DBmGIclZDw0Vux/j7X6Mufsx5u7FeLsfY+5+jLl7Md7eiaV9AAAAANBOBCkAAAAAaCeCFAAAAAC0E0EKAAAAANqJZhMAAABAG9TW1qq2ttYjr11dXa26ujqPvLa/MQxDkZGRriYeHUWQAgAAAI6jsrJShmEoKiqq0z+Ad4TNZvOKDtP+oK6uThUVFYqKiurU87C0DwAAADgOu92u8PBwj4QodK3g4OAuaSNPkAIAAACOgwCFHyNIAQAAAEA7EaQAAAAAoJ0IUgAAAACO67TTTtNLL73UJc+1cuVKpaamqrS0tEuezxPo2gcAAAD4qUsvvVQnnHCCHn300U4/10cffaTw8PAuqMo/eGWQWrhwoRYsWKCSkhKlp6fr+uuvV1ZWVov3XbJkif72t781u85ms+mNN95wR6kAAACAzzJNUw0NDbJajx8LevTo4YaKfIfXLe1buXKlXnvtNV166aWaNWuW0tPT9fjjjx9z2i8sLEwvvvii678XXnjBjRUDAAAg0JimKbO2xjP/tbF19+23366vvvpK//rXv5SamqrU1FTNnj1bqamp+uKLLzRt2jRlZmbqm2++0Z49e3Tddddp+PDh6t+/v6ZPn65ly5Y1e74fL+1LTU3V//73P91www3q16+fxowZo0WLFnV4TD/88ENNmjRJmZmZOu200/SPf/yj2e2vvvqqxowZo759+2r48OGaOXOm67YPPvhAU6ZMUb9+/TRkyBBdccUVqqqq6nAtbeF1M1JNgzBp0iRJ0syZM/Xtt99q8eLFuvDCC1t8jGEYio2NdV+RAAAACGx1tXL8/HK3vVztEV9bnn9bCgk97mMeffRR7dq1S4MGDdKvf/1rSdIPP/wgSXriiSf0u9/9Tn369FFMTIwOHDigyZMn695771VwcLDeffddXXfddVq2bJlSU1NbfY1nnnlGDzzwgB544AG98sor+vnPf65Vq1YpLi6uXe/v+++/180336w777xT559/vtasWaP77rtPcXFxuuKKK7R+/Xr97ne/01/+8heNGDFCJSUlWrVqlSTp4MGDuu2223T//ffrnHPOUUVFhVatWtUlZ0Udi1cFKbvdrl27djULTBaLRcOGDdO2bdtafVxNTY1uvfVWmaapzMxMXXXVVerdu3eL962vr292KrRhGAoLC3N97WlNNXhDLYGA8XY/xtz9GHP3YrzdjzF3P8bcN0RHRys4OFihoaFKSkqSJO3YsUOSdPfdd2v8+PGu+8bFxWnIkCGuy/fcc48WLlyoRYsW6brrrmv1NS6//HLXz+6/+c1v9K9//Uvr1q1zTYq01YsvvqixY8fqjjvukCT169dP27dv1z/+8Q9dccUVysnJUXh4uKZOnarIyEilpaVp6NChkqT8/HzZ7XZNnz5daWlpkqTBgwcf9zU7+/n1qiBVVlYmh8Nx1OxSbGysDhw40OJjevXqpVtuuUXp6emqqqrS/Pnz9cADD+iZZ55pcR3nnDlz9O6777ouZ2ZmatasWUpMTOzS99JZycnJni4hoDDe7seYux9j7l6Mt/sx5u4XSGNeXV0tm83mumxardI/53immOCQNocAwzAUFBTkqr1pL9Spp57a7P1UVFTo6aef1meffaaDBw/KbrerpqZGubm5rvv9+LkkadiwYa7LMTExioqKUnFxcbP7tKSpDpvNJpvNph07dmjatGnNHnf66afr5ZdflsVi0ZQpU5SWlqbRo0dr0qRJmjx5sqZPn67w8HANHz5c48aNc61qmzhxos4777xjrlgLDg5WSkpKm8aw1ffQqUd7gQEDBmjAgAHNLt9xxx369NNPdeWVVx51/4suukgzZsxwXW76EBYUFMhut3d/wcdhGIaSk5OVl5fX7dORYLw9gTF3P8bcvRhv92PM3S8Qx7yurq7ZqiZJkiXIba9vs9kOv347fmZtaibR9Nimn3ebPZ+k3/3ud1q+fLkefPBBZWRkKDQ0VDfeeKNqampc9/vxc0nOz0JLl48aqx9pqqPpvqZpyuFwNHtcQ0OD6z4hISFauHChVq5cqWXLlmnWrFl66qmn9NFHHykmJkZvvvmm1qxZo6VLl+rll1/WE088oQ8++EB9+vRp8fXr6uqUm5t71PVWq7XNEyxeFaSio6NlsVhUUlLS7PqSkpI274GyWq3KzMxUXl5ei7c3pd6WeNNfBKZpelU9/o7xdj/G3P0Yc/divN2PMXc/xtz72Ww2ORyO495vzZo1uuyyy3TOOedIkiorK5Wdnd3d5bn0799fq1evbnbd6tWr1bdvXwUFOQOr1WrV+PHjNX78eN15550aPHiwVqxYoenTp8swDI0cOVIjR47UHXfcoVGjRunjjz/WTTfd1Oprdvaz61VBymq1qm/fvtq4caNGjRolSXI4HNq4caOmTZvWpudwOBzat2+fTj755O4sFQAAAPB6vXv31nfffaf9+/crIiKi1VCVmZmpjz/+WGeeeaYMw9DTTz/dpgDWVW666SZNnz5dzz77rM4//3ytXbtWr7zyip544glJ0qeffqp9+/bptNNOU2xsrD7//HM5HA7169dP3377rb788ktNmDBBCQkJ+vbbb1VUVKT+/ft3a81eFaQkacaMGXrhhRfUt29fZWVl6aOPPlJtba0mTpwoSXr++ecVHx+vq6++WpL07rvvqn///kpOTlZlZaXmz5+vgoICTZkyxYPvAgAAAPC8m266SbfffrsmTpyompoaPfPMMy3e76GHHtKdd96pCy64QPHx8brttttUUVHhtjqHDRumf/zjH/rjH/+oP//5z0pKStLdd9+tK664QpJz/9XHH3+sZ555RjU1NcrMzNQLL7yggQMHavv27Vq1apVefvllVVRUKDU1Vb/73e80efLkbq3ZML1wPnbhwoWaP3++SkpKlJGRoeuuu86VKB9++GElJibqtttuk+TsJ//NN9+opKREERER6tu3r6688kplZma26zULCgqOu5bTHQzDUEpKinJzc5kqdwPG2/0Yc/djzN2L8XY/xtz9AnHMy8rKFB0d7bHX//GeJnROa99Pm83mm3ukmkybNq3VpXwPP/xws8vXXnutrr322u4vCgAAAAAaeWWQClRmXrbMnH2qG3KiFBbl6XIAAACADrn33nv1/vvvt3jbxRdfrFmzZrm5oq5HkPIi5uKPZH7xgaouu1Y6+xJPlwMAAAB0yN13362bb765xduiovxjwoAg5U3inAcINxw66OFCAAAAgI5LSEhQQkKCp8voVhZPF4AjxDYFqXwPFwIAAADgWAhSXsSIc6b2hkKCFAAAAODNCFLeJO7wjFSgtBIFAAAAfBFByps0BimzrlaqdN8BaAAAAADahyDlRQxbsBTZeDBY8SHPFgMAAACgVQQpb9M0K1Vc6OFCAAAAEOhOO+00vfTSS226b2pqqhYuXNjNFXkPgpSXMRqDFDNSAAAAgPciSHmbxs59ZgkzUgAAAIC3Ikh5maYW6GJpHwAAgNcyTVM1dof7/qs//HVbuzv/97//1SmnnCKHw9Hs+uuuu0533nmn9uzZo+uuu07Dhw9X//79NX36dC1btqzLxmjLli267LLL1K9fPw0ZMkT33HOPKisrXbevXLlS5557rrKysjR48GBdcMEFys7OliRt2rRJl156qQYMGKCBAwdq2rRpWr9+fZfV1hWsni4AP8LSPgAAAK9X22DqitnbPPLas68YoFCrcdz7zZgxQw8++KBWrFihcePGSZKKi4u1ZMkSvfbaa6qsrNTkyZN17733Kjg4WO+++66uu+46LVu2TKmpqZ2qsaqqStdcc41OPfVUffjhhzp06JDuvvtu3X///Xruuedkt9t1ww036Oqrr9YLL7yg+vp6fffddzIM5/v6xS9+oSFDhujJJ5+UxWLRpk2bZLV6V3TxrmogxTY1myBIAQAAoONiY2M1adIkzZ071xWkPvzwQ8XHx2vMmDGyWCwaMmSI6/733HOPFi5cqEWLFum6667r1GvPmTNHtbW1+vOf/6zw8HBJ0u9//3tde+21uv/++2W1WlVWVqapU6cqIyNDktS/f3/X43NycnTzzTcrKytLktS3b99O1dMdCFJehqV9AAAA3i8kyNDsKwa47fVsVpvq7fWu126riy66SPfcc4+eeOIJhYSEaM6cOTr//PNlsVhUWVmpP/3pT/r888+Vn58vu92umpoa5eTkdLre7du3a/Dgwa4QJUkjR46Uw+HQzp07dfrpp+vyyy/XNddco3HjxmncuHE677zz1LNnT0nSjTfeqLvvvlvvvfeexo0bpxkzZrgCl7dgj5S3iW9c2lddJbOmyrO1AAAAoEWGYSjUanHff7bDXzctf2uLM888U6Zp6vPPP1dOTo5WrVqliy++WJL06KOPauHChfrNb36j999/X4sWLdKgQYNUV1fXXcPWzLPPPqv58+drxIgRmj9/vsaNG6e1a9dKku666y598cUXmjJlilasWKFJkybp448/dktdbUWQ8jJGaLiM8AjnheIizxYDAAAAnxYaGqpzzjlHc+bM0bx589SvXz8NGzZMkrRmzRpddtllOuecczR48GAlJSW5mj10Vv/+/bVlyxZVVR2eGFi9erUsFov69evnum7o0KH6xS9+ofnz52vgwIGaO3eu67Z+/frpxhtv1JtvvqlzzjlHs2fP7pLaugpBygsFJTinNGk4AQAAgM666KKL9Pnnn+utt97SRRdd5Lo+MzNTH3/8sTZu3KhNmzbptttuO6rDX0ddfPHFCgkJ0a9+9Stt3bpVK1as0IMPPqhLLrlEiYmJ2rdvn/7whz9ozZo1ys7O1tKlS7V7925lZWWpurpa999/v1auXKns7GytXr1a69evb7aHyhuwR8oLBfVIlH3fLpnFhWr7xC0AAABwtLFjxyo2NlY7d+5sFqQeeugh3XnnnbrgggsUHx+v2267TRUVFV3ymmFhYXrjjTf0u9/9Tueee65CQ0N17rnn6qGHHnLdvmPHDr3zzjsqLi5WUlKSrr32Wv3f//2f7Ha7iouL9atf/UqHDh1SfHy8zjnnHN11111dUltXMcy2NqL3cwUFBaqvr/d0GTIMQyGzX1TlpwtkXHCNLDOu8HRJfs0wDKWkpCg3N7fNZzKgcxhz92PM3Yvxdj/G3P0CcczLysoUHR3tsde32Wxe8bOqv2jt+2mz2ZSYmNim52BpnxcK6tG4tK+Ezn0AAACAN2JpnxcKSkiSJJm0QAcAAIAXeP/993Xvvfe2eFtaWpoWL17s5oo8jyDlhZqCFM0mAAAA4A3OOussnXzyyS3eZrPZ3FyNdyBIeaGgHk1BihkpAAAAeF5kZKQiIyM9XYZXYY+UF3LNSFWUyax3z4FoAAAAANqOIOWFLFExki3YeaGEQ3kBAAC8QVedsQTP6qpOkwQpL2QYhhTXw3mBfVIAAAAeFx4ervLycsKUH6iqqlJISEinn4c9Ut4qtoeUn8uhvAAAAF7AarUqIiKiyw6sba/g4GDV1bHlo7NM05TVaiVI+TMjLkGmxIwUAACAl7BarR45lDcQD0D2BSzt81aupX107gMAAAC8DUHKSxnxCZIks4QgBQAAAHgbgpS3YkYKAAAA8FoEKS9lxNK1DwAAAPBWBClvFedc2qfSEpkNDZ6tBQAAAEAzBClvFR0jBQVJpkMqLfZ0NQAAAACOQJDyUoYlSIqJd15geR8AAADgVQhS3qyp4QSd+wAAAACvQpDyYk0NJ0xmpAAAAACvQpDyZk0NJ2iBDgAAAHgVgpQ34ywpAAAAwCsRpLxZ44yUSZACAAAAvApByosZcRzKCwAAAHgjgpQ3c3XtK5LpcHi2FgAAAAAuBClvFhMnGYbUYJcqSj1dDQAAAIBGBCkvZlhtUnSs80JxkUdrAQAAAHAYQcrbxbJPCgAAAPA2BClvR+c+AAAAwOsQpLycERfv/IIZKQAAAMBrEKS8XeOMFIfyAgAAAN6DIOXtGlugmyUEKQAAAMBbEKS8nMGMFAAAAOB1CFLeLu5w1z7TND1bCwAAAABJBCnv19T+vK5Wqqr0bC0AAAAAJBGkvJ4RHCJFRDkv0LkPAAAA8AoEKV/QtLyPhhMAAACAVyBI+QIO5QUAAAC8CkHKBxhHNJwAAAAA4HkEKV/Q1HCCGSkAAADAKxCkfEHTobzMSAEAAABegSDlA1yH8pYUebYQAAAAAJIIUr6BPVIAAACAVyFI+YKmPVJVlTJrqj1bCwAAAACClE8IC5dCwpxfc5YUAAAA4HEEKR9gGMYRy/sIUgAAAICnEaR8hatzH0EKAAAA8DSClI9wde6j4QQAAADgcQQpX9HUcII9UgAAAIDHEaR8BUv7AAAAAK9BkPIRh5f2EaQAAAAATyNI+QoO5QUAAAC8BkHKVzTNSJWXyqyv92wtAAAAQIAjSPmKyCjJanN+TcMJAAAAwKMIUj6CQ3kBAAAA70GQ8iVNnfuYkQIAAAA8iiDlQ4xYOvcBAAAA3oAg5Uvo3AcAAAB4BYKUL+FQXgAAAMArEKR8iMGMFAAAAOAVCFK+pOksqZIiz9YBAAAABDiClC9pmpEqLZLZ0ODZWgAAAIAARpDyJdGxksUiORxSWYmnqwEAAAACFkHKhxiWICkm3nmBfVIAAACAxxCkfI2r4QSd+wAAAABPIUj5mqYW6CUEKQAAAMBTCFI+xmjq3MfSPgAAAMBjCFK+hqV9AAAAgMcRpHxNbOPSPmakAAAAAI8hSPkYg0N5AQAAAI8jSPka19K+QzJN07O1AAAAAAGKIOVrYhvPkbLbpYoyz9YCAAAABCiClI8xrDYpOtZ5gX1SAAAAgEcQpHxRLJ37AAAAAE8iSPmipkN5CVIAAACARxCkfNDhQ3kJUgAAAIAnEKR80RGd+wAAAAC4H0HKFzUdylvCjBQAAADgCQQpH2QwIwUAAAB4FEHKFx2xR4pDeQEAAAD3I0j5oqYZqdoaqbrKs7UAAAAAAYgg5YOMkFApPNJ5gc59AAAAgNsRpHwV+6QAAAAAjyFI+SrXobwEKQAAAMDdCFI+ynUob0mRZwsBAAAAAhBBylfFsrQPAAAA8BSClK9yLe2j2QQAAADgbgQpH8WhvAAAAIDnEKR81RGH8gIAAABwL4KUr2qakaqqkFlb69laAAAAgABDkPJVYRFSSKjz6xJmpQAAAAB3Ikj5KMMwOJQXAAAA8BCClC+LpXMfAAAA4AkEKR9G5z4AAADAMwhSvqypcx97pAAAAAC3Ikj5Mg7lBQAAADyCIOXDDM6SAgAAADyCIOXLYtkjBQAAAHgCQcqXNTWbKC+Vaa/3bC0AAABAACFI+bLIaMlqlUxTKi32dDUAAABAwCBI+TDDYmF5HwAAAOABBClfR+c+AAAAwO0IUj7OYEYKAAAAcDuClK9ztUAv8mwdAAAAQAAhSPm6OGakAAAAAHezerqAlixcuFALFixQSUmJ0tPTdf311ysrK+u4j1uxYoX+/Oc/a8SIEbrnnnvcUKnnGXEJMiWZJeyRAgAAANzF62akVq5cqddee02XXnqpZs2apfT0dD3++OMqLS095uPy8/P1+uuva/DgwW6q1EswIwUAAAC4ndcFqQ8++EBTpkzRpEmTlJaWppkzZyo4OFiLFy9u9TEOh0N//etfdfnllyspKcmN1XqBpmYTJUUyHQ2erQUAAAAIEF61tM9ut2vXrl268MILXddZLBYNGzZM27Zta/Vx7777rqKjozV58mRt2bLlmK9RX1+v+vp612XDMBQWFub62tOaamhzLbHxkmGRHA4Z5WUyYuO7sTr/0+7xRqcx5u7HmLsX4+1+jLn7MebuxXh7J68KUmVlZXI4HIqNjW12fWxsrA4cONDiY7Zu3aovvvhCTz31VJteY86cOXr33XddlzMzMzVr1iwlJiZ2uO7ukJyc3Ob7HohPUENhvnoESSEpKd1Ylf9qz3ijazDm7seYuxfj7X6Mufsx5u7FeHsXrwpS7VVdXa2//vWvuummmxQdHd2mx1x00UWaMWOG63JTsi8oKJDdbu+WOtvDMAwlJycrLy9Ppmm26TENMXFSYb4Obf9BlihmpNqjI+ONzmHM3Y8xdy/G2/0Yc/djzN2L8XYfq9Xa5gkWrwpS0dHRslgsKikpaXZ9SUnJUbNUknTw4EEVFBRo1qxZruuaPlxXXnmlnnvuuaOSu81mk81ma/H1vemDaZpm2+tpXM5nFh3yqvfgS9o13ugSjLn7MebuxXi7H2Pufoy5ezHe3sWrgpTValXfvn21ceNGjRo1SpKzkcTGjRs1bdq0o+7fq1cv/fGPf2x23VtvvaWamhpde+21SkhIcEvdntbUAp3OfQAAAIB7eFWQkqQZM2bohRdeUN++fZWVlaWPPvpItbW1mjhxoiTp+eefV3x8vK6++moFBwerT58+zR4fEREhSUdd79eaWqBzlhQAAADgFl4XpEaPHq2ysjK9/fbbKikpUUZGhu677z7X0r5Dhw7RseTHGlugm8UEKQAAAMAdvC5ISdK0adNaXMonSQ8//PAxH3vbbbd1Q0XejaV9AAAAgHt53YG86ICmpX3FhWxABAAAANyAIOUPmg7htddLFeWerQUAAAAIAAQpP2DYgqWoGOcFGk4AAAAA3Y4g5S9cy/vYJwUAAAB0N4KUv4hznplF5z4AAACg+xGk/ITRtE+KGSkAAACg2xGk/EXjjJSYkQIAAAC6HUHKXzTukTJpNgEAAAB0O4KUnzCYkQIAAADchiDlL+jaBwAAALgNQcpfxDYGqZpqmdVVnq0FAAAA8HMEKT9hhIZJYRHOC+yTAgAAALoVQcqfsLwPAAAAcAuClD9p6txHwwkAAACgWxGk/Mjhzn3MSAEAAADdiSDlT5oaTjAjBQAAAHQrgpQ/YWkfAAAA4BYEKT/CobwAAACAexCk/ElT174S9kgBAAAA3Ykg5U+aZqQqymXW1Xq2FgAAAMCPEaT8SXiEFBzs/JpDeQEAAIBuQ5DyI4ZhSLFN+6SKPFsMAAAA4McIUv7G1bmPfVIAAABAdyFI+Rk69wEAAADdjyDlb5o69zEjBQAAAHQbgpS/4VBeAAAAoNsRpPyM4TpLiiAFAAAAdBeClL9hjxQAAADQ7QhS/qZpRqqsWKbd7tlaAAAAAD9FkPI3kTFSkFUyTam02NPVAAAAAH6JIOVnDItFio13XqBzHwAAANAtCFL+iIYTAAAAQLciSPmhpkN5aYEOAAAAdA+ClD/iUF4AAACgWxGk/FFsU5BiRgoAAADoDgQpP9R0KK/JHikAAACgWxCk/BGH8gIAAADdiiDlj47o2mc6HJ6tBQAAAPBDBCl/FB0nGRapoUEqL/V0NQAAAIDfIUj5IcNqlaJjnRfo3AcAAAB0OYKUv+JQXgAAAKDbEKT8VVPnPhpOAAAAAF2OIOWnDFfnPpb2AQAAAF2NIOWv4jiUFwAAAOguBCl/FcvSPgAAAKC7EKT8lMGhvAAAAEC3IUj5K1fXvkMyTdOztQAAAAB+hiDlr5qCVF2dVFXh2VoAAAAAP0OQ8lOGLViKjHZeoHMfAAAA0KUIUv4sls59AAAAQHcgSPkzDuUFAAAAugVByo/RuQ8AAADoHgQpf+Y6lJc9UgAAAEBXIkj5s8YZKZb2AQAAAF2LIOXHjLh45xfMSAEAAABdiiDlz5r2SJUUebYOAAAAwM8QpPxZ0x6p6kqZNVWerQUAAADwIwQpP2aEhkth4c4LxcxKAQAAAF2FIOXvYuncBwAAAHQ1gpS/41BeAAAAoMsRpPyc0bRPqoQgBQAAAHQVgpS/a+rcx9I+AAAAoMsQpPwdS/sAAACALkeQ8nMGzSYAAACALkeQ8nccygsAAAB0OYKUv2tqNlFeKrO+zrO1AAAAAH6CIOXvIqIkW7Dza2alAAAAgC5BkPJzhmEcnpVinxQAAADQJQhSgSCWzn0AAABAVyJIBQAO5QUAAAC6FkEqELgO5SVIAQAAAF2BIBUIXIfyskcKAAAA6AoEqQBgMCMFAAAAdCmCVCCIpWsfAAAA0JUIUoGgqdlEaYnMhgbP1gIAAAD4AYJUIIiOkYKCJNMhlRZ7uhoAAADA5xGkAoBhCZJi4p0XWN4HAAAAdBpBKlBwlhQAAADQZQhSAcKIpQU6AAAA0FUIUoHC1QK9yLN1AAAAAH6AIBUo4miBDgAAAHQVglSgaJyRMjmUFwAAAOg0glSAMJiRAgAAALoMQSpQuLr2Fcl0ODxbCwAAAODjCFKBIiZOMgypwS5VlHm6GgAAAMCnEaQChGG1SdGxzgvskwIAAAA6hSAVSGLZJwUAAAB0BYJUIKFzHwAAANAlCFIBxIiLd35RQpACAAAAOoMgFUgaZ6RY2gcAAAB0DkEqkDS2QGdpHwAAANA5BKkAYrhmpAhSAAAAQGcQpAJJ3OGufaZperYWAAAAwIcRpAJJU/vzulqputKztQAAAAA+jCAVQIzgECkiynmB5X0AAABAhxGkAk0ch/ICAAAAnUWQCjQcygsAAAB0GkEqwBixjYfyMiMFAAAAdBhBKtA0tUAvKfJsHQAAAIAPI0gFGtehvMxIAQAAAB1FkAowHMoLAAAAdB5BKtDQtQ8AAADoNIJUoGk6lLeqUmZtjWdrAQAAAHwUQSrQhIVLIWHOr1neBwAAAHQIQSrAGIbB8j4AAACgkwhSgcjVuY8ZKQAAAKAjCFIB6HDnPmakAAAAgI4gSAWipoYTJcxIAQAAAB1BkApELO0DAAAAOoUgFYA4lBcAAADoHIJUIKJrHwAAANApBKlA1DQjVV4qs77es7UAAAAAPoggFYgioySrzfk1DScAAACAdiNIBaBmh/KWFHm2GAAAAMAHEaQClatzH/ukAAAAgPYiSAUoI5bOfQAAAEBHEaQCFZ37AAAAgA4jSAWqpqV9NJsAAAAA2o0gFaAM14wUQQoAAABoL4JUoIpjjxQAAADQUQSpQNU0I1VaJLOhwbO1AAAAAD6GIBWoomMli0VyOKSyEk9XAwAAAPgUgpSXKaqqV3lNfbe/jmEJkmLinRdoOAEAAAC0C0HKi7y45qCue3+HPtiU554XpAU6AAAA0CFWTxfQkoULF2rBggUqKSlRenq6rr/+emVlZbV431WrVmnOnDnKy8tTQ0ODkpOTdd5552n8+PFurrrzUiJtMiV98UO+Jvbq1f0v2NQCvbhQRve/GgAAAOA3vC5IrVy5Uq+99ppmzpyp/v3768MPP9Tjjz+u5557TjExMUfdPzIyUhdffLF69eolq9Wqb7/9Vn/7298UHR2tk046yf1voBPO6BOll9fma31OqQqrEhUf1r3fHiMuQabEjBQAAADQTl63tO+DDz7QlClTNGnSJKWlpWnmzJkKDg7W4sWLW7z/kCFDNGrUKKWlpSk5OVnTp09Xenq6tm7d6ubKOy8h3KZBCWEyJX21v7z7X5CzpAAAAIAO8aoZKbvdrl27dunCCy90XWexWDRs2DBt27btuI83TVMbN27UgQMHdM0117R4n/r6etXXH27mYBiGwsLCXF972pj0aG09VK0Ve8s1Y2B8t75W04yUWVLoFe/dE5red6C+f09gzN2PMXcvxtv9GHP3Y8zdi/H2Tl4VpMrKyuRwOBQbG9vs+tjYWB04cKDVx1VVVemmm26S3W6XxWLRDTfcoBNPPLHF+86ZM0fvvvuu63JmZqZmzZqlxMTELnkPnXVhRJz+tfagNudXyRYVr4TIkG57rdqsgcqXFFRWopSUlG57HV+QnJzs6RICDmPufoy5ezHe7seYux9j7l6Mt3fxqiDVUaGhoXr66adVU1OjDRs26LXXXlPPnj01ZMiQo+570UUXacaMGa7LTcm+oKBAdrvdbTW3xjAMDUmJ1qbcMs1bs0PTu3FWynQ4/2w4dFAHDhwIyN9yGIah5ORk5eXlyTRNT5cTEBhz92PM3Yvxdj/G3P0Yc/divN3HarW2eYLFq4JUdHS0LBaLSkpKml1fUlJy1CzVkSwWiyuhZ2RkKCcnR3Pnzm0xSNlsNtlsthafx1s+mFMGJGlTbpm+3FeucwbEddvrmDGNz223yywvlaKObuYRKEzT9Jrvf6BgzN2PMXcvxtv9GHP3Y8zdi/H2Ll7VbMJqtapv377auHGj6zqHw6GNGzdqwIABbX4eh8PRbB+Ur5ky0JmCN+dXqaS6+2bJDKvtcHiicx8AAADQZl4VpCRpxowZ+vzzz7VkyRJlZ2fr5ZdfVm1trSZOnChJev755/W///3Pdf85c+bo+++/18GDB5Wdna0FCxZo+fLlGjdunIfeQef1iglTVnyoHKYbuvfFJTj/LC7q3tcBAAAA/IhXLe2TpNGjR6usrExvv/22SkpKlJGRofvuu8+1tO/QoUPN9vLU1tbq5ZdfVmFhoYKDg5Wamqpf/OIXGj16tIfeQdcYkx6tHUU1Wrm/e5f3Ka6HtG+nzOJDHMoLAAAAtFGngtShQ4d06NAhDRo0yHXdnj179MEHH6i+vl5jxozRqFGj2v2806ZN07Rp01q87eGHH252+corr9SVV17Z7tfwdmP6ROk/3+Vr48EqldbYFRPaPZn38KG8nCUFAAAAtFWnlvb9+9//1jvvvOO6XFJSokceeUSrVq3Sli1b9Kc//UmrVq3qdJGBKDkqWP3iQ+Qwpa/3V3TfC7kO5WWPFAAAANBWnQpSO3fu1LBhw1yXly1bprq6Oj399NP6xz/+oWHDhmnBggWdLjJQje4dLUlaua+s+14k1hmkzBJmpAAAAIC26lSQqqioUEzM4ZbZa9eu1QknnKDk5GRZLBaNGjVKOTk5nS4yUI1Jj5IkfX+wSmU13dO9z3DNSBGkAAAAgLbqVJCKjo5WQUGBJKmyslLbt2/X8OHDXbc7HA45HI7OVRjAUqKClRnnXN63Krublve5uvYd4lwCAAAAoI061cFg2LBh+vjjjxUeHq5NmzbJNM1mzSWys7PVo0ePThcZyEb3idLu4lqt2FeuM7Niu/4Fmmakamuk6iopPKLrXwMAAADwM52akbr66quVlpam119/Xd9//73+7//+T0lJSZKk+vp6ffXVVxo6dGiXFBqoxvRx7pP6Pq9S5bUNXf78RkioFB7pvMDyPgAAAKBNOjUjFRsbq8cee0xVVVUKDg6W1Xr46UzT1IMPPqiEhIROFxnIUqODlR4bor0ltVqVXa6p/WK7/kXiekhVFVJJoZTap+ufHwAAAPAznZqRahIeHt4sRElScHCwMjIyFBkZ2RUvEdDG9HE2nVi5r7x7XqBxeZ9JC3QAAACgTToVpDZs2KD58+c3u+6LL77QLbfcopkzZ+rVV1+l2UQXGN0YpNbnVaqirhuW97kaTrC0DwAAAGiLTgWpd955R3v27HFd3rdvn1566SVFR0frhBNO0Mcff3xU0EL79Y4JUZ+YYNkd0jfd0b0vlkN5AQAAgPboVJDKyclRv379XJeXLVumsLAwPfroo7rjjjs0ZcoULVu2rNNF4nDTiW45nNe1tI8ZKQAAAKAtOhWkampqFBYW5rq8bt06nXTSSQoJCZEkZWVluc6ZQuc0Le/7LrdKlV28vM91KG8JQQoAAABoi04FqYSEBO3cuVOSlJeXp/379+vEE0903V5RUSGbzda5CiFJ6hMborToYNkdplbndPHyPvZIAQAAAO3SqfbnY8eO1bvvvquioiJlZ2crIiJCI0eOdN2+a9cupaSkdLpIOI3uE6W3NxZqxb5yTcyM6bonbpqRqiyXWVsro3FGEQAAAEDLOjUjdfHFF+vCCy9UYWGhEhISdPfddysiIkKSczZq06ZNGjFiRJcUisNt0L87UKmq+i5c3hcWIYWEOr9meR8AAABwXJ2akQoKCtJVV12lq6666qjbIiMj9dJLL3Xm6fEj6bEh6hUVrAPldVqdXaEJXTQrZRiGc1YqL8fZua9nry55XgAAAMBfdcmBvJKz8UR2drays7NVU1PTVU+LIxiGcfhw3v1dfDhvYwt0kxkpAAAA4Lg6NSMlSTt27NAbb7yhrVu3ug7ftVgsGjRokH7yk580a4+OzhvdJ0rvbCrUtwcqVV3vUJita7KwEddDpkTDCQAAAKANOhWktm/frocfflhWq1WTJ09WamqqJOf5UitWrNBDDz2khx9+WFlZWV1SLKTMuBClRNmUW16v1TkVGp8R3TVP7Orcx6G8AAAAwPF0Kki99dZbio+P12OPPabY2Nhmt1122WV68MEH9eabb+rBBx/szMvgCM7lfdF6d1OhVu4r78IgxaG8AAAAQFt1al3Y9u3bdeaZZx4VoiQpNjZWU6dO1fbt2zvzEmhB0+G8aw9UqMbu6JLnNDhLCgAAAGizTgUpwzDU0NB6G26Hw+HsCIcu1TcuRMmRNtU1mFrbVYfzNjaboP05AAAAcHydClIDBw7UJ598ooKCgqNuO3TokBYtWqRBgwZ15iXQAsMwXLNSK/Z1Ufe+pkN5y0pk2uu75jkBAAAAP9WpPVJXXXWVHnroId1+++0aNWqUUlJSJEkHDhzQmjVrZLFYWjxjCp03uk+U3t9cpDU5Faq1OxRi7WT3vshoyWqV7HaptFjqkdQ1hQIAAAB+qFNBKjMzU0888YTefPNNrVmzRnV1dZKk4OBgnXTSSbrssssUFRXVJYWiuaz4UCVF2JRfWa+1Byo0uk/nmk4YFotzed+hg87OfQQpAAAAoFWdPkcqLS1Nd999txwOh8rKyiRJ0dHRslgsev/99zV79mzNnj2704WiuablfXO3FGnlvvJOBylJzuV9hw7KLC4UO9sAAACA1nXNaa5yHsIbGxur2NhYWSxd9rQ4hjGN+6RWNy7v6yyjqeEEnfsAAACAYyLx+LD+PUKVGG5Vjd3Ut7mVnX9CWqADAAAAbUKQ8mFHdu9b2RXd+5o69xUf6vxzAQAAAH6MIOXjmvZGrc6uUF1D55b3NR3Ka3KWFAAAAHBM7W42sWvXrjbft6ioqL1Pj3YakBCqHuFWFVbZ9V1upU5L60SXxNh455/MSAEAAADH1O4g9dvf/rY76kAHWQxDo3tHacEPxVq5t7xzQappj1RpsUxHgwxLUNcUCQAAAPiZdgepW265pTvqQCeM6eMMUt/kVKi+wSFbUAdXbMbESYZFamiQykoPz1ABAAAAaKbdQWrixIndUAY6Y2BimOLDrCqqtmtdbpVGpkV26HmMoCBnmCopdHbuI0gBAAAALaLZhB+wGIbOaOret7+sc09G5z4AAADguAhSfqLpcN5V2RWqbzA7/kSNQYrOfQAAAEDrCFJ+YlBCmOJCg1RZ59D6vI4fzmtwKC8AAABwXAQpPxFkOWJ5X2cO52VpHwAAAHBcBCk/Mtq1vK9cdkcHl/fFNi7tY0YKAAAAaBVByo+ckBiumNAgVdQ59H0Hl/cdXtrHjBQAAADQGoKUHwmyGDqjt3NWakVHl/c1Le0rKZJpdqJpBQAAAODHCFJ+xtW9b38Hl/c1nR1VXydVdmKvFQAAAODHCFJ+ZkhSuGJCglRe59DGg1XtfrxhC5aiYpwX2CcFAAAAtIgg5WeCLIZOdy3v6+DhvHTuAwAAAI6JIOWHmrr3fb2/Qg0dWd7X2HCCzn0AAABAywhSfmhYz3BFhQSprLZBG/M7sLyvaZ9UCUEKAAAAaAlByg8FWQydnhYpqYOH89ICHQAAADgmgpSfalre99X+8vYv74vjUF4AAADgWAhSfurE5AhFBVtUWtOgzQXtW953+FBeghQAAADQEoKUn7JaDJ3W1L1vbzuX99G1DwAAADgmgpQfG927qXtfO5f3xTYGqZpqmdXtb1YBAAAA+DuClB87MTlCEcEWFdc0aGtBdZsfZ4SGSWERzgt07gMAAACOQpDyY7YgQ6c1du9r9+G8LO8DAAAAWkWQ8nNj+kRLklbur5DDbMfyPjr3AQAAAK0iSPm54cnhCrdZVFxtb9/yPs6SAgAAAFpFkPJztiCLRnXkcN6mhhPFRd1QFQAAAODbCFIBYEzj4bwr95W3fXmfa2kfM1IAAADAjxGkAsBJKREKs1pUWG3XtkM1bXoMh/ICAAAArSNIBYDgI5b3tbl7X1PXvhJmpAAAAIAfI0gFiNFHLO8z27K8r2lGqqJcZl1tN1YGAAAA+B6CVIA4OSVCoVaLDlXZta2wDcv7wiOk4GDn1yU0nAAAAACORJAKECFWi0amRkhqW/c+wzCkWPZJAQAAAC0hSAUQ1+G8+8rauLyPzn0AAABASwhSAeSUXhEKCTKUX2nXjqLjL++jcx8AAADQMoJUAAmxWjQitR2H88bFO/8sIUgBAAAARyJIBZgx6c7ufSva0r2vcUaKpX0AAABAcwSpAHNqr0gFBxk6WFGvXcXHbmtuNJ0lxdI+AAAAoBmCVIAJPWJ534q9xzmclz1SAAAAQIsIUgFodO/Gw3n3H2d5X9OMVFmxTLvdDZUBAAAAvoEgFYBGpDqX9+WW12v3sZb3RcZIQVbJNKWyYvcVCAAAAHg5glQACrNZdEov5+G8K47Rvc+wWKTYxs59LO8DAAAAXAhSAarNh/O6Gk7QuQ8AAABoQpAKUCNSI2SzGDpQXq+9Ja0v7zNcLdCZkQIAAACaEKQCVLgtqE3L+5iRAgAAAI5GkApgo/s0du871uG8Sb0kSebmdcc/wBcAAAAIEASpADYyNVJWi6HssjrtL61r8T7GyLFSSJiUs1favM69BQIAAABeiiAVwCKCg3RyStPyvpYP5zXCI2WMnSpJcnw6112lAQAAAF6NIBXgjlze1xpjynmSYZE2fSczZ6+7SgMAAAC8FkEqwI1Ki5TVIu0rrdP+0pa79xmJydIpp0uSzE/nubM8AAAAwCsRpAJcZHCQTko+fvc+y5kXSpLMVUtklhW7ozQAAADAaxGk0Lblff0GSX0HSna7zMUfuas0AAAAwCsRpKDT0qIUZEh7S2qVXdb64byWsy6UJJlLPpJZ1/r9AAAAAH9HkIIiQ4I0vHF537FmpXTS6VKPJKmiXOZXi91UHQAAAOB9CFKQJI1Jb8PyvqAgGVPPlySZn82T6XC4pTYAAADA2xCkIEkalRYliyHtLq7VgbKWD+eV5DxTKixcysuRNq51Y4UAAACA9yBIQZIUHRKkE9uwvM8IDZcx7mxJkmPRXHeUBgAAAHgdghRcxjR271uxr+yY9zOmzJAsFumHDTL37XRHaQAAAIBXIUjB5fS0SFkMaVdxrXLLj7G8Lz5RxoixkjigFwAAAIGJIAWX6FCrhvUMl3Sc7n2SjDMvkCSZq5fLLC7s9toAAAAAb0KQQjNtOZxXkoyM/tKAIVJDg8zFH7ijNAAAAMBrEKTQzOm9nd37dhTV6GBF68v7JMnSNCu1dKHMmmp3lAcAAAB4BYIUmokNtWpIUtuW9+nEUVJSilRVKXPl526oDgAAAPAOBCkc5XD3vuMs77NYZExtnJX6bL5MR0O31wYAAAB4A4IUjnJ67ygZkrYX1ii/ov6Y9zVGT5bCI6WCPGndN+4pEAAAAPAwghSOEhdm1ZCkMEnSV/uPMysVEipj4jmSJAet0AEAABAgCFJo0eg+0ZKOfzivJBmTpktBVmnHZpm7t3V3aQAAAIDHEaTQojP6OJf3/XCoRgWVx1neF9tDxqjxkjigFwAAAIGBIIUWxYdZNTixbcv7pCMO6F27QmZhfrfWBgAAAHgaQQqtauvhvJJk9M6UBg+XHA6Zny/o7tIAAAAAjyJIoVVNQWpLQbUKq469vE+SLGdeKEkyly+SWV3VnaUBAAAAHkWQQqt6hNs0KMG5vO/LvcefldKQk6WU3lJNtczli7q5OgAAAMBzCFI4pvEZzu59szccUl553THva1gsh/dKfb5AZgMH9AIAAMA/EaRwTGf3j9XAhDBV1js0a3mO6hocx7y/cdoEKSpGKiqQ+e1KN1UJAAAAuBdBCsdktRi6e2wvRYUEaVdxrV5ec+yOfEZwiJoO6DUXzZVpmm6oEgAAAHAvghSOKzHCpjtHp8iQ9MmOEi3ZXXrM+xsTp0tWm7Rnu7Rzi1tqBAAAANyJIIU2OaVXpC4b2kOS9Pdv8rS/tLbV+xrRsTLOmCRJciya647yAAAAALciSKHNrhyWoBOTw1VjNzVreY5q7K3vlzKmnu/8Yt0qmfm5bqoQAAAAcA+CFNosyGLortG9FBdm1f7SOv19VV6re6CMXn2koadKpinzs/lurhQAAADoXgQptEtsmFV3j+0liyEt2VOmRTta3y9laWqFvuIzmZUV7ioRAAAA6HYEKbTbkKRw/d/wREnSS2sOamdRTct3HDxcSsuQ6mplLvvEfQUCAAAA3YwghQ658IR4jUyNVL3D1FPLc1RRd/Thu4ZhHD6g94sFMu317i4TAAAA6BYEKXSIxTB0+xkpSoqwKa+iXn/5KrfF/VLGyPFSTJxUUiRzzZceqBQAAADoegQpdFhkSJDuGddLVouhVdkVmr+1+Kj7GDabjEnnSuKAXgAAAPgPghQ6pX+PMF1/SpIk6T/f5WtLQdVR9zEmTJOCg6X9u6UfNri7RAAAAKDLEaTQadMHxGpcepQaTOnp5QdUWmNvdrsRGS1j9FRJkuPTeZ4oEQAAAOhSBCl0mmEYuvW0ZKVGB6uw2q5nV+bK8aMlfMaU8yTDkL5fLTM320OVAgAAAF2DIIUuEW4L0r3jUhUcZOi73Eq9s7Gw2e1Gcqp04khJ4oBeAAAA+DyCFLpMemyIbhmVLEl68/tDWp9X2ex2y1kXSpLMr76QWd76Qb4AAACAtyNIoUtN7hujqf1iZEr604oDKqw64uyo/kOk9Cypvk7m0o89ViMAAADQWQQpdLkbR/RUZlyISmsa9McvD6jB4dwv1fyA3g9l1td5skwAAACgwwhS6HIhVovuGZuqMKtFmwuq9d/1Ba7bjFPHSHEJUnmpzFVLPVglAAAA0HEEKXSLXtHB+uUZzv1S728u0qrsckmSYbXKmDJDkmR+Oo8DegEAAOCTCFLoNqP7ROu8gXGSpD9/lauDFc6lfMa4s6SQMOnAPmnTd54sEQAAAOgQghS61U9PTtKAHqGqrHPoqeUHVN/gkBEeKWMsB/QCAADAd1k9XUBLFi5cqAULFqikpETp6em6/vrrlZWV1eJ9P/vsMy1btkz79++XJPXt21dXXXVVq/eHe9mCDN0zLlV3fLRbO4pq9O9v83XTyGQZU86T+cWH0ubvZObslZGa7ulSAQAAgDbzuhmplStX6rXXXtOll16qWbNmKT09XY8//rhKS1s+d2jz5s0aM2aMHnroIf3+979Xjx499Pvf/15FRUVurhytSYyw6Y7RvSRJH20r0fI9ZTISk6VTTpckmZ/O9WB1AAAAQPt5XZD64IMPNGXKFE2aNElpaWmaOXOmgoODtXjx4hbv/8tf/lJnn322MjIylJqaqptvvlmmaWrDhg1urhzHcmpqpC4d0kOS9PyqPGWX1cpy5oWSJHPVUpmlxR6sDgAAAGgfr1raZ7fbtWvXLl144YWu6ywWi4YNG6Zt27a16Tlqa2tlt9sVGRnZ4u319fWqrz98SKxhGAoLC3N97WlNNXhDLV3tmuGJ+uFQtTYcrNJTyw/oj9MGKqjfIGnnVplLPpLlwp+4vSZ/Hm9vxZi7H2PuXoy3+zHm7seYuxfj7Z28KkiVlZXJ4XAoNja22fWxsbE6cOBAm57jjTfeUHx8vIYNG9bi7XPmzNG7777rupyZmalZs2YpMTGxw3V3h+TkZE+X0C2euriHfvLaau0tqdWrG0p192XXqujJ38hYtlA9r/uFLKGhHqnLX8fbmzHm7seYuxfj7X6Mufsx5u7FeHsXrwpSnTV37lytWLFCDz/8sIKDg1u8z0UXXaQZM2a4Ljcl+4KCAtntdrfUeSyGYSg5OVl5eXl+e8bSnWck68HP9+nDTXnKHJWiyQlJchzKV+7c/8ky4Ry31hII4+1tGHP3Y8zdi/F2P8bc/Rhz92K83cdqtbZ5gsWrglR0dLQsFotKSkqaXV9SUnLULNWPzZ8/X3PnztWDDz6o9PTWO8DZbDbZbLYWb/OmD6Zpml5VT1ca2jNc15yYqNfXF+jFtfnqO/5SZbz/NzkWzZPGniXD4v6te/483t6KMXc/xty9GG/3Y8zdjzF3L8bbu3hVswmr1aq+fftq48aNruscDoc2btyoAQMGtPq4efPm6b333tN9992nfv36uaNUdNLFQ+J1aq8I1TWYero2S5WR8dLBHGnDWk+XBgAAAByXVwUpSZoxY4Y+//xzLVmyRNnZ2Xr55ZdVW1uriRMnSpKef/55/e9//3Pdf+7cuZo9e7ZuueUWJSUlqaSkRCUlJaqpqfHQO0BbWAxDt4/upcRwq3Ir7frbqJkyJTlohQ4AAAAf4FVL+yRp9OjRKisr09tvv62SkhJlZGTovvvucy3tO3ToULOOJZ9++qnsdrueeeaZZs9z6aWX6vLLL3dn6Win6JAg3T0uVfd9uldfOXroo7SxOveHL2Xu2ymjDzOLAAAA8F5eF6Qkadq0aZo2bVqLtz388MPNLr/wwgtuqAjdZWBCmK49OUkvr83Xf/rNUP/SvRr46TwZN9zp6dIAAACAVnnd0j4EnhkD4zSmT5TshkV/HPITlX23VmbRIU+XBQAAALSKIAWPMwxDPz89Wb2ibDoUGqc/D7hMDV986OmyAAAAgFYRpOAVwm1BundcqoINU9/2GKT3dlTIrKn2dFkAAABAiwhS8BoZcaG6caTzxO630ibp+y9WergiAAAAoGUEKXiVM/vHaXJYuRyGRc/kx6qwstbTJQEAAABHIUjB69x09hClVx1UiS1Cf1q0TQ0OTvAGAACAdyFIweuERoTr7vh8hdprtanKpv99Twc/AAAAeBeCFLxS2pTJunX7+5KkdzcVak1OhYcrAgAAAA4jSMErGbE9NK5vnM7JWSFJenblAeVX1Hu4KgAAAMCJIAWvZZx5ga7d8YGyyvaros6hp7/MUX0D+6UAAADgeQQpeC2jd6Zsg4bq15v/q0jVa1thjV79Lt/TZQEAAAAEKXg3y5kXKqmmWL/c+rYk6YMfirVib5mHqwIAAECgI0jBuw05WUrprRF563VRWKEk6a9f5ymnrM7DhQEAACCQEaTg1QyLRcaZF0iSrl71ioYkhqna7tBTy3NUa3d4uDoAAAAEKqunCwCOxzh9osw5ryuoKF93RmXrzvIk7Smp1XVzdigjNkQZcaHKiA1RZlyI+sSEKMTK7wcAAADQvQhS8HqGLVjGxOkyF7ypuC/e169veFSzlueoos6hTfnV2pRf7bqvxZBSooIbA1ZIY8AKVUK4VYZheO5NAAAAwK8QpOATjInnyPz4XWnPdg0r36tXLx6k7LI67S6u1d6SWu0urtGe4lqV1jYop6xOOWV1WrGv3PX4iGBLi7NXobYgz70pAAAA+CyCFHyCER0r44xJMpcvkuPTubLdep8y40KVGRfa7H7F1XbtaQxWe4trtbukVtmltao8xuzV4JRCJYeaSo8NZvYKAAAAbUKQgs8wpp4vc/kiad0qmfkHZCT1Ouo+cWFWxYVZdXJKhOu6+gZT2WW12lNcqz0ltdpTXKPdJbUqrWmavWp+NpVr9uqIGaz0WPZeAQAA4DCCFHyG0auPNPRUaeNamZ8tkHH1TW16nC3IaHH2qqRx9qqwIUTf78vXnuJa7W9l9sqQc/YqMy7kiP1XoUqMYPYKAAAgEBGk4FMsZ10ox8a1Mld8JvOCq2VERHX4uWLDrDo53KaUlBTl9rbJNE3VN5jKKavV7hZmrw6U1+lA+Y/2XtksSm/cc5URF6oRqZGKD+N/KwAAAH/HT3zwLYNOlNIypOw9Mj//QMb5V3Xp09uCDOdyvlZmr3YX1zQGrFpll9Wqst6hzQXV2lzgnL0KtRq6bGiCLhgUJ1sQSwEBAAD8FUEKPsUwDBlnXyTzX8/K/OAtOVJ6yzJybLe/bmyYVSeFWXXSj/Ze5ZTVNgasWm04WKmdRbV6fV2BPt9ZohtO7akRqZHdXhsAAADcjyAFn2OcNlHavkXmsoUy//UnmaGhMoaNcHsdR85eTcyUHKappbvL9J/v8nWgvF6PLcnWyNQI3XBqT6VEBbu9PgAAAHQf1h7B5xiGIeOam2SMGi81NMjx9ydlbtvo6bJkMQxN6hujv53fVxcOjleQIa3OqdTPP9it19cVqMbu8HSJAAAA6CIEKfgkwxIk47rbpRNHSvV1cvz1MZl7tnu6LElSuC1I152SpL+cm6mTUiJkd5h6d1Ohbp2/S8v2lMk0TU+XCAAAgE4iSMFnGVarLDfdIw0cJtVUy/Hnh2Ue2OfpslzSYkL08KQ0/XZ8qpIibCqstutPKw7ogc/2aU9xjafLAwAAQCcQpODTjOAQWX5+v5TRX6ool+OZ38ksyPN0WS6GYej03lF6fkamrjoxQcFBhjbmV+uOj/foxdV5qqht8HSJAAAA6ACCFHyeERouy68eklLTpdIiOZ79ncySQk+X1UyI1aIrhyXohRl9dUbvKDlM6cNtJbplwS4t2lGiBgfL/QAAAHwJQQp+wYiMluX2R6TEZKkgzzkzVV7m6bKOkhRp02/Gp+rRKb2VFh2sstoGvbAqT3d/slc/HKr2dHkAAABoI4IU/IYRGy/LnY9JsT2k3P3OPVPVVZ4uq0XDkyP053MzdcOpSQq3WbSzqEb3fLJXf/7qgIqr7Z4uDwAAAMdBkIJfMRJ6ynLno1JktLR3hxzPPyazrtbTZbXIajF0/qB4/f28vprSN0aS9MWuMt0yf5fmbSmS3c+W+9U3OPRdbqX+vfagXlq5Wzll3vl9AQAAaAsO5IXfMVJ6y3L7I3L86X5p2yY5/jFLllt/K8Nq83RpLYoNs+qXZ6To7P6xemnNQW0vrNG/v83Xoh0lmjmip05KifB0iR1WVmPXmgOVWp1Toe8OVKradZZWkSQpKz5UEzKjNS49WnFh/HUEAC1xmKYshuHpMgD8CD+5wC8Z6f1k+fmDcvz5IWnDGpn/elaaeZcMS5CnS2vVwIQwPXV2uj7fWarX1xUou6xOD32xX6f3jtT1pySpZ2Swp0s8LtM0tb+sTquzK7Q6p0I/HKrWkRNrsaFBGpEaqSqHVV/vKdSOohrtKKrRK9/m68Se4ZqQGaPTe0cq3Oa93ycAcBfTNDV3S5He2lCos7NidN0pSTIIVIDXIEjBbxkDhshyy2/leP5xmWu+lMLCpf+7zav/EbIYhs7MitUZfaL05veH9NG2Yn29v0LfHqjUxSfE6+ITeijE6l0rcu0OU5vzq/RNToVWZ1cor6K+2e2ZcSEamRqpkamRyuoRqiCLRSkpKdqye7++3FOmpXtK9cOhGq3Lq9K6vCr9/RtDo9IiNSEjWienRMoW5L3fLwDoLhV1DfrLV7lalV0hSZq3tVhxYVZddEIPD1cGoAlBCn7NGHqqLDPvkuOfT8tcvsgZpi69zqvDlCRFBgdp5oieOisrVi+uOaiNB6v01oZCfbGrVNef0lOn94706Hsor23Q2gMVriV7lfUO121Wi6FhPcM1Ks0ZnhIjWl5SGRtq1bkD43TuwDjlltdp6Z4yLd1dpgPldfpyb7m+3FuuqGCLxqRHa0JGtAYlhrG0BUBA2FVUo1nLc5RXUS+rxdDpvSP15d5y/ee7AiVHBuuMPlGeLhGACFIIAMapY2T8vyqZ//mrzEVzpbBwGTOu9HRZbZIeG6LfT+mtFfvK9e9v85VfadeTy3M0PDlcM0f0VO+YELfVklNWp9U55VqdXaHNBc2X7MWEBOnU1EiNSo3U8JTwdi/NS4kK1pXDEnTF0B7aUVSjpXvK9OWeMhXXNGjh9hIt3F6ipAirxmfEaEJGtPrEuu99A4C7mKapT3eW6sXVB1XvMJUUYdU941KVFR+qmJCD+nBbiZ5ZeUCPh/fRgIQwT5cLBDyCFAKCZeyZctRUyZz9L5nz/idHWIQsU87zdFltYhiGxqZHa0RqpN7dWKg5W4q0Pq9Kv/pwt2YMjNOVJyZ0y56iBoepLQXVWp1ToW+yK3SgvK7Z7X1igjUqLUojUyPVv0eogiydny0yDEP9e4Spf48wXXdykjYcrNLSPaX6al+F8ivtendTod7dVKjMuBCNz4jW+IxoJYR7ZxMRAGiPGrtD//gmT4t3O89AHJkaoV+d0UtRIc6/3284tacOVtRrzYFK/X5ptp4+O90n9s7C+9Q3OCQZLJ3vAoZpmv7VY7mDCgoKVF9ff/w7djPDMJSSkqLc3Fzxrel6jvlvylzwpiTJuPZXCho71efGO7e8Tv/+Nl/fNK6bjw0N0k9PTtLEzOhOL32rrGvQt41d9tYeqFBF3ZFL9qQhSeEamRqpUWmRHf4HvCOf8Vq7Q6tzKrR0T5m+PVChpuZ/hqQhPcM1ISNao/tEKTKYJhUt4e8V92K83c/Xxzy7rFazluVoX2mdLIZ0zfBEXXxC/FF/p1fXO/TbT/dqd3GtescE68mz0j32956vj7mv6Yrxzq+o14IfivTpjlKF2yy6fXSKTkz23c7A3cVmsykxMbFN9yVINSJIBQbTNGW+/W+Zn82TDIssN9+j1BmX+uR4f3ugQi+tyXfNFA1MCNWNI5KV1SO0Xc+TW16n1Y2NIjblV6nhiGGICgnSqb0iNCo1Uif3iuiSma/OfsbLahu0cp9zP9XmgmrX9VaLoZGpEZqQEaNTUyMUHORdTTk8ib9X3Ivxdj9fHvMv95bpr1/nqcbuUGxokH49tpeG9Wz9h9tDVfW6e+FeFVXbNTw5XL+b1FvWLlgR0F6+POa+qDPjvb2wWnO3FGnlvvJmy/INSZcM6aGrTkzwyGfIWxGkOoAgFThM03Tul1rxmRRkVcJDz6qkV4ZPjnd9g6kFW4s0e2OhauwOGZLOzIrRT4YnKia05ZW7DQ5T2w5VO7vs5VRof2nzJXtp0cHOLntpkRqUENYlS/aO1JWf8fyKei1r7Py374j3EWGz6Iw+UZqQEa2hPcMDvkkFf6+4T255nf7zXYHW51UpMtiiuDCr4sOsig9v/PNHlyNsFq9vfuMLfPEzXt9g6pXv8vXhD8WSpKFJYbprbKri23Cm3q6iGv32072qsZs6s1+Mbjst2e2fI18cc1/W3vF2mKZWZ1do7paiZr90PDE5XOcNjNPqnAot2lEqSRrQI1R3jeml5CiWikoEqQ4hSAUW09Eg88U/yly7QkZIiCy3PyplDfZ0WR1WWFWv/3xXoKV7nGvrI4ItuvrEBJ3TP05BFkNV9Q36LrdSq7MrtPZApcpqG1yPtRjNl+yldPNfpN3xGTdNU3tKarV0d5mW7S1TYZXddVuPMKvGZTg7/2XGhQTkD638vdL9Kusa9M7GQi34oci19LQtgoOMo8LVkZebghhnqx2br33GCyrr9dTyHG0rrJEkXXJCvK4ZntiuX1ytzq7QE8uy5TCln56UqIuHuLctuq+Nua9r63jX2h36Ylep5m8t0oFy58+1QYY0LiNaFwyKV9/4w6tWVuwr0wur8lRZ51CY1aKbR/XUxMyYbn8v3o4g1QEEqcBj2utlvvC4zI3fSmHhsvz6cRl9+nm6rE7ZnF+lF9cc1O7iWknOrn9xYVZtPFjZ7Ie7iGCLTk1xzjqd0ivCrWvsu/sz7jBNbcqv0tLdZVq5v1yVR+zz6h0TrAmNTSoCaZN2V425vXFNCEtADmtwmPpsZ6neWF+g0sZfUJyUEqFfTBqkwsJCFVXVq6ja3vhfvYqq7K7LR+5BPJ5Qq6XlsPWj67ztnDl38aV/O9fmVOjZlQdUXudQRLBFd5zRSyPTIjv0XB/+UKwX1xyUJN0zrpfG9InuylKPyZfG3B8cb7xLqu36cFuxPt5eovLGv4sigi06OytWMwbGqUcrjZkKKuv1zIoDrlmrSZnRunFkz4D+5Q1BqgMIUgGqrlbWvz2h2k3fSZHRstzzpIyUNE9X1SkNDlOLdpTojfUFKj/iB7VeUTbXkr3BieEe+2HYnZ/x+gaH1hyo1NLdZVqTU6H6IxaHD04M04SMaI1Jj1Z0iH//g9GRMXeYpnLK6rS9sEY7Cqu1o6jGFdCn9Y/VRSf0aNMSJH/2fV6l/rU2X3tKnOPSKypYN5yapBGpkerVq1ebfnNcXG0/ImjZmwWtptuq6tseuCKCLUcFrbjGsJUUYVO/+FC/XOrqC/92NjhMvbXhkN7eWChJ6hcfqnvH9er0L3VeXnNQC34oVnCQod9P7aOBbmqL7gtj7k9aG+99pbWat6VIS3eXuf6N6xlp03kD4zS1X6zCbMf/5UqDw9Q7Gws1e+MhOUwpJcqmu8b0Uv8egdlinyDVAQSpwGQYhnpGRynn1zdI+3ZKcQmy3PukjB5Jni6t08pqG7Roe4mCLNLItEilRXvH2Uue+oxX1DXo6/3lWrK7TBsPVqnplYMM6ZRezmWNvWOC1Ts6RJF+FqyON+amaSq/sl47Cmu0vbBG24tqtLOwRtXHWKMWHGTorKxYXXxCfKu/6fRXueV1evW7fH2939k5MyLYoiuHOZfS2oKMLv+MV9f/OHA1n9kqqrarsMquuobjv1ZShFWT+8ZoSt9YJUX6z/fN2//tLKmx609fHtD3B6skSef0j9UNpybJ1gVNcRocpv6wLEercyoUExKkp6e5py26t4+5vzlyvB0Oh74/WKV5W4q09kCl6z4DE0J1weB4nZ4W1aH9zZvzq/SnFQd0qMquIEP6yUmJunDw0d0j/R1BqgMIUoGpabwPbNuqhqd+K+Xul5JSnDNTMXGeLs8vecNnvLCqqUlFmWuW5UhxoUHqHROi3jHBSmv8s3dMiGJCgnxyj9WPx7yk2t4YmKpd4enIfXNNgoMM9YsPVVaPUPWPD1X/HmHKq6jTWxsK9cMh5zIQq8XQmf1idMmQHkqM8J8fzFtSVd+gtzcUasEPxbI7TFkM5+zcVcMSFH1EcxdPfMZN01RVvaPFma2my/tKa12zW4acm86n9ovV6b0jfb7LpTf8vdKazflVevrLAyqqtivUaujWUcma0MX7UKrrHbrv073aVVyrtOhgzTq7+9uie/OY+yPDMJSQ1FPvfL1Nc7cUuv7tMiSd3jtSFwyO1+DE8E6/TkVtg174Jk8r95VLkk5KDtevRvcKqBUIBKkOIEgFpma/4Sk6JMese6XCfCk1XZa7n5AREeXpEv2Ot33G95XWavmeMm0rrFF2aa0OHdGo4seigi3qHROitMZg1TsmRGnRwUoIt3ptwKqoa9DOolrl1Vn17Z4CbS+sbtaMo4nVIqXHhiorPlT9ezj/6x0T0uJvNU3T1Pq8Ks3ecMi1rt5qkab0jdWlQ3r41UyH5PyN/+e7SvXf9QUqrWncB5UcrhtO7ak+sUfP9HrbZ7xJrd2hr/eX67Ndpfo+r8p1fUSwRRMyojW1X6z6xbfv+ARv4Y1jbpqm5m4p0mvrCuQwnR1R7x2fqj4x3bM6oLCqXnd/sleFVXad2NPZFr07D1z1xjH3VxW1DVq0s0QfbS9VQYWzQ21IkKGp/WJ03qD4Lm8SZZqmPt1ZqpfWHFRdg6mYkCD98owUjUjt2F4+X0OQ6gCCVGD68Xib+blyPPVbqbRIyhwgy52Pygjt/G94cJi3f8ar6huUXVqn7LI67S+t1f5S558HK+rVWrVhVktjuHIuDWwKWkkRti5vH38stXaHdhU5l+Y17W1q6tp0JENSWkyw+vcIVVZ8mPr3CFVGXEi7ZyVM09SGg1WavbFQGxuXLAUZ0uS+Mbp0SA+/aKW74aBzH1TTb3+b9kGd2iui1fDs7Z9xSTpYUafPd5Xq852lzX55kBkXoqn9YjQ+I8an9g5625hX1DXoL1/lalXjwenjM6J166jkNu1X6YzdxTX6zaJ9qrE7NLVfjH7ejW3RvW3M/dHBijrN31qsz3aWqMbuHOO40CCdOzBO0/rHKaqb/x/dX1qrP6044Pr777yBcfrpyYldsiTVmxGkOoAgFZhaGm8zZ68cT98nVZZLg06U5Ze/k2Hz/R8IvYWvfsZr7Q4dKK9zBaumP3PL69Ta1pTgIEOp0UeGK+dSwZTI4E7/ptjuMLW3pFbbC6sbQ1ON9pXWNjtssUnPSJuGpcapd4SUFR+qvvEhXd6RadPBKr218ZBrpsNiSJMyY3TZ0B7d3lK/OxxvH9Sx+NJnvMHhDMOf7izR1/srmnVmPL13pKb2i9WJPcPd+guBjvCmMd9ZVKOnlucor6JeVouhn52apGn9Y902a70mp0KPL3W2Rf+/4Ym6dGj3tEX3pjH3Nz8cch6g+/X+wwfopseG6Kdn9NXwWIdbm0XVNTj0n+8K9EHjeWeZcSH69ZheSuummVVvQJDqAIJUYGptvM3d2+X40wNSbbU0fJQsN/9GhjVw1gd3J3/7jNsdpnLLnaEqu7QxaJXVKqesrtXN/0GGlBIV7Np7lRbt/DM1OrjF9tUNDlM55XWN+5mcwWlPcW2zLoRN4kKDlNUjzLU8Lys+VDFhNreN+ZZ85wzVd7nODdAWQ5qQEa3LhiYoNdr7A1VVvfM8qPlbD++DOjsrVlef2Hwf1LH46me8vLZBy/aU6dOdJc32DiaEWzWlX4ym9I3x2mMDvGHMm5ZDvbj6oOodppIibLpnnGc6n320rVj/XO1si/7rMb00LqPr26J7w5j7kwaHqW9yKjR3c5G2Hjp8gO5JKRG6cHC8Tk6JaFM30O6yJqdCf/4qV2W1DQoJMvSzET11Zr8Yr13W3hkEqQ4gSAWmY423+cMGOf78iFRfJ+O0CTKuv0OGxb+ns90hUD7jDQ5nJ7zsphmsI5YK1rTSDc+Qc/aod0yw0qJDZEqNrcdrW3xMZLBFWfGhzYJTfNjR+7U8MeY/HKrW7A2HXB2lLIY0Nj1alw/tod5e+JvM1vZBXX9qT6W3sA/qWPzhM76rqEaf7SzR0j1lzc67OrFnuKb2i9HpvaO86swqT495jd2hf3yTp8W7nYeij0yN0K/O6NXtS6+O5V9rD2r+1mLZLM626IMSuzbQeXrM/UWN3aHPdzoP0M2rcP4carVI4zNidMGgOGXEOfctesN4F1Xb9dzKA1rfuPJgdJ8o3TYq2e863RKkOoAgFZiO2xb6+9Vy/O0JqaFBxoRpMq65xS9/++JOgf4ZN01ThdV219JAV9AqrW127tePhRzZQa8xOCVH2tr0efTkmG8vrNbsDYVaneNcImdIGpMepcuHJrQ7oHSXlvZBXX9Kkkaktr4P6lj86TNe1+DQ1/sr9NnOEn2fd/jYgAibReNdDSpCPP73oifHPLusVrOW5WhfaZ0shnTN8ERdfILnW0Y3OEzNWp6jVdkVig4J0tNnp3fpvkV/+px7QlG1XR/+UKyF24tdv6yIDLZoWv84nTsw7qgued4y3g7T1NzNRfrv+gI1mFJiuFV3jemlwUn+s5+cINUBBKnA1JbxdnyzTObLf5JMU8Y5l8hy8U/dXKV/4TPeMtM0VVrb0CxcOUwpq3F5Xmsd9NrCG8Z8V1GNZm885NpzJDl/m3nF0B6u37i6W2f2QR2LN4x3dzhYUafFu8r0+a4S5VceblCREetsUDEh03MNKjw15sv3lOn5VXmqsTsUGxqkX4/tpWE9I9z2+sdTY3fovk/3aWdRjVKjg/XUWeldNnvgr5/z7ranuEbzthZr2Z5SNS00SI606fxB8ZrSL0ahrcz0ett4by+s1h+/PKC8inpZDOmKoQm6bGgPr99P2RYEqQ4gSAWmto63Y9lCma//zfmYi/+fLOdc6q4S/Q6fcffzpjHfXVyjtzcWus4okaTT0iJ15bAE9XVT6+2u2Ad1LN403t3BYZr6Pq9Kn+8s1Vf7y1179awWQ6elRWpqvxgNT45w6w9U7h7z+gaHXvk2Xx9uK5EkDU0K011jU73yrJ2iarvuXrhHh6rsGtozXA93UVt0f/+cdyXTNLUur0pztxRpXe7hA3QHJ4bpgsHxGpUaedz/X7xxvKvqG/Ti6oOuJa0nJIbpzjG9fP5MQYJUBxCkAlN7xtvxyRyZ777ifNzVN8syabo7SvQ7fMbdzxvHfG9Jrd7eeEgr9pa7louNTI3UFcN6dNvm/K7cB3Us3jje3aWitkFL9zhnqXYWHW5Q0SPcqil9nQ0q3NEG351jnl9Rr6e+zNH2whpJ0qVDeujqExO8+jfxexrbolfbHZrcN1q/PD2l08sxA+lz3hGmaSqnrE4bDlbp4+0l2lvi/P/DYkhn9I7SBYPjNTCh7X/XefN4L9ldqn98c1DVdocigi36+WnJGt2n6xucuAtBqgMIUoGpvePtmPNfmR+97XzsDXfIcvqk7i7R7/AZdz9vHvP9pbV6Z2Ohlu8tc7X5PbVXhK4YltCuHzKOp6v3QR2LN493d9pVVKPPd5Vq6e7SZvv9hjU2qDijGxtUuGvM1+ZU6NmVB1Re5/yB8Y4zemlkmm8cUvrtgQo9tsTZFv2a4Qm6fGhCp54vUD/nralvcGhHYY22FFRry6FqbSmoVnltg+v2UKuhqf1idf6guA51v/T28c4tr9OfVhxw/YLhrKwY/ezUnl7VlKatCFIdQJAKTO0db9M0Zb71kswvPpAsFllu+Y2Mk053Q6X+g8+4+/nCmGeX1erdjYVauudwoDopJUJXDuuhwYkd38Sc17gP6qumfVA2i64YlqDpAzq3D+pYfGG8u1Ndg0PfZFfo052lWp9b6ZpxDHc1qIhRVnxolwbY7h7zBoeptzYc0tsbCyVJ/eJDde+4Xl7bDr41C7cX6+/fONui3zWml8Z3oi16oH/Oy2rs2nKoWlsLqrU5v1o7impc57A1CQ4ylBUfqpGpkTorK7ZT+9N8YbztDlNvfn9I720qlCkpLTpYvx7bS5ke2gfbUQSpDiBIBaaOjLfpcMh89c8yv1osWa2y/PIhGYOHd3Ol/oPPuPv50pjnltfpnY2FWry71BWoTkwO15XDEjSkHV2hunsf1LH40nh3t4LKen2xq1Sf7SxVfuXhf2PTmxpUZEQrxsv3pZVU2/WnFQf0/UFny+dz+sfqhlOTZAvyvd+0S9Ir3+Zr7pYiWS2GHpvSWyd0sNtaIH3OTdN5lt/WAudM05aCauWU1R11v5jQIA1ODGv8L1x940K77Bc2vjTe3+dV6pmVuSqutstqMXTdKYk6d0Ccx7t7thVBqgMIUoGpo+NtNjTI8c9Z0ndfSyGhstzxqIx+g7qxUv/BZ9z9fHHM88rr9O6mQn2xq1RN5xoP7RmuK4b20LCe4a3+g9zSPqjhyeG6oYv3QR2LL453d3OYpjYerNJnjQ0qmg6rthhSTEiQYsOszj9DrYe/DrMqNjRIMaGH/7S2sg+pu8Z8U36Vnv7ygIqr7Qq1Grp1VLImZMZ02fN7gsN0tkX/en+Fohrboqd0YB+bP3/O6xsc2lHkXKa3tfG/0iOW6TVJiw52BacTksLbfCRFR/jaeJfV2PWXr/NcR1+MTI3QL05P6ZJfnHQ3glQHEKQCU2fG26yvl+P5x6TN66TwCFnufkJGWmb3FOpH+Iy7ny+PeX5Fvd7bXKjPdpa4WgWfkBimK4YlaHhy80C18WCVXl578Ih9UDZdf0rPbtkHdSy+PN7uUFHXoOV7yvTZzlLtKKpp12Ojgi3NglVs6OHwlZmSKEd1uWJDnfdprY10W5imqblbivTaugI5TOcPzL8Zn+qVh0l3RK3dofs/26fthTXqFRWsp85Ob/fhwf70OS+rbdDWgirXbNOOwhpXN8omNouh/j1CXbNNAxPD3Nru3xfH2zRNfbStRK98m696h6m4MKtuPyNFJ6V4zxEBLSFIdQBBKjB1drzN2ho5nv2dtHOrFB3rDFPJad1Qqf/gM+5+/jDmBZX1en9zoRbtKHXtQxiUEKYrhvVQr6hgt++DOhZ/GG93Ka62q7jarpIau0pqGlRSY1dpTYNKqu0qqW1QaY1dJdV2ldY2yNHOoQy1GooNtTYLXDFNwetHlyOCLa6wXVHXoL98latV2c7P0/iMaN06KllhNt9cytea4sa26AVVdg1JCtMjk3u3a7mir37OTdNUbnm9NjcGp60F1cpuaZleSJAGHbFMr198iEeXc/rqeEvOrpF/XHFA+0vrZEi66IR4XTM8sdUZZk8jSHUAQSowdcV4m1UVcjx9v5S9WwoOkTH9MhlnXSjD5lubkN2Fz7j7+dOYF1bV6/3NRVq0o8S1PMyQZEpu3Qd1LP403t7CYZqqqG1wha2SmsaQdUT4qmwwVFBerZLqhqNmE47HapFiQqyKDQtSSXWDChv3dvzs1CRN6x/rM3s72mtvSa1+s2ivquodmpgZrdvPaHtbdF/5nNc3OLSzqFZbjghOrS3TawpOJySGKyWq+5bpdYSvjHdrau0O/Wttvj7ZUSJJ6t8jVHeN6dWhZaXdjSDVAQSpwNRV422Wlcjxjyel7ZudVyT0lOWKG6Thp3nVX8TegM+4+/njmBdV2zVnc6EWbncGKnfvgzoWfxxvb3fkmDscDlXbHSqpPjpsOUNY8yBWVe846vmSImy6Z1yvbjvTzJusy63UI4v3y2FKV52YoCuHta0turd+zstrGxqbQjiD0/ZjLNNrCk6DEsPdukyvI7x1vNvrq33len5VrirqHAq1WnTzyJ6a1Ne79h22J0h5/44vwAcY0bGy3P0Hmd8scx7ae+igHC88IZ1wsixXzpSRwnI/oCvFh1l1w6k9demQHsqvrO/ydtrwXYZhKNwWpHBbkHpFH/+33bV2h8pqG0NWdYPqGhw6KSVCEcHe/YN1VzkpJUK3jErWC6vy9Ob3h5QcadNEH2uokVdep2V7y/Tl3nLXwbdHig4JagxMzuCUFR/qs10Xfd0ZfaKU1SNUz648oE351Xruq1yty63UTaN6Ktzme//PEaSALmIYhozTJsgcPkrmR+/I/HSutPk7OR75hYzJM2TMuFJGuHdvsAR8TUzjHhigo0KsFiVaLUqMsHm6FI85KytWueV1en9zkf76dZ4SI2ztOm7AE4qr7fpyb5mW7SnTtsLmTUtSj+imNygxTKlRwfyixYskRtj02JQ+endTod7acEhL9pRp66Fq3TWmlwZ04UHs7sC/PkAXM0LDZFz8/2SOnSrH2/+W1n8j89N5MlctlXHxT2WcMUmGhd+EAQC8x/+dlKjc8np9tb9cf1iarafOzmjTjJ47VdY16Ov95Vq2p0zfH6xyNSCxGNKJPcM1PiNaI1Ij+eWKDwiyGLpiWIJOTA7XMysOKK+iXr9ZtFePTOmtYT1955fOfNKAbmIk9VLQzx+QuWGtHLNflg7mOA/yXfqxLFfdKCNzgKdLBABAkmQxDN0xOkWFn9VrW2GNHl2yX0+dneHxvUN1DQ6tyanQsj1lWpNT2Wy/08CEUI3PiNaYPtGKC+NHWl80ODFcz07P1N9W5Sm/sl6DE717JvTH+NQB3cwYdqosg0+U+fkHMhe8Je3eJscTv5YxZoqMi/+fjOg4T5cIAIBCrBbdPyFNd3+yV7nl9frD0mw9OqV9bdG7QoPD1IaDVVq6p0xf7y9v1hAkLTpYEzKjNS492is7vqH9IoODdPfYXqq2O7y2JXprCFKAGxhWm4yzL5J52gSZ7/9H5leLZa74XOa3Xzn3Tk2eIcPK/44AAM+KDbPqwUlp+s0ne7W5oFp/+TpPd45ue1v0jjJNU9sKa7R0T5m+3Fum0prDLcoTwq0anxGt8RnRyogNYb+TH2pqEuNr+MkNcCMjNl7G9XfInHCOHG++KO3dIfOdf8tcvsjZ3W/IyZ4uEQAQ4PrEhOiecal6dPF+LdtTppQom64+sW3toNtrX2mtlu0u07K9ZTpYcfgYmqiQII3tE6XxGdEalBgmC+EJXoggBXiA0W+QLPf9UeaKz2TOeV3Ky5bjuYekk06T5fIbZCQme7pEAEAAa2qL/vyqPM3eUKjkyGBN7qLzfvIr6rW8sePeniPalYdaDZ2eFqVxGdE6KSXC55Z5IfAQpAAPMSwWGePOknnqaJkL3pL5xQfSulVybPxWxtkXyTjnUhkhoZ4uEwAQoM7MilVeRb3e3VSoF1blKinCpqE9O9YMoKzGrhX7nB33NhdUu663WqRTekVqfHq0RqVFKsRKV1v4DoIU4GFGeKSMK34mc9xZcrz1krRlvcwP35a58gsZl10nY8RY1oMDADzimuEJyi2v04p95frDsmzNOjtdadEhbXpsVX2Dvsl2dtxbl1uphsaGe4akoY3tys/oHaUoD3cGBDqKIAV4CaNXH1nueFT67ms53v6XVJgv88WnZS75SJYrb5TRO9PTJQIAAozFMPSrM1J0qKpePxyq0WOLs/X02emKCWv5AOP6BlPf5jrD0zfZFaprONyuvF98qCZkRGtsepR6hAfuAcjwHwQpwIsYhiGdcoYsQ0+RuWiOzI/flbZtkuOxO2RMmCbjgqtlREZ7ukwAQAAJsVp034Q03fPJXuVV1OuJZTl6bGof1+0NDlOb8qu0bE+Zvtpfroq6w+3Ke0XZND4jWuMyots8kwX4CoIU4IWM4BAZM66UecZkme+8InPtCplLPpK5ermMC6+RMf5sGRaWQgAA3CM21KoHJ6bp3kV7taWgWn9emaufjY/S+2sPavmeMhVV2133jQ+zalx6lMZnxKhfPO3K4b8IUoAXM3okybj5Xplbv3fun8rZK/ONf8hc+oksV82UMWCop0sEAASI3jEh+s24VD38xX4t31um5a+vcd0WEWzR6N7OduVDksIVRMc9BACCFOADjEEnyvLgczKXfixz3htS9m45nr5PxqjxMi65VkZ8gqdLBAAEgBOTI3Tbacn669d5CrZaNCo1UuMyonRKSoRsQXTcQ2AhSAE+wggKkjF5hsyR42XO/a/M5Z/I/GaZzHWrZEy/TMZZF8qwBXu6TACAn5vSL1YnJkcqK72XygoLZJrm8R8E+CF+dQD4GCMqWpb/u1WW+5+RsgZLdbUy5/5Xjod+LnPdKv5BAwB0u6RImyKC+X08AhtBCvBRRno/We55UsbP7pJi46WCPDleeFyOPz8sMzfb0+UBAAD4NYIU4MMMw5DltAmyPPZ3GedcIlmt0qbv5HjkF3K882+Z1VWeLhEAAMAvEaQAP2CEhsly8U9lefh56cSRUkODzEVz5XjgZjk+/0BmZYWnSwQAAPArLG4F/IjRs5eCfvGgzA1r5HjrZSn/gMy3XpT57isyTj5dxtip0qATOYMKAACgkwhSgB8yho2QZfBwmcs+kbnsE+f5U6uXy1y9XIpPkDF6ivO/xGRPlwoAAOCTCFKAnzKsNme79EnnSvt2ylzxmcxVS6WiQzI/mC3zg9nSwGEyxk6VcfJoGSEhni4ZAADAZxCkAD9nGIaUniUjPUvmZdfL/O5rmSs+k7asl37YIPOHDTLD/ilj5DgZY6ZKmQOcjwEAAECrCFJAADFswTJGjZdGjZdZWCDzq89lrvhcOnTw8DLAlN7OWarTJ8qIjvN0yQAAAF6JIAUEKKNHoowZV8qcfrm0baNz6d/alVLufpnvvCLz/dekYSNkGTNVGnqqDCt/XQAAADThJyMgwBkWi7OT36ATZV51k8w1y2V++Zm0e5u0bpUc61ZJ0bEyzpgkY8xUGSm9PV0yAACAxxGkALgY4REyxk+Txk+TmbNP5srPZH61WCorkfnJHJmfzJH6DnQGqpHjZISFe7pkAAAAjyBIAWiRkdpHxmXXy7zo/0kb18jx5WfShjXSrh9k7vpB5uyXZJw6RsaYM6UBQ2hQAQAAAgpBCsAxGVardNLpCjrpdJmlxTK/Xuxc+peXLfOrxc4Zq8TkxrOpJsuIT/R0yQAAAN2OIAWgzYyYOBlnXyzzrIucM1MrPnMe8luQJ3PeGzLn/0864SQZY86UcdJpMmw2T5cMAADQLQhSANrNMAyp3yAZ/QbJvOJnMteudJ5NtW2jtOk7mZu+kxkRJWPUeGcr9T79PF0yAABAlyJIAegUIyRUxujJ0ujJMvNzZa78XObKL6TiQzIXfyhz8YdS70znLNVp42VExXi6ZAAAgE4jSAHoMkZSiowLfyLz/KukzeudS//WfS3t3y3zrRdlvvtvmSedpqozz5OZ0EuKjvV0yQAAAB1CkALQ5QxLkDT0FBlDT5FZUSbzm2XOpX/7dslcs0KFa1Y475jQU0bWYOcywazBUq8+zscCAAB4OYIUgG5lREbLmDxDmjxD5r6dMr9aLOvOLarfs0M6dFDmoYPS10tkSlJYuJQ5UEbWYBn9Bkl9B8gI5awqAADgfQhSANzG6NNPlvQsJaek6MCunTJ3bpW5Y4vMnVukXduk6ipp83cyN3/nDFaGReqd4QxV/QY7Z63iEzmzCgAAeBxBCoBHGGHh0pCTZQw5WZJkNjRIOXtk7twq7dji/LMw37kccN8uafFHznAV28MZrLIag1VapvOsKwAAADfipw8AXsEICpL69HO2Sp90riTJLC6Udm5xzlrt2CLt3yWVFMpcu0Jau8IZrIJDpMwBMpr2WfUdJCMi0qPvBQAA+D+CFACvZcT1kEaMlTFirCTJrK2R9mxvXA64Vdq5RaqqlH7YIPOHDc5gJUkpvRubWDTOWiWlsBwQAAB0KYIUAJ9hhIRKA4fJGDhMkmQ6HFJetnO2qnG/lfIPSLn7Zebul5YvcoarqBhXZ0Cj3yApPUuGLdij7wUAAPg2ghQAn2VYLM6W6b36SOPPliSZ5aWHlwPu3Crt2S6Vl0rrVslct8oZrKxWZ5hqmrHqN0gGZ1oBAIB2IEgB8CtGVIx00ukyTjpdkmTW10v7dh7eZ7VzizNY7dzq7Bq4aI7zgYnJMjIHOPdbZQ5w7tey2Tz4TgAAgDcjSAHwa4bN5pxx6jdIOvsimaYpFeTK3LHVNXOlA/ukgjyZBXnSN8ucs1ZBVql3pjNU9R0gI3Mge60AAIALQQpAQDEMQ0rqJSOplzR6siTJrKxwNrHY/YPMXduk3dukijLndXu2S4s/dIariCgps7+MjAEy+jbOXkVGe/T9AAAAzyBIAQh4RkRk8zOtTFM6dFDmbmeoMndvk/bulCrLpY3fytz47eEOgYnJztmqvo1LAnv3ZUkgAAABgCAFAD9iGIYzICUmS6PGS5JMe72UvccZqnZtk7lnm5SXc8SSwKWHG1n07tu436o/SwIBAPBTBCkAaAPDapMy+svI6H/4wODWlgQ2zWJJzZcEZjbutcrsz5JAAAB8HEEKADqoU0sCk1IaZ60Gysjsz5JAAAB8DEEKALpIm5YE7t4mHcyR8nNl5udKq1paEtjYzCKRJYEAAHgrghQAdKNOLQmMjJYGDJHRf4iMAUOktAwZliDPvRkAAOBCkAIAN2t1SeCuHxoDVuOSwIoy6duvZH77lTNYhUVIWYNlNIYrpWfJsPLXOAAAnsC/wADgYc2WBJ42QVLjksC9O2Vu2yRz+yZpx2apulLasEbmhjXOYBUc4jxseMAQGf2HOptYBId49L0AABAoCFIA4IUMq80ZkvoNks65RKajwbnXattGmds2Sds3SRXl0pb1MresP7zPKmPA4RmrrEEyQsM9/VYAAPBLBCkA8AGGJUjq009Gn37S1AtkOhxSXrbMbRulbZuc4aq0SNqxWeaOzTL1jmSxOB8zYIiqTxsns0eKFBHp6bcCAIBfIEgBgA8yLBapVx8ZvfpIE6c791kV5DmXAf6w0fnnoYPOPVd7tuvQormSYUip6YebV/QfIiMmztNvBQAAn0SQAgA/YBiG82yqpBRpzFRJkllUIHP7Zmn7JgXt3Cp79h7n8sDsPTIXf+h8YM9UZ6hq3Gdl9Ej03JsAAMCHEKQAwE8Z8YkyTpsg4/SJSklJ0YEftriaV5jbNkrZe6SDOTIP5kjLFzn3WfVIcu6vGjBExoChznDGWVYAAByFIAUAAcKIjpVOHf3/27vz6CirBP3j35uNLCQkAROykI0QECEEEBi1BRHcGVzaURs5P6dVus+AOqfPmbFnGOx2HNCxu7W1j/YZ57TTSrtgxmlHEWRslcUFZJVFlEBICARCAqEIIXvq/v54KxVCFihIKhXyfM6pk3rfet+qW5d7qnjqLi9m4tWA53pW+77D7vUsYHFgHxwvxx4vhw2rnWA1KK41WI24whlOGBTUm29DREQkIChIiYj0UyZqIIybhBk3CQBbVwv7v29dcn1/AZw8gd38BWz+wnMtq0hnnlVKunOB4JQMZzsyqjffioiIiN8FXJBatWoVy5cvx+VykZ6ezoMPPkh2dnaHxx48eJB33nmHoqIiKioqeOCBB7jtttv8XGIRkUuDCY+A0eMxoz0XCm5sgKKC1mBV+D3U1ji9WPu+c45pOTl+CKRkYFLTnb8p6TA0xVnGXURE5BIUUEHqq6++YunSpcybN48RI0awYsUKlixZwgsvvMCgQYPaHV9fX09iYiJXXXUVr7/+ei+UWETk0mVCwyBnjDNXCrBNTc6S66UHnEUrSg9AaTFUHvPe7M7NzrEAwSFOmErJgFRPL1ZKBsQP0bwrERHp8wIqSH344YfMmDGD6dOnAzBv3jy2bt3K6tWrueOOO9odn52d7e2teuutt/xZVBGRfseEhDjD+VIzYMo0735bUw2lJdhDxVDaErAOOL1XpQec7Y1n9F5FRDnDAVPTPX8zIFnDA0VEpG8JmCDV1NTE/v372wSmoKAgxo4dS0FBQbe9TmNjI42Njd5tYwwRERHe+72tpQyBUJb+QPXtf6pz/+vpOjdR0eBZQr2FtRYqKzzh6oCz5HrpASg7BLWnvRcOhjOHB17WOjSwJbAlpjgBro+xzc1q436kzxX/U537l+o7MAXMt1NVVRVut5vY2Ng2+2NjYzl8+HC3vc57773Hu+++693OzMzk2Wef5bLLAuvaKUOHDu3tIvQrqm//U537n9/rPDkZxoxrs8s2NtJ4qJjG4n2e214aiwtpPnbUCV6VFbBjc2u4CgkhNDWD0IwRhGYMJzQjm9CMbIKHJF7UfyhsczO2vg5bV4utr8NdV+vc99zcLY91tu05r3W7Dltf632+Q0FBhA0fxYDLcxlweS5hl+cSMiTxwutSzos+V/xPde5fqu/AEjBByl/uvPNOZs2a5d1u+SKuqKigqampt4rlZYxh6NChlJWVOb/oSo9Sffuf6tz/Aq7OwwfCqDznBhgg+HS1Zxig54LBhzzzr+pqvaGrjcgo7+IWZnAiNNZj6+uhvs65NdRBfT32jPtn7qexoWffY3MzDQXf0lDwLdXvv+3six+CGX45ZvgozPBRMCyrT/a2BaKAa+P9gOrcv1Tf/hMSEnLeHSwB8wkeExNDUFAQLperzX6Xy9Wul+pihIaGEhra8SpSgdQwrbUBVZ5Lnerb/1Tn/hfQdR4ZBSNGY0aMpqWfqWV4IIcOYA8Vtc63OloKNadhr+fiwhfzusZA2ADnNiDcuZ1x33S4fwCEeR4/437LfhMeQcKgGI5+/QV233fYwu/hUJGzGEfl59hNnzuvHRYGGSM8wepyyBqFiY65yIrs3wK6jV+iVOf+pfoOLAETpEJCQsjKymLXrl1MnjwZALfbza5du7j55pt7uXQiIuJvxhgYnACDE7zXugJneKCzemAxHDoAJys9YcgTZgaEt7lvOtlPWDiEhXX7nANjDCFJSQRNmYadPNUpc10tFO/FFn7vBKvC76GmGgq+dZaXbzk5Idnprcr2hKukYboAsohIgAqYIAUwa9YsXn75ZbKyssjOzmblypXU19dz3XXXAfDSSy8RHx/PnDlzAGeBikOHDnnvV1ZWUlxcTHh4uMaQiohcokxoKAzLxAzL7O2inDcTHgGjcjGjcgGwbjccPYwt/A5awtWRg1B+GFt+GNZ/5rkAchRk5ji9VtmjIHMkJiKyV9+LiIg4AipIXX311VRVVZGfn4/L5SIjI4OFCxd6h/YdO3aszS+HlZWVPP74497t5cuXs3z5ckaPHs2TTz7p59KLiIicHxMUBEmpmKRU+MENANjT1bB/D7bQMxywqMBZ4XD3NuzubU6wMsZZMj5rFLSEq8uSAnYlL+tuhtPVUF0Fp05CdRX2VOt9GhswYyZC7iRMcHBvF1dExCfGaqAl4Cw2ceay6L3FGENSUhJHjhzRGFg/UH37n+rc/1Tn/tVd9W2bm515YYXfQ0u4Ona0/YHRg5xQNXyUE7Aysp25XT3ANtR7QpEThmx1FVSfhFOnoPoktiUgnapy/p4+BedTB7GDMT+4AXPtDZh431fRVRv3P9W5f6m+/Sc0NLTvLTYhIiIirUxwMKRlYdKyYPqtANiTJ7xDAW3hd3Bgn9O7883X2G++dnqtgoMhbXjr6oBZozDxQ9o9v3W7nYsmnzrphKGze4tOVTlBqWW7uspZ+fBCREbBwEEQHQMDYzDRg2BgDDQ2YL9eC67j2A+XYVfkQ+6VBE29CcZMwASpl0pEApeClIiISB9hBsXBhKswE64CPAtvlBSesYjFd3DyBBQVYIsKsJ984JwYPwRSM50gdGYwcrt9L0RwiBOC2oSiaE9QGuTZF+O9T1R0l8u82x/+LXbbeuy6/4M9O2H7RtzbNzoXaL72BqenKnbwhVSXiEiPUpASERHpo0xoqHdYH3iWjD9e7l0Z8Myl16k81vGThEe0hp4ze4vO7j2KjnHCUkRkt87JMqGhmMlTYfJU7JFD2HX/h/3qU+cCze+/hV2+DMZNJmjqzTA6T6sYikjAUJASERG5RBhjYEgiZkgiTJkGnLH0etkhiIg6Ixh5eo86ubZibzBJqZh7H8LeORe75SvsulWw7zvYtgH3tg1w2VDMtTdirpmBiYnr7eKKSD+nICUiInIJO3vp9b7AhA3AXDUdrpqOLT3g9FKtXw0VZdg/L8W+/xYmbwpm2s3OewvQVQtF5NKmICUiIiIBy6SkY370E+xdD2A3f45du8qZA7blS+yWLyEhGabdRPOd9/d2UUWkn1GQEhERkYBnBgzAXDMTrpmJLdmPXbcKu2EtlB/G/d9/5PB7b2AmXo2ZehOMuKLP91LZ+jo4VIw9WAQVR5x5cHlTtJKhSABRkBIREZE+xaRlYebOx979Y+zGdU4vVUkh9uu1znLqScMwU2/CXHU9Jmpgbxe3S9ZaOFkJB4uxB/fDwSInPJUfbnMNLgswOAFz/SxnJcPIqF4rs4g4FKRERESkTzLhEU5gmnYzg6tPUP4/bzhB6shB7Dt/wP55KebKazBTb3ZWN+zlXirb3AxlpdhDRXBwvxOYDhY5S9J3ZFA8DMvExA3GblvvrMj43/+F/eBtzA9mOqEqIcm/b0JEvBSkREREpM8LGzGa4AcexX33j52eqXWrnKFx61c7C1WkpGOm3YyZcp1fenNsbY3z+oecsGRL9sPhEmhsaH+wCYKhKZhhWZCWiUnNhGEZbVYmtPfNw25Y41wb7MhB7KfLsZ996CwNP/N2yOn7wxlF+hoFKREREblkmMgozPRbsdfdAvv3YNeuwm7+AkoPYN96Bfvua5hJ12Km3QIZ2RcdPqy1cOKYd0ie08u0HyrKOj5hQIQTkoZlQmqmE55S0jBhA7p+X2EDMFNvwl57I+z+BvcnH8CuLfDN17i/+drpuZo5GzNpakAtaS9yKVOQEhERkUuOMcZ7sWJ778PYDauduVRHDmK//AT75SeQloWZejNmylRMeOQ5n9M2NUHZQWxJERwqah2ad/pUxyfEDXECTmomJi0ThmXCkKEXdVFhYwxcMZ7gK8Y7FzD+9APs+s+cIPfHF7H/8zrmulud3reY2At+HRE5N2PtGTMZ+7GKigoaGxt7uxgYY0hKSuLIkSPon6bnqb79T3Xuf6pz/1J9+9/51rm1Fvbudlb82/IVNHm+9wdEYKZMw0y7CZM23Dm2prp1AYiW0HS4BJqa2j9xUJCzwMWwLE9vU5bT2xQd0xNvt/37On3KudbWZyvAddzZGRLqvKeZf+0MFexmgdLObW0NfPcNdtdW7Pc7nPedku708qVkQEraRYfXQBAo9d0fhIaGctlll53XseqREhERkX7BGOPMJcq5wplz9NVn2HX/B0dLnXC1bhWkpENdLRwv7/hJIiKdXqZhWZDqCU3JwzChYf59M2cwUdGYW+7G3nAHdutXzjyqooLWnrdRuc48qrET+3ygsNY6wzR3bcHu2gr7dkNzc9tjjhyEzZ6VDgHCBkByGiYlDVIyvH+JidW8MrkoClIiIiLS75iBMZgb78DecDvs2en06GxdD6UHWg8anNA6NG9Yy9C8xID9z7cJCcFMnoqddK0zP+wv7zvv6fsduL/fAQnJTg/VVddjwiN6u7jnzel12t4ank4ca3tAYgpmzATMmAnO8aUHnLBVegAOH4SGeijeiy3e6zzect7AaCdYJadBajomOd1ZlCTi3MM8RUBBSkRERPoxYwyMysWMysVWubC7v8HEDXaG5gX4Nag602Z+2PFy7OoV2HUfQ/lhZ8GN997AXHujs3z64PMbwuRP1lo4XOIEp51b2vc6hYXByFxPeJrYbgl4M2Zi63M1N0PFESgtwZYWe0JWCZQfgepTTojes9M5tuWk+MucQJWaDsmev4mpWsRD2lGQEhEREQFMTCzmr67r7WJ0KzM4AXP3j7Gz7sOu/wz7yXInUH38HvaT9zHjr8LccDtkjezVnjZbVwPf7fD0Om2ByrN6nRKSMWMnOiEp54pzrnLYwgQHw9BUGJqKmXh16+s11MORQ9jSYm/IorTEmWNWWQGVFdidm51jAYKDnTKkeHqtPH8Zktjnh0vKhVOQEhEREbnEmfAIzPTbsNNugZ1bcH/yPny/A7vlS+yWLyEzBzPjrzETr8GE9Px/D51ep4OtwWnvbmg+YyGP0DAYOdbpcRo7AZOQ3K2vb8IGQPpwTPrwtuU6fcozLLAESos9fw9A7WlnxccjB2HzF13Mv/IELM2/6hcUpERERET6CRMUBOMmETxuEvZQEfaT5div1zqLU/zhOec6W9ffhpl6EyYqultf29bVwvfbsTu3enqdKtoekJDkBKcxE2HkmPPudepOJioacsZgcsZ49znXCjvuBKzDB+CQ52+X869iWnuuEpIwg+IgNh4GxUNsfK8uTiLdR0FKREREpB8yqZmYv30Me9f/cy5cvGYluI5j/7wU++EyZ1GKGbMxSakX9PzWWqcXp2WRiIJvO+h1GuMNTyaxe3uduosxBuKHQPwQzNiO5l8dwB5qDVlUHIHqqo7nX7WIivYEqziMJ1wRG485I2wxKA4TonlZgUxBSkRERKQfMzGxmL++D3vzD7Gb1mH/8oFz7ay1q5yLGI+ZSNDM2TA675zD1Zxepx2t4ensZeQvG+oZrnel0/MzwP+9Tt2l7fyra7z7bX29c+Fmz+qBHK/AnqwEl+fW1OhcxLllGOEZz9kucA2M8QasyqRUmsPCYZAncLWErphYvwzH7GnW7QZj+tSQyL5f6yIiIiJy0UxoKObqGdirroeCXbg/+QC2b4RdW3Dv2uJcdHjmbMxfXYcZEA44vU72yEHszpa5Tt+2vWhxSGhrr9PYKwO216k7mQEDID0bk57d7jFrLdScdgLVyeNYV2vAsicr4eSJ1sDV3OT0bFVXYQ8Vc3rX1tbnafOCxhu4iB3c2qs1KK41cMXGQ3SsE/7OwVoLjQ3Q2AiN9dBwxv0z9tkO9rU9rgEaG7ANDd773ltH+5qaCPrtG8576SMUpERERETEyxgDI8cSPHIstvwI9rMPsV984gzT+9PL2PeWYn9wI5VBhuav13XS6zTBM9cpt0/3OnU3YwxEDXRuKWl01vdirXWWZz953BO6XEQ3N1B16AD2RCW09HBVnXCWhj910rkdLOq8h8sEQUwsDIpzlpBvE2bOCENNjT32/s+psRdf+wIoSImIiIhIh0xCEua+edjZc7BffoL9dDkcL8eu+h9OtxwUEuoM0xvrCU+JKX1qeFYgMsZAdIxzS83EGENMUhKnjxxxQpaHdbudXiuXE668PVxn3eekC6zbc7/Sh4IEOaErNBRCBzjz2kJDndUKQ0M922HO4hlhYd7tNrcz9psO9rU5ro9du01BSkRERES6ZCKjMDfcjr1+Fmz/GrtxHVFDU6jJGuUsU+4Z6if+ZYI8vUwxsUBW5z1c7maoOukZOnjcGX7ZWZg5czs4WKG4CwpSIiIiInJeTHAwTLiaoInXEJeURN1ZPSQSmExQcOtcqbOunSUXTpdiFhERERER8ZGClIiIiIiIiI8UpERERERERHykICUiIiIiIuIjBSkREREREREfKUiJiIiIiIj4SEFKRERERETERwpSIiIiIiIiPlKQEhERERER8ZGClIiIiIiIiI8UpERERERERHykICUiIiIiIuIjBSkREREREREfKUiJiIiIiIj4SEFKRERERETERwpSIiIiIiIiPlKQEhERERER8ZGClIiIiIiIiI8UpERERERERHykICUiIiIiIuIjBSkREREREREfKUiJiIiIiIj4SEFKRERERETERwpSIiIiIiIiPlKQEhERERER8VFIbxcgUISEBFZVBFp5LnWqb/9Tnfuf6ty/VN/+pzr3P9W5f6m+e54vdWystbYHyyIiIiIiInLJ0dC+AFNbW8vPf/5zamtre7so/YLq2/9U5/6nOvcv1bf/qc79T3XuX6rvwKQgFWCstRQVFaGOQv9Qffuf6tz/VOf+pfr2P9W5/6nO/Uv1HZgUpERERERERHykICUiIiIiIuIjBakAExoayt13301oaGhvF6VfUH37n+rc/1Tn/qX69j/Vuf+pzv1L9R2YtGqfiIiIiIiIj9QjJSIiIiIi4iMFKRERERERER8pSImIiIiIiPhIQUpERERERMRHIb1dgP5m1apVLF++HJfLRXp6Og8++CDZ2dmdHr9+/XreeecdKioqGDp0KPfffz8TJkzwY4n7rvfee4+NGzdSWlpKWFgYOTk5zJ07l+Tk5E7PWbNmDb///e/b7AsNDeXNN9/s6eJeEvLz83n33Xfb7EtOTuaFF17o9By18YuzYMECKioq2u2/8cYbefjhh9vtVxv33e7du/nggw8oKirixIkT/MM//AOTJ0/2Pm6tJT8/n08//ZTTp08zatQoHn74YZKSkrp8Xl+/D/qLruq7qamJZcuWsW3bNsrLy4mMjGTs2LHMmTOH+Pj4Tp/zQj6b+pNztfGXX36ZtWvXtjln3Lhx/Mu//EuXz6s23rlz1fk999zT4Xlz585l9uzZHT6mdu5/ClJ+9NVXX7F06VLmzZvHiBEjWLFiBUuWLOGFF15g0KBB7Y7fs2cPL774InPmzGHChAl88cUX/PrXv+bZZ58lLS2tF95B37J7925uuukmhg8fTnNzM2+//TaLFy/m+eefJzw8vNPzIiIiePHFF/1Y0kvLsGHDeOKJJ7zbQUGdd3yrjV+8Z555Brfb7d0uKSlh8eLFXHXVVZ2eozbum/r6ejIyMrj++uv5zW9+0+7x999/n48++ogFCxaQkJDAO++8w5IlS3j++ecJCwvr8Dl9/T7oT7qq74aGBoqKivjhD39IRkYG1dXVvPbaa/zqV7/i3//937t8Xl8+m/qbc7VxgLy8PObPn+/dDgnp+r+QauNdO1ed/+d//meb7W3btvEf//EfTJkypcvnVTv3LwUpP/rwww+ZMWMG06dPB2DevHls3bqV1atXc8cdd7Q7fuXKleTl5Xl/ebjvvvvYuXMnq1at4ic/+Yk/i94nnf1L2YIFC3j44YfZv38/o0eP7vQ8YwyxsbE9XLpLV1BQ0HnXn9r4xYuJiWmz/b//+78kJiaqjXej8ePHM378+A4fs9aycuVK7rrrLiZNmgTAI488wrx589i0aRPXXHNNh+f5+n3Qn3RV35GRkW3+kwjw4IMPsnDhQo4dO8aQIUM6fV5fPpv6m67qvEVISIhP9ac23rVz1fnZdb1p0yauuOIKEhMTu3xetXP/UpDyk6amJvbv39/mwyMoKIixY8dSUFDQ4TkFBQXMmjWrzb5x48axadOmnizqJaumpgaAgQMHdnlcXV0d8+fPx1pLZmYmP/rRjxg2bJg/inhJKCsr46c//SmhoaHk5OQwZ86cTv9zozbevZqamvj888+57bbbMMZ0epzaePcpLy/H5XKRm5vr3RcZGUl2djYFBQUdBqkL+T6QztXU1GCMITIyssvjfPlskvZ2797Nww8/TFRUFGPGjOG+++4jOjq6w2PVxruXy+Vi27ZtLFiw4JzHqp37l4KUn1RVVeF2u9v9ShAbG8vhw4c7PMflcrXr/h40aBAul6uHSnnpcrvdvPbaa4wcObLLIWPJycn83d/9Henp6dTU1PDBBx+waNEinn/+eQYPHuzHEvdNI0aMYP78+SQnJ3PixAneffddfvGLX/Dcc88RERHR7ni18e61ceNGTp8+zXXXXdfpMWrj3aulrfrSji/k+0A61tDQwJtvvsk111zTZZDy9bNJ2srLy2PKlCkkJCRQVlbG22+/zdNPP82SJUs6HDqmNt691q5dS3h4eJs5VB1RO/c/BSnpF1599VUOHjzIU0891eVxOTk55OTktNn+2c9+xl/+8hfuu+++ni5mn3fmMIX09HTvh/r69eu5/vrre7Fk/cPq1avJy8vrctK92rhcKpqamvjtb38L0OHCKmfSZ9PFObNnNS0tjfT0dB599FG+/fZbxo4d24sl6x9Wr17Ntdde2+mcyxZq5/6nGWh+EhMTQ1BQULtfKF0uV6djWWNjYzl58mSbfSdPntTYVx+9+uqrbN26lV/+8pc+/+IeEhJCZmYmZWVlPVS6S1tUVBTJycmd1p/aePepqKhgx44dzJgxw6fz1MYvTktb9aUdX8j3gbTVEqKOHTvGokWLzjms72zn+mySriUmJhIdHd1p/amNd5/vvvuOw4cPX1AQUjvveQpSfhISEkJWVha7du3y7nO73ezatavNr8NnysnJYefOnW327dixgxEjRvRoWS8V1lpeffVVNm7cyC9+8QsSEhJ8fg63201JSQlxcXE9UMJLX11dHWVlZZ1+caqNd5/Vq1czaNAgn5eOVxu/OAkJCcTGxrZpxzU1Nezbt6/Tz/YL+T6QVi0hqqysjCeeeKLTeTpdOddnk3Tt+PHjVFdXd/q5oTbefT777DOysrLIyMjw+Vy1856nIOVHs2bN4tNPP2XNmjUcOnSIP/zhD9TX13vnM7z00ku89dZb3uNvvfVWtm/fzvLlyyktLSU/P5/CwkJuvvnmXnoHfcurr77K559/zt///d8TERGBy+XC5XLR0NDgPebsOn/33XfZvn07R48eZf/+/fzud7+joqLC51/5+6ulS5eye/duysvL2bNnD7/+9a8JCgriBz/4AaA23lPcbjdr1qxh2rRpBAcHt3lMbfzi1dXVUVxcTHFxMeAsMFFcXMyxY8cwxnDrrbfy5z//mc2bN1NSUsJLL71EXFycdxU/gKeeeopVq1Z5t8/1fdCfdVXfTU1NPP/88+zfv59HH30Ut9vt/WxvamryPsfZ9X2uz6b+rqs6r6ur409/+hMFBQWUl5ezc+dOfvWrXzF06FDGjRvnfQ61cd90Vectampq2LBhQ6e9UWrnvU9zpPzo6quvpqqqivz8fFwuFxkZGSxcuND7S0HLl3KLkSNH8thjj7Fs2TLefvttkpKS+Md//EddX+c8ffzxxwA8+eSTbfbPnz/f+0F+dp1XV1fzyiuv4HK5iIqKIisri8WLF5OamuqvYvdplZWVvPjii5w6dYqYmBhGjRrFkiVLvEt0q433jJ07d3Ls2DHvMsNnUhu/eIWFhfzrv/6rd3vp0qUATJs2jQULFnD77bdTX1/PK6+8Qk1NDaNGjWLhwoVt5jMcPXqUqqoq7/a5vg/6s67q+2/+5m/YvHkzAI8//nib8375y19yxRVXAO3r+1yfTf1dV3U+b948SkpKWLt2LadPnyY+Pp7c3FzuvfdeQkNDveeojfvmXJ8r4FyLy1rbaRBSO+99xlpre7sQIiIiIiIifYmG9omIiIiIiPhIQUpERERERMRHClIiIiIiIiI+UpASERERERHxkYKUiIiIiIiIjxSkREREREREfKQgJSIiIiIi4iMFKRERERERER8pSImIiJzDmjVruOeeeygsLOztooiISIAI6e0CiIiIrFmzht///vedPr548WJycnL8WKKetWnTJp577jlee+01wsPD+eMf/8iBAwd48skne7toIiJynhSkREQkYNxzzz0kJCS02z906NBeKE3P2bt3L2lpaYSHhwNQUFDAmDFjerlUIiLiCwUpEREJGOPHj2f48OG9XYweV1hYyIgRIwBoaGiguLiYO++8s5dLJSIivlCQEhGRPqO8vJxHHnmEuXPnEhQUxMqVKzl58iTZ2dk89NBDpKWltTl+165d5OfnU1RURHBwMKNHj2bOnDmkpqa2Oa6yspJ33nmHb775hlOnThEXF0deXh4//vGPCQlp/apsbGzk9ddfZ926dTQ0NJCbm8tPf/pTYmJizln2qqoq7/3CwkKuvPJKqqqqKCwspLm5mcTERKqqqhgwYAADBgy4yJoSEZGeZqy1trcLISIi/VvLHKknnniC9PT0No8ZY4iOjgZag1RaWhq1tbXceOONNDY2snLlSoKCgvjNb35DbGwsADt27OCZZ54hISGBGTNm0NDQwEcffYTb7ebZZ5/1DiGsrKzkn//5n6mpqWHGjBmkpKRQWVnJhg0bWLx4MVFRUd7yZWZmEhUVxeTJkykvL2flypVMmTKFn/3sZ+d8j/fcc8951cXdd9993seKiEjvUY+UiIgEjH/7t39rty80NJQ333yzzb6ysjJ+97vfER8fD0BeXh4LFy7k/fff54EHHgDgjTfeYODAgSxZsoSBAwcCMGnSJB5//HHy8/N55JFHAHjrrbdwuVw8/fTTbYYV3nvvvZz9W+PAgQNZtGgRxhgArLV89NFH1NTUEBkZ2eV7W7RoEQAbNmxg06ZNPProowC8+eabxMXFceuttwKQmJh4HjUlIiK9TUFKREQCxkMPPURSUlKbfUFB7a/UMWnSJG+IAsjOzmbEiBFs27aNBx54gBMnTlBcXMzs2bO9IQogPT2d3Nxctm3bBoDb7WbTpk1MnDixw7lZLYGpxcyZM9vsu/zyy1mxYgUVFRXtetLOlpubC8DHH3/MmDFjyM3Nxe12U1ZWxi233OJ9XERE+gYFKRERCRjZ2dnntdjE2WGrZd/69esBqKioACA5ObndcSkpKWzfvp26ujrq6uqora1tN7eqM0OGDGmzHRUVBcDp06e7PK+6uhq32w3A7t27ueuuu6iqqqKkpMT7+lVVVYSFhXlX8hMRkcCmICUiInKeOuodA9oNATzbz3/+c2+4A1i6dClLly71bv/TP/0TANOmTWPBggXdUFIREelpClIiItLnHDlypMN9l112GYD37+HDh9sdd/jwYaKjowkPDycsLIyIiAhKSkp6tLyPPvooDQ0NbNq0ifXr1/PYY48BsGzZMqKjo7ntttsA2gxXFBGRwNbxT2siIiIBbNOmTVRWVnq39+3bx969e8nLywMgLi6OjIwM1q5d22bYXUlJCdu3b2f8+PGA08M0adIktmzZQmFhYbvX6a6FbUeNGkVubi61tbXk5OSQm5tLbm4ux44dY+LEid7ts5dlFxGRwKUeKRERCRjbtm2jtLS03f6RI0e2Wc1u6NChPPHEE22WP4+Ojub222/3HjN37lyeeeYZFi1axPTp02loaGDVqlVERka2WV58zpw57NixgyeffJIZM2aQmprKiRMn2LBhA0899ZR3HlR32LNnDzNnzgTg6NGjuFwuRo4c2W3PLyIi/qMgJSIiASM/P7/D/fPnz28TpKZOnUpQUBArVqygqqqK7OxsHnzwQeLi4rzH5ObmsnDhQvLz88nPz/dekPf+++/3XkMKnOF0Tz/9NMuWLeOLL76gtraW+Ph48vLyuvXCuC6Xi6NHj3qDU0FBAREREQwbNqzbXkNERPxHF+QVEZE+o+WCvHPnzmX27Nm9XRwREenHNEdKRERERETERwpSIiIiIiIiPlKQEhERERER8ZHmSImIiIiIiPhIPVIiIiIiIiI+UpASERERERHxkYKUiIiIiIiIjxSkREREREREfKQgJSIiIiIi4iMFKRERERERER8pSImIiIiIiPhIQUpERERERMRH/x8/zqiGkwVl+QAAAABJRU5ErkJggg==",
            "text/plain": [
              "<Figure size 1000x800 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "epochs = np.arange(0, len(H.history[\"loss\"]))\n",
        "plt.figure(figsize=(10,8))\n",
        "plt.style.use(\"ggplot\")\n",
        "\n",
        "\n",
        "# plot training and validation loss\n",
        "plt.style.use(\"ggplot\")\n",
        "plt.plot(epochs, H.history[\"loss\"], label=\"train_loss\")\n",
        "plt.plot(epochs, H.history[\"val_loss\"], label=\"val_loss\")\n",
        "plt.title(\"Training Vs Validation Loss\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "T86_p0ouvdOe"
      },
      "source": [
        "# Adding Dropout"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "id": "3v6wBm6kH8QS"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "329/329 [==============================] - 4s 9ms/step - loss: 0.5122 - accuracy: 0.8393 - val_loss: 0.1848 - val_accuracy: 0.9445\n",
            "Epoch 2/20\n",
            "329/329 [==============================] - 2s 8ms/step - loss: 0.2240 - accuracy: 0.9339 - val_loss: 0.1367 - val_accuracy: 0.9594\n",
            "Epoch 3/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1713 - accuracy: 0.9488 - val_loss: 0.1180 - val_accuracy: 0.9647\n",
            "Epoch 4/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1451 - accuracy: 0.9555 - val_loss: 0.1056 - val_accuracy: 0.9689\n",
            "Epoch 5/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1261 - accuracy: 0.9614 - val_loss: 0.1038 - val_accuracy: 0.9699\n",
            "Epoch 6/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1123 - accuracy: 0.9659 - val_loss: 0.0988 - val_accuracy: 0.9708\n",
            "Epoch 7/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.1009 - accuracy: 0.9679 - val_loss: 0.0914 - val_accuracy: 0.9738\n",
            "Epoch 8/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0929 - accuracy: 0.9707 - val_loss: 0.0988 - val_accuracy: 0.9718\n",
            "Epoch 9/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0874 - accuracy: 0.9726 - val_loss: 0.0940 - val_accuracy: 0.9732\n",
            "Epoch 10/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0826 - accuracy: 0.9745 - val_loss: 0.0918 - val_accuracy: 0.9752\n",
            "Epoch 11/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0773 - accuracy: 0.9755 - val_loss: 0.0871 - val_accuracy: 0.9750\n",
            "Epoch 12/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0716 - accuracy: 0.9770 - val_loss: 0.0858 - val_accuracy: 0.9761\n",
            "Epoch 13/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0725 - accuracy: 0.9770 - val_loss: 0.0906 - val_accuracy: 0.9761\n",
            "Epoch 14/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0675 - accuracy: 0.9780 - val_loss: 0.0881 - val_accuracy: 0.9765\n",
            "Epoch 15/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0620 - accuracy: 0.9799 - val_loss: 0.0874 - val_accuracy: 0.9771\n",
            "Epoch 16/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0620 - accuracy: 0.9796 - val_loss: 0.0892 - val_accuracy: 0.9767\n",
            "Epoch 17/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0574 - accuracy: 0.9808 - val_loss: 0.0872 - val_accuracy: 0.9773\n",
            "Epoch 18/20\n",
            "329/329 [==============================] - 2s 8ms/step - loss: 0.0551 - accuracy: 0.9823 - val_loss: 0.0898 - val_accuracy: 0.9767\n",
            "Epoch 19/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0552 - accuracy: 0.9825 - val_loss: 0.0858 - val_accuracy: 0.9781\n",
            "Epoch 20/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.0532 - accuracy: 0.9827 - val_loss: 0.0910 - val_accuracy: 0.9766\n"
          ]
        }
      ],
      "source": [
        "from keras.layers import Dense , Activation, Dropout\n",
        "\n",
        "# model is a 3-layer MLP with ReLU and dropout after each layer\n",
        "model = Sequential()\n",
        "model.add(Dense(hidden_units, input_dim=input_size))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(dropout))\n",
        "model.add(Dense(hidden_units))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(dropout))\n",
        "model.add(Dense(num_labels))\n",
        "model.add(Activation('softmax'))\n",
        "\n",
        "\n",
        "\n",
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer='adam',\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "\n",
        "\n",
        "H = model.fit(x_train, y_train, epochs=20, batch_size=batch_size,validation_split=0.3) #you can specify a validation set or not (it's optional). You can add \"validation_split=0.3\" in case you want to add validation set."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "Qw08sGzlH8NG"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 0s 1ms/step - loss: 0.0776 - accuracy: 0.9788\n",
            "\n",
            "Test accuracy: 97.9%\n"
          ]
        }
      ],
      "source": [
        "loss, acc = model.evaluate(x_test, y_test)\n",
        "print(\"\\nTest accuracy: %.1f%%\" % (100.0 * acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "id": "6TxJdXo2r_HJ"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1875/1875 [==============================] - 3s 2ms/step - loss: 0.0358 - accuracy: 0.9905\n",
            "\n",
            "Test accuracy: 99.0%\n"
          ]
        }
      ],
      "source": [
        "loss, acc = model.evaluate(x_train, y_train)\n",
        "print(\"\\nTest accuracy: %.1f%%\" % (100.0 * acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "id": "tr_430pGr8G8"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x25e3fcef3d0>"
            ]
          },
          "execution_count": 28,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x800 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "epochs = np.arange(0, len(H.history[\"loss\"]))\n",
        "plt.figure(figsize=(10,8))\n",
        "plt.style.use(\"ggplot\")\n",
        "\n",
        "\n",
        "# plot training and validation loss\n",
        "plt.style.use(\"ggplot\")\n",
        "plt.plot(epochs, H.history[\"loss\"], label=\"train_loss\")\n",
        "plt.plot(epochs, H.history[\"val_loss\"], label=\"val_loss\")\n",
        "plt.title(\"Training Vs Validation Loss\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "74KOkwqNu62I"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9p1TO2zBwVy0"
      },
      "source": [
        "\n",
        "Combine both L2 and Dropout\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "id": "LKuGd6aDwz10"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "329/329 [==============================] - 3s 7ms/step - loss: 0.9046 - accuracy: 0.8419 - val_loss: 0.4976 - val_accuracy: 0.9447\n",
            "Epoch 2/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.5013 - accuracy: 0.9292 - val_loss: 0.3784 - val_accuracy: 0.9522\n",
            "Epoch 3/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.3954 - accuracy: 0.9414 - val_loss: 0.3168 - val_accuracy: 0.9563\n",
            "Epoch 4/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.3413 - accuracy: 0.9485 - val_loss: 0.2826 - val_accuracy: 0.9617\n",
            "Epoch 5/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.3111 - accuracy: 0.9510 - val_loss: 0.2646 - val_accuracy: 0.9626\n",
            "Epoch 6/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2976 - accuracy: 0.9533 - val_loss: 0.2595 - val_accuracy: 0.9628\n",
            "Epoch 7/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.2870 - accuracy: 0.9552 - val_loss: 0.2376 - val_accuracy: 0.9692\n",
            "Epoch 8/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.2803 - accuracy: 0.9563 - val_loss: 0.2416 - val_accuracy: 0.9659\n",
            "Epoch 9/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.2753 - accuracy: 0.9570 - val_loss: 0.2369 - val_accuracy: 0.9681\n",
            "Epoch 10/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.2667 - accuracy: 0.9597 - val_loss: 0.2468 - val_accuracy: 0.9647\n",
            "Epoch 11/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2680 - accuracy: 0.9582 - val_loss: 0.2309 - val_accuracy: 0.9692\n",
            "Epoch 12/20\n",
            "329/329 [==============================] - 3s 8ms/step - loss: 0.2636 - accuracy: 0.9600 - val_loss: 0.2296 - val_accuracy: 0.9703\n",
            "Epoch 13/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.2616 - accuracy: 0.9590 - val_loss: 0.2341 - val_accuracy: 0.9677\n",
            "Epoch 14/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2563 - accuracy: 0.9600 - val_loss: 0.2329 - val_accuracy: 0.9687\n",
            "Epoch 15/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2585 - accuracy: 0.9606 - val_loss: 0.2280 - val_accuracy: 0.9689\n",
            "Epoch 16/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2564 - accuracy: 0.9605 - val_loss: 0.2225 - val_accuracy: 0.9714\n",
            "Epoch 17/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2587 - accuracy: 0.9597 - val_loss: 0.2288 - val_accuracy: 0.9698\n",
            "Epoch 18/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2526 - accuracy: 0.9620 - val_loss: 0.2213 - val_accuracy: 0.9708\n",
            "Epoch 19/20\n",
            "329/329 [==============================] - 2s 7ms/step - loss: 0.2492 - accuracy: 0.9616 - val_loss: 0.2241 - val_accuracy: 0.9713\n",
            "Epoch 20/20\n",
            "329/329 [==============================] - 2s 6ms/step - loss: 0.2570 - accuracy: 0.9598 - val_loss: 0.2251 - val_accuracy: 0.9703\n"
          ]
        }
      ],
      "source": [
        "from keras.regularizers import l2\n",
        "from keras.layers import Dense, Activation, Dropout\n",
        "\n",
        "model = Sequential()\n",
        "\n",
        "# First hidden layer with L2 regularization and ReLU activation\n",
        "model.add(Dense(hidden_units, input_dim=input_size, kernel_regularizer=l2(0.001)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(dropout))\n",
        "\n",
        "# Second hidden layer with L2 regularization and ReLU activation\n",
        "model.add(Dense(hidden_units, kernel_regularizer=l2(0.001)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(dropout))\n",
        "\n",
        "# Output layer\n",
        "model.add(Dense(num_labels))\n",
        "model.add(Activation('softmax'))\n",
        "\n",
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer='adam',\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "H = model.fit(x_train, y_train, epochs=20, batch_size=batch_size, validation_split=0.3)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 1s 2ms/step - loss: 0.2111 - accuracy: 0.9724\n",
            "\n",
            "Test accuracy: 97.2%\n"
          ]
        }
      ],
      "source": [
        "loss, acc = model.evaluate(x_test, y_test)\n",
        "print(\"\\nTest accuracy: %.1f%%\" % (100.0 * acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1875/1875 [==============================] - 3s 2ms/step - loss: 0.1965 - accuracy: 0.9789\n",
            "\n",
            "Train accuracy: 97.9%\n"
          ]
        }
      ],
      "source": [
        "loss, acc = model.evaluate(x_train, y_train)\n",
        "print(\"\\nTrain accuracy: %.1f%%\" % (100.0 * acc))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x25e8394c390>"
            ]
          },
          "execution_count": 32,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x800 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "epochs = np.arange(0, len(H.history[\"loss\"]))\n",
        "plt.figure(figsize=(10,8))\n",
        "plt.style.use(\"ggplot\")\n",
        "\n",
        "\n",
        "# plot training and validation loss\n",
        "plt.style.use(\"ggplot\")\n",
        "plt.plot(epochs, H.history[\"loss\"], label=\"train_loss\")\n",
        "plt.plot(epochs, H.history[\"val_loss\"], label=\"val_loss\")\n",
        "plt.title(\"Training Vs Validation Loss\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.11.4"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}