Galaxy_Zoo2_CNNs.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "tb7RkxGAjv0X",
    "outputId": "f4c5e26d-3d74-4c0c-efb6-5c6513c2a654"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount(\"/content/gdrive\", force_remount=True).\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from astropy.table import Table\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "import pickle\n",
    "import cv2\n",
    "from google.colab import drive\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from scipy import interp\n",
    "from itertools import cycle\n",
    "\n",
    "#To use in Google Colab\n",
    "#drive.mount('/content/gdrive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 50
    },
    "colab_type": "code",
    "id": "NUh-8h5Hjv0c",
    "outputId": "5db76956-5bf0-4e32-a337-3de44b9b9a0e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 447 ms, sys: 20.7 ms, total: 467 ms\n",
      "Wall time: 556 ms\n"
     ]
    }
   ],
   "source": [
    "#Reading the labels\n",
    "%%time\n",
    "d = Table.read('/content/gdrive/My Drive/galaxy_zoo/galaxy_zoo_labels.csv')\n",
    "ID = np.asarray(d['GalaxyID'])\n",
    "len(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "awOWS1l7AzRM"
   },
   "outputs": [],
   "source": [
    "#!unzip -uq '/content/gdrive/My Drive/galaxy_zoo/galaxy_zoo_images.zip'\n",
    "#### FOR GOOGLE COLAB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 67
    },
    "colab_type": "code",
    "id": "8sdylaAcjv0i",
    "outputId": "e2733a51-832b-4fd6-80c4-7d9772760125"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000\n",
      "CPU times: user 1.97 s, sys: 300 ms, total: 2.27 s\n",
      "Wall time: 2.76 s\n"
     ]
    }
   ],
   "source": [
    "#Loading images from the database\n",
    "%%time\n",
    "imgdata1 = []    #list to store cropped data\n",
    "#file = open('/content/gdrive/My Drive/galaxy_zoo/data_120_12k_centered.txt','wb') #To store and save in a file\n",
    "for i in range(10000,11000):\n",
    "        \n",
    "    path='/content/images_training_rev1/'\n",
    "    im = cv2.imread(path+str(ID[i])+'.jpg')\n",
    "\n",
    "    '''\n",
    "    NOTE: This part of the code is to identify the brightest point as the centre from cropping but we are using image center as the crop center at the moment.\n",
    "    \n",
    "    \n",
    "    im = Image.open('/content/images_training_rev1/{0}.jpg'.format(ID[i]))    #reading individual images from ID(filenames)\n",
    "    pix = im.load()\n",
    "    pix_val = im.getdata()\n",
    "    pix = np.asarray(pix_val)    #pixel data in array\n",
    "    n,m = np.shape(pix_val)\n",
    "    \n",
    "    \n",
    "    RGB_sum = np.zeros(179776)    #array of R+G+B for all pixels\n",
    "\n",
    "    for i in range(179776):\n",
    "        RGB_sum[i] = sum(pix[i,:])\n",
    "\n",
    "    centre = np.argmax(RGB_sum)    #Finding brightest point in array\n",
    "   \n",
    "    for i in range(centre, n):    #loop to identify \n",
    "        if RGB_sum[i] <= 40:\n",
    "            temp_w = i - centre\n",
    "            break\n",
    "        else:\n",
    "            continue\n",
    "    '''\n",
    "    crop_img = im[151:271, 151:271]\n",
    "    #cv2.imshow('Image', crop_img)\n",
    "    #cv2.waitKey(0) \n",
    "    \n",
    "    imgdata1.append((crop_img)/255)\n",
    "    if i % 2000 == 0:\n",
    "        print(i)\n",
    "    \n",
    "#pickle.dump((imgdata1), file)\n",
    "#file.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4tVTDzcHjv0m"
   },
   "outputs": [],
   "source": [
    "## Load saved data file\n",
    "f1 = open('/content/gdrive/My Drive/galaxy_zoo/data_120_centered.txt','rb')\n",
    "data1 = pickle.load(f1)\n",
    "f1.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "iRVBmdPTjv0w"
   },
   "outputs": [],
   "source": [
    "#Reshaping the data in a format that can be used for training\n",
    "X = np.asarray(data1).reshape((len(data1), 120, 120, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "vCYJoG-gjv03"
   },
   "outputs": [],
   "source": [
    "## Reading labels for Q1 of classification tree categories\n",
    "n = len(X)\n",
    "q1 = np.zeros(3*n).reshape(n,3)\n",
    "for i in range(n):\n",
    "    q1[i] = [d['Class1.1'][i],d['Class1.2'][i],d['Class1.3'][i]]\n",
    "    \n",
    "maxi = []\n",
    "for i in range(n):\n",
    "    maxi.append(np.argmax(q1[i,:]))\n",
    "\n",
    "y = np.zeros(3*n).reshape(n,3)\n",
    "for i in range(n):\n",
    "    y[i,maxi[i]] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "CLHwDN-zjv0-"
   },
   "outputs": [],
   "source": [
    "## Reading labels for Q6 of classification tree categories\n",
    "n = len(X)\n",
    "y = np.zeros(2*n).reshape(n,2)\n",
    "for i in range(n):\n",
    "    if d['Class6.1'][i]>=0.5:\n",
    "        y[i,0] = 1\n",
    "        #y[1] = 0\n",
    "        \n",
    "    else:\n",
    "        y[i,1] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QkroqkZ3jv1E"
   },
   "outputs": [],
   "source": [
    "#Data Split\n",
    "X_train, X_test, y_train, y_test = X[:9000], X[9000:], y[:9000], y[9000:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "AvXMgQB3jv1L"
   },
   "outputs": [],
   "source": [
    "#Define architecture of that will be used for CNN\n",
    "\n",
    "def cnn_alexnet():\n",
    "    cnn = tf.keras.Sequential()\n",
    "\n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=96, kernel_size=11, strides=4, padding='same', activation='relu', input_shape=(120,120,3)))\n",
    "        \n",
    "    cnn.add(tf.keras.layers.MaxPooling2D(pool_size=3, strides=2, padding='same'))\n",
    "\n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=265, kernel_size=5, padding='same', activation='relu'))\n",
    "\n",
    "    cnn.add(tf.keras.layers.MaxPooling2D(pool_size=3, strides=2, padding='valid'))  \n",
    "\n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=384, kernel_size=3, padding='same', activation='relu'))\n",
    "\n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=384, kernel_size=3, padding='same', activation='relu'))\n",
    "\n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=256, kernel_size=3, padding='same', activation='relu'))\n",
    "\n",
    "    cnn.add(tf.keras.layers.MaxPooling2D(pool_size=3, strides=2))  \n",
    "\n",
    "\n",
    "\n",
    "    cnn.add(tf.keras.layers.Flatten())\n",
    "    shape=cnn.output_shape[1]\n",
    "    print(shape)\n",
    "    cnn.add(tf.keras.layers.Dense(shape, activation='relu'))\n",
    "    cnn.add(tf.keras.layers.Dense(1024, activation='relu'))   \n",
    "    cnn.add(tf.keras.layers.Dense(512, activation='relu'))    \n",
    "    cnn.add(tf.keras.layers.Dense(3, activation='softmax'))\n",
    "    \n",
    "    return cnn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "5OJXEcyzvcjf"
   },
   "outputs": [],
   "source": [
    "#Define architecture of that will be used for CNN\n",
    "\n",
    "def cnn_lenet():\n",
    "    cnn2 = tf.keras.models.Sequential()\n",
    "\n",
    "    cnn2.add(tf.keras.layers.Conv2D(6, kernel_size=5, padding='same', activation='relu', input_shape=(120,120,3)))\n",
    "    cnn2.add(tf.keras.layers.AveragePooling2D(pool_size=2, strides=2, padding='valid'))\n",
    "    cnn2.add(tf.keras.layers.Activation('relu'))\n",
    "\n",
    "    cnn2.add(tf.keras.layers.Conv2D(16, kernel_size=5, activation='relu', padding='valid'))\n",
    "    cnn2.add(tf.keras.layers.AveragePooling2D(pool_size=2, strides=2, padding='valid'))\n",
    "    cnn2.add(tf.keras.layers.Activation('relu'))\n",
    "\n",
    "    cnn2.add(tf.keras.layers.Flatten())\n",
    "\n",
    "    cnn2.add(tf.keras.layers.Dense(units=120, activation='relu'))\n",
    "    cnn2.add(tf.keras.layers.Dense(units=84, activation='relu'))\n",
    "    cnn2.add(tf.keras.layers.Dense(units=3, activation = 'softmax'))\n",
    "    return cnn2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "BO6gb2ju1aSa"
   },
   "outputs": [],
   "source": [
    "#Define architecture of that will be used for CNN\n",
    "\n",
    "def cnn_new():\n",
    "    cnn = tf.keras.Sequential()\n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=64, kernel_size=3, strides=4, padding='same', activation='relu', input_shape=(120,120,3)))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "    cnn.add(tf.keras.layers.ReLU(max_value=None, negative_slope=0.0, threshold=0.0))\n",
    "            \n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=128, kernel_size=3, strides=4, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "    cnn.add(tf.keras.layers.ReLU(max_value=None, negative_slope=0.0, threshold=0.0))\n",
    "    #cnn.add(tf.keras.layers.Conv2D(filters=128, kernel_size=3, strides=4, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.MaxPooling2D(pool_size=3, strides=2, padding='same'))\n",
    "            \n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=256, kernel_size=3, strides=4, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "    cnn.add(tf.keras.layers.ReLU(max_value=None, negative_slope=0.0, threshold=0.0))\n",
    "            \n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=256, kernel_size=3, strides=4, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "    cnn.add(tf.keras.layers.ReLU(max_value=None, negative_slope=0.0, threshold=0.0))\n",
    "    #cnn.add(tf.keras.layers.Conv2D(filters=128, kernel_size=3, strides=4, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.MaxPooling2D(pool_size=3, strides=2, padding='same'))\n",
    "\n",
    "            \n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=256, kernel_size=3, strides=4, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "    cnn.add(tf.keras.layers.ReLU(max_value=None, negative_slope=0.0, threshold=0.0))\n",
    "            \n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=256, kernel_size=3, strides=4, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "    cnn.add(tf.keras.layers.ReLU(max_value=None, negative_slope=0.0, threshold=0.0))\n",
    "            \n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=256, kernel_size=3, strides=4, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "    cnn.add(tf.keras.layers.ReLU(max_value=None, negative_slope=0.0, threshold=0.0))\n",
    "            \n",
    "\n",
    "    cnn.add(tf.keras.layers.Flatten())\n",
    "    shape=cnn.output_shape[1]\n",
    "    #print(shape)\n",
    "    cnn.add(tf.keras.layers.Dense(shape, activation='relu'))\n",
    "    #cnn.add(tf.keras.layers.Dense(1024, activation='relu'))   \n",
    "    cnn.add(tf.keras.layers.Dense(512, activation='relu'))    \n",
    "    cnn.add(tf.keras.layers.Dense(3, activation='softmax'))\n",
    "    \n",
    "    return cnn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "u9F7Cm51p5oZ"
   },
   "outputs": [],
   "source": [
    "#Define architecture of that will be used for CNN\n",
    "\n",
    "def cnn_new2():\n",
    "    cnn = tf.keras.Sequential()\n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=64, kernel_size=10, strides=2, padding='same', activation='relu', input_shape=(120,120,3)))\n",
    "    cnn.add(tf.keras.layers.MaxPooling2D(pool_size=3, strides=2, padding='same'))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "\n",
    "    \n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=128, kernel_size=5, strides=2, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.MaxPooling2D(pool_size=3, strides=2, padding='same'))\n",
    "    cnn.add(tf.keras.layers.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "    \n",
    "    cnn.add(tf.keras.layers.Conv2D(filters=256, kernel_size=5, strides=2, padding='same', activation='relu'))\n",
    "    cnn.add(tf.keras.layers.MaxPooling2D(pool_size=3, strides=2, padding='valid'))  \n",
    "\n",
    "    #cnn.add(tf.keras.layers.Conv2D(filters=256, kernel_size=5, strides=2, padding='same', activation='relu'))\n",
    "\n",
    "    cnn.add(tf.keras.layers.Flatten())\n",
    "    shape=cnn.output_shape[1]\n",
    "    #print(shape)\n",
    "    cnn.add(tf.keras.layers.Dense(shape, activation='relu', activity_regularizer=tf.keras.regularizers.l1(0.0001)))\n",
    "    #cnn.add(tf.keras.layers.Dense(1024, activation='relu'))   \n",
    "    cnn.add(tf.keras.layers.Dense(500, activation='relu'))    \n",
    "    cnn.add(tf.keras.layers.Dense(3, activation='softmax'))\n",
    "    \n",
    "    return cnn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ktxEpz7OvvrT"
   },
   "outputs": [],
   "source": [
    "#Define architecture of that will be used for CNN\n",
    "\n",
    "def cnn_vgg16():\n",
    "\n",
    "    model = tf.keras.models.Sequential()\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1),input_shape=(120,120,3)))\n",
    "    model.add(tf.keras.layers.Convolution2D(64, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(64, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.MaxPooling2D((2,2), strides=(2,2)))\n",
    "\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(128, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(128, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.MaxPooling2D((2,2), strides=(2,2)))\n",
    "\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(256, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(256, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(256, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.MaxPooling2D((2,2), strides=(2,2)))\n",
    "\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(512, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(512, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(512, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.MaxPooling2D((2,2), strides=(2,2)))\n",
    "\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(512, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(512, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.ZeroPadding2D((1,1)))\n",
    "    model.add(tf.keras.layers.Convolution2D(512, (3, 3), activation='relu'))\n",
    "    model.add(tf.keras.layers.MaxPooling2D((2,2), strides=(2,2)))\n",
    "\n",
    "    model.add(tf.keras.layers.Flatten())\n",
    "    model.add(tf.keras.layers.Dense(4096, activation='relu'))\n",
    "    model.add(tf.keras.layers.Dropout(0.5))\n",
    "    model.add(tf.keras.layers.Dense(4096, activation='relu'))\n",
    "    model.add(tf.keras.layers.Dropout(0.5))\n",
    "    model.add(tf.keras.layers.Dense(3, activation='softmax'))\n",
    "    \n",
    "    return model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 588
    },
    "colab_type": "code",
    "id": "ngcQeoFElgz9",
    "outputId": "9b55f6ea-a8bf-4678-b4e5-2443c0dccafa"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2304\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_111 (Conv2D)          (None, 30, 30, 96)        34944     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_72 (MaxPooling (None, 15, 15, 96)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_112 (Conv2D)          (None, 15, 15, 265)       636265    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_73 (MaxPooling (None, 7, 7, 265)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_113 (Conv2D)          (None, 7, 7, 384)         916224    \n",
      "_________________________________________________________________\n",
      "conv2d_114 (Conv2D)          (None, 7, 7, 384)         1327488   \n",
      "_________________________________________________________________\n",
      "conv2d_115 (Conv2D)          (None, 7, 7, 256)         884992    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_74 (MaxPooling (None, 3, 3, 256)         0         \n",
      "_________________________________________________________________\n",
      "flatten_32 (Flatten)         (None, 2304)              0         \n",
      "_________________________________________________________________\n",
      "dense_92 (Dense)             (None, 2304)              5310720   \n",
      "_________________________________________________________________\n",
      "dense_93 (Dense)             (None, 1024)              2360320   \n",
      "_________________________________________________________________\n",
      "dense_94 (Dense)             (None, 512)               524800    \n",
      "_________________________________________________________________\n",
      "dense_95 (Dense)             (None, 3)                 1539      \n",
      "=================================================================\n",
      "Total params: 11,997,292\n",
      "Trainable params: 11,997,292\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#Check the summary of CNN here and assign optimizer and loss functions!!\n",
    "###########################\n",
    "cnn = cnn_alexnet()\n",
    "cnn.summary()\n",
    "\n",
    "adamax = tf.keras.optimizers.Adamax(lr=0.002, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.0)\n",
    "\n",
    "#sgd = tf.keras.optimizers.SGD(lr=0.005, decay=1e-6, momentum=0.9, nesterov=True)\n",
    "#cnn.compile(optimizer=adamax, loss='mean_squared_error', metrics=['accuracy', f1_score])\n",
    "cnn.compile(loss='kullback_leibler_divergence', optimizer='adam', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 706
    },
    "colab_type": "code",
    "id": "Hob3Jpwclxyn",
    "outputId": "4346a8e1-37ad-4c12-a832-4cd2c4a046e8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 8500 samples, validate on 1500 samples\n",
      "Epoch 1/20\n",
      "8500/8500 [==============================] - 12s 1ms/sample - loss: 6.9047 - acc: 0.5575 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 2/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 3/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 4/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 5/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 6/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 7/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 8/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 9/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 10/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 11/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 12/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 13/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 14/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 15/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 16/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 17/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 18/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 19/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n",
      "Epoch 20/20\n",
      "8500/8500 [==============================] - 10s 1ms/sample - loss: 7.1014 - acc: 0.5594 - val_loss: 6.9738 - val_acc: 0.5673\n"
     ]
    }
   ],
   "source": [
    "# Train your model\n",
    "################\n",
    "\n",
    "#history = cnn.fit(X_train, y_train, epochs=, batch_size=128, verbose=1)\n",
    "history_lenet = cnn.fit(X, y, validation_split=0.15, epochs=20, batch_size=128, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 50
    },
    "colab_type": "code",
    "id": "x3xjAIt-mArS",
    "outputId": "5c3de8ff-9b03-481c-9046-95e993ce15fd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4000/4000 [==============================] - 3s 719us/sample - loss: 1.1519 - acc: 0.7247 - f1_score: 0.7247\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.72475"
      ]
     },
     "execution_count": 204,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate your results\n",
    "\n",
    "score = cnn.evaluate(X_test, y_test, verbose=1)\n",
    "score[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "qGIvxi477Hsl"
   },
   "outputs": [],
   "source": [
    "### Save your results\n",
    "#cnn.save('cnn_test.h5')\n",
    "#load cnn model\n",
    "cnn1.save('Lenet_8.5k_20Epocs_SGD(0.01)_cnn_test.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 752
    },
    "colab_type": "code",
    "id": "S9nwM41aj-Mq",
    "outputId": "587601d6-05fa-4327-8e5a-a5f04ca5193f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'acc', 'f1_score', 'val_loss', 'val_acc', 'val_f1_score'])\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plot learning curves\n",
    "##########\n",
    "\n",
    "history = history_lenet\n",
    "# list all data in history\n",
    "print(history.history.keys())\n",
    "# summarize history for accuracy\n",
    "plt.plot(history.history['acc'])\n",
    "plt.plot(history.history['val_acc'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()\n",
    "# summarize history for loss\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "ydd3tfWsijy7",
    "outputId": "c82edb9e-3915-40f2-a517-59d3cad88387"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4000, 3)"
      ]
     },
     "execution_count": 201,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_test = cnn.predict_proba(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "9C6YpOIlikcd",
    "outputId": "c21d2277-80dc-482a-ff73-d0095376ebdc"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfYAAAFnCAYAAABU0WtaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzsnXd4FNX6xz8zW7LZ9EogFOm9CShI\nDyWhKYoFVMRrV+yIioKi2FARFbH+rt5rueL1iiiKAiqKqCgIUqVogAQSICF1s33m/P7YZJPNzoag\nhJAwn+fhIXP2nTNnzpTvvOe85xxJCCHQ0dHR0dHRaRTI9V0AHR0dHR0dnZOHLuw6Ojo6OjqNCF3Y\ndXR0dHR0GhG6sOvo6Ojo6DQidGHX0dHR0dFpROjCrqOjo6Oj04jQhf0Usn37dqZNm0ZGRgbp6elc\ndtllbNy4sb6LVStWrFiBzWY76fmuXr2aWbNmnfR8q6MoCldddRVpaWns3r074LelS5fSq1cvMjIy\n/Ndm/PjxLFu2rE7Kcu+99/LNN9/USd71yX//+1//39OmTWPHjh11erylS5dy9dVX1+kxQvFXn4eM\njAzy8/NrtKlaj7Wxr8r3339PTk7OCZfr79KxY0cOHz4M/L3yL1q0iAcffPCkl++MQ+icElRVFQMH\nDhRr1qzxp61cuVKcc845wm6311/Bakl6errIzc2t72L8ZXJyckSnTp2E2+0O+u2jjz4S06ZNC0jL\nzMwU/fr1E3/88ccpKmHDxuv1ij59+pzSY2pdt1NFXT0PR48eFaNGjfrL+19zzTViw4YNJ7FEtaND\nhw4iNzf3b5f/xRdfFA888MBJLNmZie6xnyIKCwvJy8ujZ8+e/rTRo0fzySefEB4eDsDbb7/N2LFj\nycjI4Oabb6agoACA+++/n4ULFzJ16lT69+/Pc889x4cffsiECRNIS0tj69atAJSUlDBz5kzS09MZ\nMWIEH330kWZZQtmtXLmSiRMnoqoqAHPmzGH+/PnMmjWLffv2MXXqVDZu3FjjcTp27MiyZcuYOHEi\ngwYN4l//+hcAZWVlTJ8+nTFjxjBixAhmz56Nx+MJ8LqKioq44447SE9PZ+zYsbz++uvHzbc6u3bt\nYvLkyWRkZHDBBRfw/fffoygKU6dORVVVJkyYwK5du457vVq3bk3r1q3Zs2cPAH/88QdXXnkl6enp\nTJgwgW3btvltX3/9dUaMGEF6ejpPPvkkonzOpw8++ICMjAzS0tK4++67cTqdAEydOpVPPvmEO+64\ngzfffNOfz++//86gQYNQVZVff/2VSZMmMWrUKC699FKys7MBn5d66623Mm3aNJ5++ulanX/Fftdf\nfz0zZ85k5MiRjB8/nv3799d4P1TU+2uvvUZ6ejqKorB582YuuugiMjIyGDt2LD/++CMA//jHPygt\nLSUjI4Ps7GzS0tLYuHEjBw8eZNCgQbz99ttMmDCBwYMHs2LFCgBcLhd33HEHgwcP5pprruHZZ5/l\n/vvvDzonIQRPPvkkaWlppKen83//938Bvz/66KOMHj2acePG+a9Xfn4+1157rb/+33rrLb99Wloa\nL730Eunp6eTk5JCZmcmUKVMYM2YMo0aN4rPPPvPbrl27lnHjxpGens6NN95IUVHRCT8PVeuvwrMN\n9TxMnjyZnJwcMjIycLvdAZ5wqPusgueff57169czc+ZMVqxYgcvl4qGHHiI9PZ0xY8bw1FNPoShK\nUP0uWrSIhx9+mBtvvJFBgwYxc+ZM1qxZw0UXXcSgQYNYs2YN4HsPvfzyy/79qm8DNZb/RK4pUON1\nWbhwIenp6aSnp3PVVVdx5MiRGtPPOOr3u+LMQVVVMWnSJDF+/Hjx3//+V2RlZQX8vnnzZjFkyBCR\nn58vhBDi0Ucf9X+53nfffWLixImirKxM7N69W3Tu3Fm8+uqrQgghnnrqKXHPPfcIIYSYNWuWuPfe\ne4WiKOLYsWNi6NChYvfu3UFlqcnupptuEkuWLBE7duwQI0eO9LcmVHyRH2//Dh06iGeeeUYIIcSW\nLVtE9+7dhdfrFe+++664//77hRBCeDwe8dBDD4mdO3cGeF1z5swRc+bMEUIIUVhYKIYNG+b3PkLl\nWxVFUcSYMWPE8uXLhRBCbN26VfTr10+UlpaK7Oxs0blzZ81ro+X5bdy4UfTu3VtkZ2cLRVHE6NGj\nxX//+1//b4MGDRIej0ds2LBBjBo1SpSWlgqXyyUmTZokVqxYITZs2CAGDBggDh8+7D+3p556Sggh\nxJVXXimWLVsmPv/8c3HFFVf4j/nCCy+IefPmidLSUtGvXz+xbt06IYQQy5cvFxdeeKG/rL169RL7\n9u0LOo+azv+jjz4SXbp0EZs3bxZCCPHcc8+JW265pVbX85VXXvEfY/z48eKzzz4TQgjx8ccfi5Ej\nRwohRFD9Dh8+XGzYsEFkZ2eLLl26iHfeeUcIIcSKFSv8Ht0777wjJk+eLDwejzh48KAYMGCAuO++\n+4LOa9myZWLy5MnC7XaL0tJSMXToULFlyxZ/XWzbtk0IIcQjjzwiZs2aJYTwPT8PPfSQEEKIrKws\n0bVrV5GTk+Mv2+zZs/3533jjjeK1114TQgjxyy+/iB49egi32y3KysrEOeec46+Lxx57TMydO9df\nL7V9HqrWX8V+oZ6H9evX++u0qn2o+6w6FfUuhBCvvfaauP7664XH4xEOh0NMmjRJLFu2LGifF198\n0f/uKSgoEN26dfOf5zvvvCOmTJkihPC9hxYvXuzfr+p2RTlDlb+217Sqxx7quuzZs0eMHj3a3/r2\n9ttvi48//jhk+pmI7rGfIiRJ4q233mLUqFG8/fbbjBw5knHjxrFq1SoAvv32W9LT00lISADgkksu\n4YcffvDvf95552G1Wmnfvj2qqjJ8+HAAOnTowNGjRwFYs2YNV111FbIsEx8fz6hRo/z5V6Umu4cf\nfpg33niDuXPn8tBDD/lbE2q7P8AFF1wAQNeuXXG5XBw7doz4+Hg2b97MunXrUFWVRx55hM6dOwfk\n+91333H55ZcDEBsby6hRowLqQCvfqhw8eJD8/HzGjRsHQPfu3WnWrFmAdx2K3377zd/Hfu655/Lo\no4+yaNEimjdvTmZmJseOHePiiy8GoE+fPv7zWbt2LUOHDiUyMhKz2cw777zD6NGj+eabbxg7dixN\nmjQBYMqUKUHXYtiwYezcuZOioiLAF2+QkZHBr7/+SpMmTRg4cCAA48ePJysry993etZZZ3HWWWcF\nncPxzr9t27b06tULgPT0dDZv3gwc/3oOGzbM//eyZcsYM2aMvx4qWhJqwuv1ctFFFwG+a1dxHhs3\nbiQ9PR2j0UhqaipDhw7V3H/t2rWkp6djMpmIjIxkxYoVdO/e3X9O3bp1A6Bz585+D2327NnMmTMH\ngBYtWpCUlMTBgwc1z+nll1/m2muv9Z+Ty+UiLy+PTZs2kZKSQocOHQCYOXOmZjzIidRfBbV5HqrX\ngdZ9VhPffvstl156KUajEYvFwoQJEwKep6r07t2bhIQE4uLiSEpKYsiQIUDg++VkUtM1rSDUdYmO\njqagoIDly5dTXFzM1KlTmThxYsj0MxFjfRfgTCIqKorbb7+d22+/nfz8fJYuXcrdd9/NJ598QkFB\nAcnJyX7b6OjoAOGKiIgAfB8IsixjtVoBkGXZ33ReWlrKnXfeicFgAHxNnRkZGUHlqMkuJSWFnj17\nsnnzZr+wnMj+FecJ+H9XVZUxY8ZQXFzMCy+8QGZmJueff37QS7KgoIDo6OiAOqj6UtHKt/r+UVFR\nSJIUkEdBQQEtWrTQPJcKevXq5W/eX7JkCcuXL/eff0lJCU6n0y9oADabjaKiIgoLCwOuW8WHUGlp\nKatXr2bdunWAr+nR4/EEHNNqtXLeeefx7bff0qdPH0pKSujTpw+fffYZ2dnZAXVqNpv9XTMxMTGa\n51DT+VffLzo6mpKSEn9Za7qesbGx/r+XL1/O22+/TVlZGaqqBjUHa2EwGDTv15KSkoC8mzRpotls\nW1hYGHBfVOQFEBkZGXCciqbmbdu2sWDBAnJzc5Flmby8vID7pWpdfP/997zyyisUFhYiSRJCCFRV\nDTqu2WzWPL8Tqb8KavM8VK8DrfusJgoKCgLOMyYmJuhjuIKK9wuEvl5/hwULFrB69WoAnn766Rqv\naQWhrkuzZs1YtGgRb775JvPmzaNfv3488sgjNG3aNGT6mYYu7KeIw4cPc/DgQfr27QtAYmIiN9xw\nA19++SV79+4lMTHR77mBr785MTHxhI6RnJzM4sWL/R7GX7HbtWsXO3fupFOnTrz//vtcccUVf/k4\n1Zk8eTKTJ0/myJEj3HbbbSxbtgyjsfIWrKiDZs2aASdeBwkJCRQXFyOE8ItbUVGRvxWktlx88cW8\n+eabrF69mlGjRpGcnExERARffvllkO22bdsoLCz0b1f8nZyczIUXXsh9991X47HS09NZvXo1hYWF\npKenI0kSycnJtGnThqVLlwbZV/Qha1HT+efk5ATcX8XFxf6Xfm2v55EjR5g9ezYffvghnTt3Zv/+\n/aSnp9e4T01ERkZSVlbm387Ly9O0i4uLC6jj/Px8LBZLjXnPnDmTadOmMWXKFCRJYvDgwZp2Ho+H\nO++8k+eff56hQ4fidrvp0aOH5nEdDgfFxcWkpKQE5HEyn4dWrVpp2lYvS8XfcXFxIfM/Ge+UqlQX\n+eLi4lrvO2PGDGbMmOHfPt41rem6APTv35/+/ftjt9uZP38+zz77LAsWLAiZfqahN8WfInJzc5k+\nfTrbt2/3p23dupWcnBy6d+/OsGHD/C948HmNoZomQ5GWlsaSJUsAX/PnE088oTnkKJSdqqrMmTOH\n+++/n9mzZ/PKK6/4mzaNRqPfw6vtcaqyePFi/ve//wE+z6x58+YBniX4miw/+OADwOdtrF69WrMZ\nMxTNmzcnJSXFH5y1adMm8vPzA14ItcFoNHLbbbfx7LPP4vF4SE1NJSUlxS/sBQUF3H333djtdtLS\n0vjmm28oLi7G6/Uyffp01q1bR1paGqtWrfJ7y1999VVAMGAFw4cPZ/PmzXz11Vf+FoGePXuSl5fH\nli1bAMjOzmbmzJnH9Y6Pd/779u1j586dgC9Qsk+fPkDtr2dBQQFWq5U2bdrg9Xr916qsrAyTyYSq\nqic0BKx79+6sWrUKVVXJzc1l7dq1mnZpaWl8/vnnuN1u7HY7l19+eY0fOADHjh2jW7duSJLExx9/\njMPhwG63B9lVpFc05//73//GZDJht9vp06cPeXl5/uDUl19+mcWLFwN19zwYjUbsdjterzeoDrTu\ns+oYjUZKS0sB3/P0v//9D0VRsNvtfPLJJyf8TqlKUlKSP/A0OzubTZs2aR5fq/zVOd41rem6rFu3\njkceeQRVVbFarXTq1AlJkkKmn4nown6K6N27N/PmzWPu3Lmkp6czatQonnzySRYuXEhqaio9evTg\nhhtu4IorriAjI4PS0lLuuuuuEzrGnXfeSWlpKenp6YwbNw5VVenYsWOt7f7zn/+QlJTE0KFDadas\nGZdffjmPPvoo4BuPOnnyZFasWFHr41Tlggsu4JNPPiE9PZ2MjAxMJpO/z7xquUpKSsjIyODKK6/k\nhhtuOCFRliSJ5557jnfffZcxY8bw2GOP8cILL2g28x2P8ePHExYWxpIlS/z5vvfee/6yDRgwAKvV\nSq9evbj22muZOHEi48aNo0uXLowfP56uXbty0003MXXqVMaMGcO//vUvRowYEXScyMhIf79zRf+3\nxWLhxRdfZN68eYwZM4bp06eTkZFx3JfU8c6/d+/e/Otf//KLxMyZM4Ha3zedOnViyJAh/jkY0tLS\n6NWrF1OnTiUpKYk+ffowfPhwzRe+FlOmTCEsLIyRI0fyyCOPMG7cOM1zHDt2LIMGDWL06NFceOGF\nXHzxxZx99tk15n3HHXcwffp0JkyYgN1u57LLLmPOnDlkZWUF2EVHR3PdddcxceJEJk6cSMuWLRk5\nciQ33XQTQggWLVrkj3jfvXu3/5msq+ehY8eOxMTEMHDgwIDx6KHus+qkp6dz991389ZbbzF16lRS\nUlIYN24ckyZNYtiwYQHdSSfKpZdeyqFDhxg9ejQLFizQbK0JVf7qHO+a1nRdzj77bJxOp7++V6xY\nwR133EG/fv00089EJFGbTjIdHZ0GzdKlS/n0009DDhOsL6p2G8yfPx9FUXjggQfquVQ6Og0b3WPX\n0dGpF77++msmTZqE2+2mrKyM7777zt9qoaOj89fRg+d0dHTqhWHDhvHdd98xZswYZFlm2LBhmqM4\ndHR0Tgy9KV5HR0dHR6cRoTfF6+jo6OjoNCJ0YdfR0dHR0WlENJg+dq9XobAweByqzskjLs6q1/Ep\nQK/nukev47pHr+NTQ1JS1Anv02A8dqPRUN9FaPTodXxq0Ou57tHruO7R6/j0pcEIu46Ojo6Ojs7x\n0YVdR0dHR0enEaELu46Ojo6OTiNCF3YdHR0dHZ1GhC7sOjo6Ojo6jQhd2HV0dHR0dBoRurDr6Ojo\n6Og0InRh19HR0dHRaUTUqbDv2bOHkSNH8u677wb99uOPP3LxxRdz2WWXsXjx4rosho6Ojo6OzhlD\nnQm73W5n3rx5DBgwQPP3xx57jEWLFvH+++/zww8/8Mcff9RVUXR0dHR0dM4Y6myueLPZzBtvvMEb\nb7wR9Ft2djYxMTE0bdoUgKFDh/LTTz/Rrl27uiqOjo6OzqnFo4IqQALM2tOvytmlSA4vqAJJFXjP\nigarKdguz47562yMewtRUiNRzorGndZSM8/w17dhyLH5t+3Xd0dNjQzO85AN62tb/dtq0wjsN/fU\nzDNs2R+Yfzns33Ze0BbGa89hHjVrHQgB5QuClz45CGQpyM60PhfLR3v9255zUnBe0iH42J/+Sdiq\nA0h2ry9foHT+YNRka5CtcXs+EfM3IJUf29M1gbJZ52iWM2LhJkw/5/rLWXbn2XgGNA2ykw/ZiJ32\npc9OgNosgqJ3x2jmaX1lC5b3dyOVl7Ps5p44L++kaRs/+iOkUg9xmbkAiAgTSPDnn7dq2p8IdSbs\nRqMRo1E7+7y8POLj4/3b8fHxZGdn11VRdHR0TjFybhlSsQvJo2I4UIKnRyJqy+ggO6nETcSLm1FS\nrEhegZAlHDd018zTvOoAYav2Y1mxH0+3BOw3dMc9slWwod0DSS+REG9BmA2IMAOFX1yomaflwz1Y\nF/3mK4sQOC7pgP323pq2sVeswJBZ4heXov+MRWkTE5znf3YR9djP/m3XiBaULErTzDP6nu8wbTvm\n3y5cMhZvjyQA3MDRckE0ZpcS88/t5CWEcyTPiVeSsYf4WIgocWM85vRvlyKhatjKBpmoQhcAxVFh\nfNG3KQXRFqJV3/l9EWaijaICYBjRGrlXpegpzSMxAJ7YYHE13dTbL5YAnjgrSMHCLvdrhiGl8p5Q\nEywo1fKTS90woT1y/+Y+YS/H2yoOEaZxTj1TMMwb4t8WUSa8GmW0uksxX9kS87h4upX8So+SDXQv\negGxWqbYkuC3s0kSXgmMc9TADNa9F5QnAF0Fo6//gt1xnRBIvo+6FZXn/qnzSfYr/X0b158PwKXV\nsnjgQ0/A9hu3aB+qJhrM6m7w11a50Tkx9Do+NfzlenYpkGeHYpfv5RkbBs1D5PXVAbC5fUKkCkhv\nDZHmYLtdx+CTP+CLfdApHka2gos7aud54cew7hAkhoNBgp+nQkQVD9OrwgNrfX+/tb0yfd4g6JMa\nmJeiwsPr4cPdlWnhRiIfPE/72IfKYMV+AEzbjxFTqoBWPb67EwC5oFzcLIbQ9S0bIN/h34xQJSJC\n2ZZ5ffVeTnyslZykKCpSCoAcQDYYYFSVD46zm0BSFAJYD8QA2/AJ95+vplNokvmzdSxml4LJKIFB\nRgUqSwWkt/P9qw3zh9XOrrsVlpxfo8mvcrl4tomFNhoGJo2Pi74ptTt+SqTvXzmSUOlUsIsOBXsC\nzLod206Ep4xU2yEKLD6HsMeKrXhlI3ajlQmZy9mWqP0xCNDk2yM0LTtMUZjvIyzWVVy78v0NuhVs\n10x/Mbz/cfctyd9PVEIrJI2PodpSL8KenJxMfn6+f/vIkSMkJycfd7+8vNK6LNYZT1JSlF7Hp4Cg\nevaqRCz4FctnmSgpERBmoHTuAJR2sUH7Rs5bT/j7lULomNQe2zxtIYy/91sM2ZXHKVhxIcpZwV6z\neUMOMc9u8G3syMdhANvQZtplX3fI90e5GOYfKUFEVX4sGPYXw4VtMW04TFV5tJU4ceQW+z4wyr1H\nqcgJd/Qi/vM/kcu9MeFRyA9xD1odHiKqbJfllOD6+WCw15yWSlK4ERzlHp5TCXlfh5e5qNpIbbe7\nKcgrJdMg86PJwE6jzNIwE81VlWOfTSI/zoKlPF9neIjX561n+/7Vhg6VLZfuMANuDZNURQW3glzu\nhWc3jyL9mwN4BqVqWINp81EkW6XX5+mVFHCNKpBsbkyb8/zbR5pHEdk6mpFuL9HlLekmIWijqFj/\nuZ2wrytbVcuu6UrEJZ0pLCwLyjf2ii/8TeGS5KHkzWEgQZg9B1nxYFDsWMoOEbYtm+Z7/kN03E4k\nSQTlcyL0yttyXBstQfd6rBhNduxlqRijHRzqOJnipL54jZVevrHIRbt566mQWTXegi1E837YikzC\n1hwE1YBqi8U+uSPuCVW+iL7y/ffQyF3E3LAajkTR4lgJAEJ4cdm/Q3WvZ+GiFxl3/gTkvyjR9SLs\nzZs3x2azcfDgQVJSUlizZg3PPvtsfRRFR+cvIdncyHkOjL8XoMaG4TlPWwijZq3D8smfACiJ4RQt\nGRvkZVr/uR3rv8u9zPKXt1Si9YrH1x9bddujhCyjqN6vKUK8PA2BdpLDi5xjQ20W3C+rNI3AkFv5\nMjfsK/Y3HQMoZ/lE1rTpaOCOHpWIZzZSdm/fyuLEWnx/VCmn5BW+cmp5K9VCfSMWb8FxZedgO5MM\nkkRegoVfzm7C1i6JfBoTTjM1+PyNY9tibBXLT31T2HdWDEZFxWsIjineLRsgzvdBoiXoLcubrLMM\nMiPyygj7rVIw1QQL3q6JAHglX/PuOR6FPFli6JJdcNRO1/3FNClwEjm9J95OlWIfga8117C/GOvi\nLSBJmH/Koey23ji7x6OF+Y+CytYKwJUajlDDg+ykQidhuysdLPVIKe744P59AGOPRAxRvt8kNQ+j\neJ3wz3Kxy4mY8jdgzN+IN87nNZtGbEEQhlTRlhGi1RoA7VNANUXhSSlvUhcC4+FMFEMbVGsrhCEW\nJTUSSZTgjesKkgFhCEcNT0ayuTFmlVScIarV6L8n1YjmIPmurTCEY8h2IhVXPmf2VlFExloIuuuT\nVVwLRlWpDAljk4jqVr5jXOHGeVH5RyoSstWIJcLEiy/+gsejQnlYhCWuO+432gOwudTFz7+u54En\n7mN/fiaSJLFv734iTcHdPLWlzoR9+/btzJ8/n0OHDmE0Glm5ciVpaWk0b96cUaNGMXfuXGbMmAHA\n2LFjad26dV0VRUen1khlHqRCJ8bfC0CScI+sFqAkBBELN2H9v8qmNk+3BIpCCLv524P+vw35Dl9A\nVTWcE9ogZ5cSvrRyZIhpSx7es4NbsVyjW2FZ9mdleZ2hhb26YKMhbEBQYJNleSa2mX01Tav3a6op\n2i84Yfa9QL2tohGxYRj3FuK4uitoiKZrTGvM3x/ENb4NwmTwuYoauu4a0RIlNYpii4EjisDTLYHd\nSRGUSHBnlAUhSUhCIAPKgRu1z7U6nRN9/8qpKupGIYgWgmFuhS5elaEFDgxCpY1HQUZCRJgwGTV8\nKq+K1C4KYTGCsTy/Ekd1Kx/jzqpVMZWzYih9ZsjxDQH3aI24Aw1EnCU4WE0IzFnLMdgPYcr9Ftme\ni8F2ADUiFaOtSvOy3fdf1d5rU0Gl1+wX9XJUUxSyx9dq4o3rhly6HyWuKyhOhCkad8txuFNHoUS1\n9YvvXyIevNoxhUEoZ4XVztAkawYfaiEizYhq3V2qKnj11U3k59u5bnEfAP74o4B27eIpKirk0YUP\n8e67/wagY8dOLFiwiHPOObd2ZQtBnQl7t27deOedd0L+3q9fPz744IO6OryOTiWKiiGzGKnEjTGz\nGKVlFJ5zg6Nfw9/ZSeSTG/zbni7xQcIu55Th6R0ouJLdC2WewL7mikO3iESu0i8btiYb+gY2oarN\nIlGreQAVwlgdEWZEjbf4PTKvRtN6Be6BzfC2j/UJt+wTIi2UNjGU3dITDBJyngNPj0REYrCHB1Cy\ncCiSRwVJQk20oCZYNO1cE9uRN9HXJyyAHFmiRJIok2CfQSbg1ffU4KD9fzfIHJYl9htkEspbGpaf\n1xLOq/mtLSSJ6p86ZiFoq6gMdCuc7Q39IeQBxru9hAkwofFtEW4ADPh+rQGjHPRyr28kTymyzdeU\nbsr7GWP+RoQpGkl1YTz6C8KSiDn3m5D7y+7C4ERDGI52V6FEtUZyFeJN6osa7nuuhNGKakkEyYgw\nh75HzwQ2bcolP98ekNayZQw7dmznsssu5OjRI5jNZu688x5uu+0uwsJq+cFRAw0qeE5H50SRj9qJ\nnLc+oH/Q3bcJxdWF3aWgNI1EjTAhl/n6J7W8YbVZBO7USJQWkRiyfUOKjJnFhL+/C8d1wQE8QR5u\nCMH09EnGNbw5ylkxeDvF4x6i3X/qGdCUY+suC33CVQg1zKc6SusY7Lf2qp1tR+220yIJDsoy+wwy\nHgl+NhmIUwXPRfz9l1Qo2nsVvEgclSXO9ip096rcbncRLXxxDPl5pWjHjTcCFBeS4sKY/yuS6vKn\nGUv2IpceQHbkgtFK2IFlf/kQjo7XI9tzcTcdhrAkokS1RhgjUCJbgsFCUlIUNj0m57isWpUZlGY2\nG2jTpi0RERGcc05/nntuER06hAhY/Qvowq7TYJHyHES8tpXw/+wCQEkOp+DbwMEjxs1HsT3UP0DY\nJafXNySq6njhMAPukS0R84w+7xtfX7P5u4O4BzarbFL19/36/hdGCckrcE7WfihtDw/wHQ9Q4yya\nY28BPOc1C9lPX9/8YZCwl5/37waZXUYDxvLxTIdlmQ8sx/Fgq9DJq3DAIHO2RyEhVNdAOQcMMv09\nCu0UlbhyWzOCdLei1VIfgBH6cS5RAAAgAElEQVQapqgrTgyl+wAJyVOM7CwASca6dT6m/I2olkRk\nZ/5xswmFN7YzCBXJXYQ3oQ+eJuchKS6EKQIlvClKfHcUayoYtT9AdU6ca6/tTcuWMXz55V52/fA2\nrXtfACQSHh7Oxx9/TkpKU2T55M4Vpwu7zumNomL5+A9cI1tWBlsB8pEyom9bg2l75Rhgw9HyPmxT\n5UPiHt0KJAklOdz3O2D8o4jwD/bg+EfXoMOJKDPklfeHGiXUuLBKUa9CwSfng+X4j49WZPvpigB2\nGWR+NhnYYpQ5ZJD51nxir4jOXoUsWWak20uxLHGuR0EFZtjd+sIUVZBchZjyN4AQWLc+hTBYMB/5\n4bj7hRJ1V+po3+/uYl8zeFgc7tRRCIMVSXhxN01DVBmfrXPqaNIkgn79TLz//itsWPUz+VlbYNqL\nADRrpt0y93fRhV3ntCTykZ8I/6ByPKu3bSze3pXCLhW7KXl2CNZ/7iD8w0q78Hd24rywHSKu3Lbc\n05RLfV64p0cixt8LcITwsAs/m1i7AtZC1E9nnECRLPGF2YhdgkVWMwXH8Rq6l0fgZxlkrnO4qWhk\nt0nQ3asy1uU9Xu9zo0VyFSI7fZHwsj0XVDdhBz7BUJaNGhaHpHoJy/oUANUcp91nXQVvdHuQZIzF\nu3E3HY6QjRiLdlGU/iXCYEEYrWDSDlzUOX1wuVw8//yzvPjic3g8HsKjm5DaaWidH7dhv510GiVJ\nXf4dlCaXuH1N5GYDmGSUDnEAKC0Dh45ZlmfiuKZb0P75P0/R9LwbIwLINEhsMBrwSBLrTAY+tpho\nVj4kK0cjMl2LaQ43zRXBRS4PLY7TbN4oEQK8diJ/nY0wxyA5j2HZ/z9fX7MhHGPBFpBkJMV5/Lyq\nUFXU3Sm+l7wpfyO2Po+hRLbC02yE9nA/nQbF+vU/MWPGbezd63M8pk79B/SYQ5j1rw9jqy26sOvU\nC3JuGbGXfY6IC6PonTGI6PIoYiFQI014u/qaDaVSN6adBRgOlBDzzk6K3xgVkI/SKgphlnGNbY0r\nrQXu4S20D9gIRf2YJPGV2cBt0eEYhSCxXHwPhxDuUILezqvQ1avSy6twidNLcqjx7o0V1YNsy8Kc\nu4aon+/2JVmSkZ1HNc2NhdqzikG5py1UjKV/4m42AmPBVsp6P+zzsIUXYY7FG98LIUkIS7Iu4I2U\nvXv3cMEFGQghaNu2Hc89t4gBAwYGTRdbV+jCrnNKMW7JI27KisqEfAdSobNS2CWJY79c7vvbrZDU\ny7fkb+RTGyj4bGLQi9A9shX5v009FUWvFwTwcZiRH00GBPBOuBmzEEQJwbEqTedeSeJw9XHrQIKq\nMtKt4AHOd3npWT7cK1wI4s8U/VYVUF0YC3eU92urmA+uxFCyF4M9R3OX6qKummOxd70D2WNDNcfi\nST7Xdy9KBlRrc4QkI8KPP3umzplB+/YdmDLlSlJSUrjzzplYLNrDQusKXdh16g634htDXcVbNv+U\ni7d1NMZ9Jf40w1E7pm35uMYHTkZt+d9eXMOaU/LMEM0x4o0RL7DS7BPytWYDu43Bsd1uSeJYlQ+c\neFXlHI/CkzaXP0AtURVnTn+3ECAUZPshjEW7sOx5C4q2ECdbMZbsPf7+5ajmWCSPDXezNGz9n/eJ\ndniTvzdhis4ZweHDuTz44H3cfPOt9O3rG2Z69tk3EB9vxes99fePLuw6JxU5x0b4WzuwvucbgnZs\n5UWoLcr7wYXAfkN3nONbkzB6qX+f2GkrcVzYLkjYnZd3CrnkYUPlpXAT7wHRsVY2mwxEVBFgFSjR\nWN6ygnSXl8EeLwKY6PKCgKTymdbOCET5uplCwZi3kaj1t2Eo3Yekajdvhnq5eaPa4k08GyWyFUpU\nG1/0eHiTOiu2TuNFVVXeffffPProQ5SUFJOTc4gVK75CUQSPP/4DhYVOzGYDgwa14IUX0oFTM3GR\nLuw6Jw3ThsPEXL2Squs5yKVuVPCt5GWQQQK12mpkzvRW2B4feErLeiqwA4XlM66tCDMyv+pkLeWr\nYpXVIOSTnR7S3F56ehRan2nBa14HBtsBzDlfEb7zZQz2g8ffBxAGiy+YrfdtlFo6oUS1xhvTERGe\ndPyddXROgL179zBjxu2sX/8jAKNHZzB//nNIksSGDYcoLPQFVbrdCps25ZKQEA5B8yLWDbqw65wY\nHpWw1QcwZBbjGts6cGUtj0rps0OInrHWnyQftGFdexD7VV3AWulb5m+4POQUpw2ZrUaZz8xGnj/O\njGtzbC76ehWiVUEzNXD++FihOVV640VxE7nxAQwlexGyibBDq2o0F0hICFRTNLKnhJKBr+BqOTFg\n+FdSUhROfVY0nTrA7XazaNFCFi58BrfbTWJiEk888TQXXHCRf6nVlSv/DNhn5Mg2GI0yurDrnF6U\neUjq95+AJCU1MkDYPec1QyoIHPoTc+e3OC5uHzjLGzQ6UfcC/eIjOKQRed5MUckxyLRQVKY5PMyL\nDCPPob16W2NCcpcglx3EfPg7VFM0xqIdCKNvvTLjsc2EHVpZq3wUayqelCHYe9yDEt2+bguto3Mc\nioqKeO21xbjdbi6/fCoPPzyPuLjAqZbHj2+Pogi+/PJPDhwoJj1dazH7ukMXdp1aEfXQj0FphqN2\npCIXIrbSOxWxYQiTjORR8fRMomT+INSWjW8RiIogNwk4aJCYHRkY9TrQ7WW4W+F2LQGPrLv50+sL\n49H1GIt2YizcUT507NvKOcxPAFufxxDGCDxNBqLENq74Cp2Gi81WSliYBZPJRHJyMs899xIxMTEM\nHqw92Uzfvs3o27cZjz46lD17CkhNjdK0qyt0YdcJRggMB0r86xgD2Gadg21OfxLPW+JPi3hhM0rL\naFxjzqrcV5bI39Lwh59tMsrcGmWhuzewmXy3UWanRqR6gE1+KXENuUtcCGT7IQyl+yuHF6peX1O5\n0YrBnkPE5kd9yWHxyK6C42bpje2MN6E3si0LT5NBAMjOPDyJfXC3PP+MXwFM5/Rl1aovuO++Gfzj\nH9dz++13ATB+/Pm12leSJDp2PPVT+erCruNDUeG5jSQ9ud6flLdzmv/vimU81dgw5CKfJ+bMOAtX\nRu3Wfm5ItE2IpLQ8qO2P44h4nCo41+MlT5a5w+4iw31q+tD+LpIzH2PxHlDdRGx+FGGMQLbnnNDw\nMCBI1F0tL0DIRoQ5Fk/SObjOmgSG02sJUx2d2nD06FFmz76XZct8I3i++molt956x0lfsKUu0IVd\nBxSV+LHLILsy2EitmDDGrfimcS3HNuscvN0TArz5hs4hWWKB1YxNklhWbaWywW4vVzgDh1N5gNFu\nb8MLclO9hO9+ncgN99d+F0si3ugOAEgeGxICb1xX5LJslMjWODteh2pJQLU2B7lBrqemoxOAEIL3\n33+XuXMfpKioCKvVyv33z+a6625qEKIOurDrAHKeg9KH+xNb7IEZ3/rSStxYX9yM+ZfDFL07xm/r\nmnBqg0DqAgF8FGbkG7OR/9Ww5OjhvNKGNUZcCGRHLqaj6zEeXY8552sM9kNIXjtCMiCJ4NYET/IA\nJNcxDLYsyno/jBLZyifmiX1B1l8POmcWRUWFXHvtVXz//XcADBo0nAceeJKmTVtw+LCdpCQrYWHB\nz4XT6eXAgWKEEBgMMu3axfkj5OsD/ck90/CoxE77kpL5g/0Tx6gpEagpEVDsDTCNeHWrbxrXRkKB\nBI9EWHg/XFvMz/V4GehWGOP20rNa3/ppg8eG7DxK7JfpKLFdMOeuAUDIppATtQBBol46YBHOtpeD\n3LhGJ+jUH7m5pdjtXjweBY9HpW3bOKzW4PursNDBmjUH8HhUvF6FpKQIRo/WdhjeeWcr332Xhder\noqqCq6/uQVpa6yA7IQSjR/8HRVFRFIGqCtauvUpTXJcs2cHs2d+iKL48L7usK08/PQKA6OgY3G43\nCQkJzJv3FK++6mHs2C/9+65ceTm9e6cE5blnzzFGjnwPgIULR9G+fXyQzalEF/YzCOsrW4hY9BvO\nCW1Qm0cGG7T2Na9728ZQdltv3KNaNvhFKjYbZZ6MCOOALLNPYyGYKxxu+npVpjg9p5937nVgLNpJ\n2IFlWHe8EPSzwXHE/3dVUfcNKRO4U4YgzDG4Wp6PJ2WIbyESydjgr6nO8RFCIATIGhMgqarg228P\n4HR6cTq9uN0Kkyd31cxn1apMXn99k992/Pj23HnnuZq2U6d+wtatR6vsezm9egWLYFZWCTfdVLle\nRN++TUMK+5YtR/n008plmYcPP0vTTpIktm49QtX1i1RVYNBYP8HrVSkpqRyxkZu7i6ysdrRs2QpZ\nlnn55TewWiNISEjgjTfeC9hXhFggqeoHxF13rWbRog2MHt2We+8dQGTkqY8x0YX9DCH2yi8wbfI9\ndJblmZi/P4TSMgr7Nd1wjy4PgDPIAQFzDQ0H8F+LCSOwwGrmYIjVzLp6FRaXOOminIZeuddBWNan\nRK+7/rimSlRrSs99DiQDnuT+gOzzwHXhblBUiIWWd1lS4mLZst0UFjo5csRGcnJESGF95JG1vP/+\ndpxOLw6Hl6efHsm0aT2C7CQJJk9eGpB2ySWdMWg8L0eO2Fi7Nsu/3b176IVuTKbA/T0e7efLWO0D\nu6ws9JwO1YVZVUM/s0ajHHBMRREYNMI+KvN0AWtYtepn3O6vWLJkKZIk0aJFS79t9Wui1nIGyMzM\nIr7+eh9z5w6plf3JRhf2MwTHxe2x3XU2cVN9zUpykQu5yKXtuTcgNhllMuIijmt3rsfLlQ4PQz0K\nKafZ9KyyLYuEpcFryFfFk9Ab11mTcHS+WW8+P8UIIfweq8ulkJRk1RTBzMxCVq/eh93uwW730KFD\nPJdc0kUzz5tu+pwffzyI3e7Fbvfw9ddX0rlzYpDdsWMO7rnnK/92165JIYXd5fJSUGWCKKfTq2kn\nSRIWiwGnU6liqxAREXxOFkugRDgc2nkCGKuNIPGG6M6q/gFgs4XuQqou7IoS+tk1GCQ8VbLyelXM\n5mBl9wXA7QU+A4oRQqJz5654vV5MpsBnKyEhnJSUCCRJQpLAZNIOEA0PN9K+fbxvwT9Jol27OO69\n9zzNFpNTgS7sjQz5cBmRc3/CeUkH3CMqvzxdE9th3J4fYFt2S0+8XU79GMu/iwNYajHxrcnAJyGC\n3y53uMk0yCywuWinqKdP9LoQUHKAsD++QHbmE7lpjqaZEtECT5OBlJ73ih5tXgfY7R7y8+0UFTkp\nKnLRrFkk7doF9ouqqmDv3gIuuuhD8vLs/vRt226kSZPgj8kdO/KYM+db//a4ce1CCnthoZPDh8v8\n23l5djp3DrarLkylpaG92+oiHErYAcLDTdWE3UuExmyQ4dXiURyO0CKcmhpJ69axmEwyRqNMWJj2\nfRsTE8YFF3TAaJQxmw3ExoZe0nTy5K6ce24qsixhMMiaHz8VLF8+GUnyCbfBIAXVB0B+fj7ffLMA\n+BCArl178MwzL9C3bx/NPN9778KQx6tKu3bx/PDD1bWyPRXowt5IkI+UkTD8f/5tERsWIOwAIsyA\n0jQCz7kplM4dEDCMraGQJUv0TQhuZchweXim1EWTEH1gpxSvA2PxblDdWHcswnTke1SLrwnTWOxb\n9U5rOhYlshXFI5eiRLXVlwr9C5SVefj8870cPVqGxyMoLXXy0EPaTaGvvPIr8+dXzqZ4xx3n8OCD\ngwJsJAk6dkwICgBzubQFMzw88HVqt4cWweqCmZdXpmkX7N2Gns2vupCFKifAsGGtsNs9WCxGLBaj\nZl80wDnnNGPJkosIDzcSHm4kIcEaMs9XXx0X8reqNGkSyRtvjK+VbY8eTejRo3Yr7/XsWbOd3W5n\n+PDzOHLkMBaLhXvvfZCbbpqO0dj4ZLDxndEZiJTvIPr2NQFphsxipDwHktPrj35X2sdR8PXF9VHE\nv4VNguesZl6yBk7F2sar0l5RecTmpE19Na8LgTF/AzHfXIbsOhbSTGt2Nm9MJ4QhDE/KYMr6PK73\njYfghx+y2bbtKHl5dvLy7Fx1VXf69m0WZOd0ern11soIZpNJZs6cwZp919W9xIqVuKpSsV9srIUD\nB4r96S6X9iRE1T8A7PbQwlrdtqDAEdLuyiu7YTIZSEgIp2XL0PNH3HxzH/7xj16Eh/vEunpfdlVe\ne612IpycHEFa2vG7uhoCVquVK66YyoYNG3j22edp3brhD90NhS7sjQEJil8bRfjbO4l4dSsApu3H\nSBz6X/LXT6nnwv01jkgST0SEhRyadn+Zi7vt9bCQiqpgKNqBKX8jlj/ew5S/IaSpMISjhjdBLsum\neMRSVKsvQji+RXvyShv/o1dU5CQy0qwpMEeOlPHoo2txOLwUF/uawhctytDM54MPdrJkyQ7/dt++\nTTWFPT7egslUGUDl8ag4nd4g7xggLi5Q2IuKgoW9gpgY3wdlVJSZ8HATSoigy9TUaK6/vjfh4Uas\nVlONIjx37hBmzx6E1WoiPNyoOTYaIDLSzHPPjQ6ZT1WiosKIOrVTkp/WeL1eXn/9FVq2bOWfAvae\ne2ZhMBjqdYz5qaDxv10aK0L4PTyR4Jvu1XlZB7+wAxS+OwYR3fCm8zw7PiJkRPs/ix1McIf2hOoC\n2X4Yc/bnRP18V412rhbjcHS8Dk/yADCGbrLEEgWlDWdJUbdb4dgxOw6HF7vdi9ks06GDdmzGbbd9\nyZo1B7DZXNjtXtauvYpOnYL7RT0ehQ8//N2/3aZNbMjjJyUF1mXV/u6qSJJEcnIEhw5V1m1JiUtT\n2BMTraSmRhETE0ZcnKXG+bwXLx5DcrL1uGLQqlUMjz8+vEabCpKTG4cXfLqybdsW7rrrNrZu/Y2k\npGSGDx9BREREo2x21+LMOMtGhGF3AfEXLgeC1zRXE8JxXNYBx5WdUdqGflGeztwTGRYg6i0VlQfL\nXJzv8nLKIgKEiuQpIXL93Vj2/0/bxGhFGMLxJPenrNdslDjtccCnM6WlLgoKnBQWOsjMLGLs2Haa\nAUcbN+YwceKH/u3+/VP59NPLQuTp5ujRyv7i7OwSTWEP7mPWFmvQEnbt/miASy/tQlmZm+bNYzAY\ngo9TwZAhLdm8+fhDCgHNQDmd0xO73c4zzzzJq6++hKIoNG/egmeeWUhExJl1DXVhbyhorIdu2FuI\nt1eVcaVGGdvDA05xwf4+dqBrQiRl1YaGnLIpXRUnsiMPc85XhP++2Lc4igaeuB44utyCq82U07o/\n3OXyUlTkZN++IhISrCFnwRo3bgm7dlXGBaxf/w/atIkLsguOjA7dYlJ9Mo6srGJNu+qBZqWlbpxO\nr+aHRa9eKVxzTU+SkiJISrLWOJZ61qyBACQlRZGX13BaRXT+Pt99t4Z77rmDAwf2I8syN954C/fd\nN5vIyIY9pPevoAt7AyHm1m+C0sLf341Y+ge2R8+rhxKdPM5KCu4Y/LHAVvei7rGR8HEvZOdRzZ+F\nMQLJW0bh2G/wxveu12FnQgiOHi3j6FE7x445GDZMe1W9Tz7ZzfXXf+7fvvLKbiH7aKsHkB09WhZC\n2KuPZQ4d7V11yFRYmCHkGGWLxciiRRlYrUYiIkwkJlpDBnv1759K//6pIY+po+PxeLj33rs4cGA/\nnTt3ZeHCRZx9dt/6Lla9oQt7A8F2fz+k8pekIbuU6Ad+wLI8k7Ibu9dzyf46xRK0T6wU9QhV8FNh\nWZ1MICO5iwk78CmGwq2EZX+BoSwrpK2zzWRK+z9fcz/5KcTrVenY8eWAMcyZmbdqTlUZHR04cuDI\nkdDN1tWFveq46qpERJhJSYkgPNwX6NWqVeigsJkzB3DnnecQGWkmIsIccoIOWZa47DLtMd46OrVB\nCIHH48FsNmMymXj22RfYtGkjt9xye9BEM2caurCfphj+KELOLcMz2OepKB3Lm1MdXv/scUVvjMQz\nsOF5MvtkiXM1xqLvO2Y78cy8dmTHYWR3Ccaj6zEW/Y4wRWA8thnzkR9qlYWzzWRKB752ypvXs7NL\n+PrrfWzfnkdOTikDB7Zg+vRgL8NolImLCw8Q9qysYrp0SQqyrR7tffhwGTabW/MjoGnTSJo3jyIm\nxkJSkjWoL7uCFi2i2br1xlqdU2Li6fExpNO4yc7O4t5776J585Y888xCAAYPHsrgwUPruWSnB7qw\nn4Ykdfk3APm/aAxVUwV5W6aCqWFOYOKGIFG/1OnhpdLQw42CUL1IioPE9//aR407ZQje+J54mg7F\nnVq7oUQnQl6enYMHS8jKKiYrq4Trr++t2Xe8Y0ce9977tX+7epN3Vdq2jQ3or96/P5SwhyPLEnFx\nFuLiLJjNMoWFTk1hnz9/BPPnjzjR09PRqTcUReGf/3yNJ56Yh91eRmxsLLNmzSY+vuHNoFmX6MJ+\nmhE1a53/75hrV+PtEo8aH479tl6+RI1pHxsCR2SJMbHWgIj3wW4v7xc7qHFAnuLGeGwT1m0LMOd+\ni6SGnnlLNcUgjOG4Wl/qGz/uKsCTdA6epHN8K5sZw0/eCdVAevp7HDxYGbg1blw7zb7r6t51Tk7o\nFou2bePYuDGX1NQoEhOtmh8KAC1bRpOTc2e9zVGto1NX7NixnRkzbmPTpl8BOP/8C3n88ad1UddA\nF/bTCSHwtovFOeYsLF/sx7QtH9M23/zufmFvgPSNjyCr2rj0850e/q+qly4EYfs+QHIVYcrfgOzI\nw1i0A9mZV2Pe3tguFJ6/vi6KHYDN5ubjj3dx6FApubk2YmIsPPqodrNfs2ZRAcJ+8GCpprDHxwd+\naBw+HFrYH354CI8/Pvy4Y6krFqvQ0WksCCF48sl5vPTS83i9Xpo2bcb8+c+RkTG2vot22qIL++mE\nJOG4thvWFzcHJOdtvrKeCvT32GyUSa+28lo7r8IHxQ5aVAuQi/tsIMbC7TXmp4bFU3reKygx7VGi\n25308tZEQYGDGTMqV9lq3To2pLA3bx7FL79Ubh88WKJpl5xsZfLkrqSmRtGtWxKpqaGnDQs1M5mO\nTmNHkiSys7NQFIVrrrmeBx98mKgordUWdCrQ3xanIWpqZR903o6rTusx01rYJGiTGCxSOXmlmjec\nZdcbAaLuaHcVkupBDU/Gk3we7mZpYAjT2PNvltPmZvnyPWzadJjt2/PIyipm7twhmityVfeujxyx\nIYTQ9KDbt0+gc+cEmjePpmXLGNq2DfbWAWJiLLz4YvrJORkdnUZEUVEh+fn5tGvXHoB5857immuu\np18/7SVrdQLRhb0+EYLIuetRk8Kx31rZ1O4a0RLXiJaI2JMvZnXFboPMWrOBN8LN7K/W7P5CiYMp\nGitNxa4YETTXet6UHDCd3AklQk18cv31n/H11/sD0qo2oVclIsIUsIa13e6ltNQdNLwMYMaM/syY\n0f/vF1xH5wxDCMGnn37MAw/cS2JiIqtXr8VsNpOYmEhiYuglW3UC0YW9vlAFSd3eBiD/h8DpORuS\noLuBxVYzT0YEl7m9V2FdoT1oLXRDwVbiPxsUZF80+vOTJuqzZ6/ht9+OkJVVzOHDZezdewsxMYHB\naosXj+GBB9awdOkuf1rVecarIkkSEyd2omnTSJKTrbRoERO0VraOjs5f59Chg9x//wxWrvwCgNat\n21BYWEiTJrVbtlWnEl3Y6wHpmIPEwf8FwN0nmaiHf0KyeZALnRR+fH49l672ZBok+scHCvEkpweb\nJPGYzUmrin50IQjL/A+Rv85BduYH5ZN3RV6tm9qFEPz6ay4//XSIHj2SGTpUewa2X3/N5ddfD/u3\ns7JK6N49UNjj48N59dWxTJjQnptvXoHTqZCTE3oaUr3ZXEfn5KOqKm+99X889thcyspsREVFM2fO\nI1x11T+Q5YY5rLe+0YW9HghfspviF4ZhfWMb5l8rpzNVmjaMhQr+MEhMjrEGRbp/UVhGH2/5kpZC\nELHhXqy7XguZj6vl+ZQMe7fWx/3++ywmTapclOXyy7uGFPaoqMAPhQMHikPOMd6+fTz/938T6Ngx\ngaZNz7x5pXV06pOpUy9j9eqVAIwZM56nnnqWpk2Dl+XVqT26sNcD9lt6giRhXp+LaXvlIhwFX02q\nx1IdHwE00ZjX/e4yF/fb3SAE5kOriPxlJgbbfs083E3TKB34qn9t8qooisqffxYSHx+uOYNZ9QC2\njRtzQ5a1at+3JEF+fujVwzp0SAi5DKmOjk7dMnbsBLZu3cKTTz7rXzdd5++hC/spQCpwYvrlMO60\nFmA2+KPc1fIJSopfGYF7aPP6LGKNaA1bAzjP6eBfW96mVeaHmHO+1tjTR8l5r+BuOQ5h1l5Kdv/+\nIp5++idWrvyT0lI3Tz2VxjXXBI/bb98+HpNJxuPxtQrs2VNAUZEzaM5zgFtv7cvUqd1p0SKa5s2j\n9f5wHZ3ThPXrf2L//kwmT74CgMsvn8r550/Uh7CdRHRhr2PC39hG5MJNlMwf7BP1Ktin9wyIhj8d\neTXcxEORlcI5JPs7LtnzITfv+QSD/WDI/VzNx2Dr/zyqtWmN+e/Zc4xBg/4dkLZlyxFNW7PZQIcO\nCezZc4y+fZuSkdEu5AxrvXoFtwjo6OjUHyUlxcybN5d///ufWCwWzj13AK1bt0GSJF3UTzK6sNch\nFXO+A1iWZyLn2VFaRvs8d1k6bcen/2mQGBAfiawqNLcdZPFX87lm+5tYlNDTuZZ1vwdvQm/czUac\n0Kpo7dvHc+utfXnppY3+tFDCDvDeexNJTLTqHriOTgPi88+XM2vWPRw+nIvJZOKWW27X+9HrEF3Y\n6wjz14HLgprXHcK87hAAeZuugBBzfdc3Ow0yw+KsdMvbxra3e4S0c7SbhjBF4m4xDk+T80AKHb1a\nVubhp5+yEQJGjWoT8JskSTz00BCSkiJ4+OHvAEhNjUJVhaY33qxZ6NnZdHR0Ti8OH87l/vvvYcWK\n5QD06dOPhQtfolOnzvVcssbN6akujQD30Obk/zwF+aAN085jRM35EYDCJWNPS1EXwIfAV9nvIt66\nKfA3oxXJa6d46Lu4m3LSZsYAACAASURBVI8GQ3CfthaZmYU8+OAafvghG6dToXfvJkHCXsENN/Qm\nLs7CRRd10r1xHZ1Gwt1338ZXX60iIiKS2bMf5uqrr8Ng0J/vuub0U5jGglFGRJlRWkYRP8n3tZr3\n25VB/eynA+9ZTNwVZeGs4n3s+ypQ1Mt63Ie914N/Kd+4OAvffLMfUT6c/bffjlBQ4AiKbgcwGGQm\nT+76l46jo6Nz+lB1quW5cx8nLMzCY489RWrq6Rsg3NjQR/+fbOwewr7YV7ltkMjbMpW8ndNOS1EX\nwF1RPg988dfT/ekFEzeRd1XJXxZ18K0N3rVr5ZrhQvjGouvo6DQ+3G43CxbM5+qrr0CUf8136NCR\nt956Vxf1U0ydeuxPPPEEW7ZsQZIkHnjgAXr0qOyzfe+99/j000+RZZlu3brx4IN/XUBOFwyZxcSP\nXwZA3pjWvsTTsNm9guLi3Ww48A47slbTpeB3f7q76fCTtnrawIEt2L49j9TUKEaNakPr1tpD3nR0\ndBouGzb8zIwZt7Nrl+898uuvG+jb95x6LtWZS52pzi+//MKBAwf44IMP+PPPP3nggQf44IMPALDZ\nbPzzn/9k1apVGI1GrrnmGn777Td69Tq9h34djwpRVxN8HrBkcyMizfVZJE0kx1FiV08gqeh3tOS7\neNh/TtqxrruuN1dc0Y0OHRJCDk3T0dFpmNhspTz++CO8+eYbCCFo3boNCxa8qIt6PVNnwv7TTz8x\ncuRIANq2bUtxcTE2m43IyEhMJhMmkwm73Y7VasXhcBATE1NXRTklVB3aBmD5zy7CvtxP8dsZ9VSi\nanhsJL2vPbzkYHRrTN1m0qTTSPK8J3f8d6tWDfu66ujoaPP5559zww03kpNzCKPRyPTpd3D33fcS\nHh4cQ6NzaqkzYc/Pz6dr18pgqPj4ePLy8oiMjCQsLIzp06czcuRI/p+9+w6PouoCOPzbkt5DEnrv\nvXekq4BIL0EUpIjSe/2QIj2QUBURBUEBQQQUpAnSu3RCbyHUJKT3LfP9sWE3S0g22ZRN4L7P4+PM\n7JSTMebsvXPnXBsbGz766CNKliyZXaFkvwTdVJ6SUjdkQf4yHqfZZwg51sOSUekpwq7hvrNRiu2J\nciuKD37JlVfznrg5QXDqk6CYcuJEIFqthEIhQy6XU6dOQZRKMYxDEN5Gp06d4unTJ9SsWQtf3+VU\nqVLV0iEJSXLsAfCrwRSg64pftWoVe/fuxdHRkb59+3Lz5k0qVKiQ5jk831CnPNcIHqabnjRODcW+\nBwcrPCq8edKRHKFOgNCb8HsriDfUo7/tVpYK/W4iJb13fh/wTPb2Wmbucd++fxEZaShiExY28Y3l\nXoVc/rv8lhD3OGtJkkRAQAAlSpQAYOrUqZQoUYJ+/fqJV9jSLRTI/t/NbEvsXl5ehIQYpugMCgrC\n01M3QvrevXsULVoUd3d3AOrUqcO1a9dMJvbgTLQmc4RWwmniMehYmqh5TTLV+s0MWUIo7ttrIk8M\nM9q+pNZIRjdfTFW1licKLYfDYnHUSgQnfe7p6ZSpe6x+NbNbkrCwGFQqldnne1tl9j4Lpol7nLXu\n37/L2LEjuX37FidOnMPV1Q1PTyc6duxJaGjqEywJb5aR301zvgRkWz9p48aN2bdPNxWfv78/Xl5e\nODrqpsQsXLgw9+7dIz4+HoBr167pvwXmJdZ7HxpXmJPLiFrYVJfULUWrxmV/B6Ok/m31IShHq7Ct\nPY+gkGgOhsdy82UMBbRSGifKuOS9MoAYLCcIeZxKpWLpUl+aNWvIiRPH0Go13Llz29JhCSZkW4u9\nVq1aVK5cGW9vb2QyGdOnT2fbtm04OTnx/vvvM2DAAPr06YNCoaBmzZrUqVMnu0LJFna/3MBx3llC\nzn1i6VD0rF6cxHWfYbBerNIOh5G6b9MjYxPomqDO1us3bFiEhAQ1Go2kf9YuCELedPHieUaPHs71\n69cA6NGjFzNnziVfPjHFcW4nk15vZuViualr7dUo+IQWRdCUdEFVJz+JzYtaLiBNPJ4bjJ/pF/si\ngEDnYvwaEcsHiRqTpxDdlzlD3OfsJ+5x5qxYsZTZs6ej1WopVqwEixYtoXnzlkb7iHuccVN+1z2a\nnNvdKt3H5Kqu+LeZ7GWcftnm0GPs1/jjPOyQxeJx2feRUVI/0uwnZGMlAp2LUUatSVdSFwRBeKVq\n1WrIZDKGDh3J0aOnUyR1IXfLvWXRcjGHJReJGVET5c1QbPYHABBy6VOLxGJ/8RusXxzTr8eV/pTW\ntfvp10+GiYEtgiCkLTg4mIMH9+Pt3RuAZs1acPbsZYoWLWbhyARziMRuhpiJdZAcrXEeddiwMYff\n13Y8ORSbgD+RqyL124I/ecFvDk6okyZgmB4dn6MxCYKQt0iSxObNG5k+fQphYWGULFma+vUbAIik\nnoeJxG6GV2ViFbd0I8/DtnyUY9eWJUbisr89VqGXjLaHeD9ij70TI5wNVZ8GxYlXzQRBeLMHD+4z\nfvxojh7VPUZs2rQF+fPnt3BUQlYQiT0TohY0QVPSBckph+rBSxIevxnPkhTeegfqfDWRrF0Z7Gwo\nBnMqNJr0D88QBOFdoVar+f77b1m4cC5xcXG4ubnxzTfz6NGjl366VSFvE4k9A+TBsWjdbUGh63ZX\nV/M0cUTWsrvxrX5Z7VqZiNbb0drrarvPdrAmNul/ytWRcZTW5OzLDrGxKoYP34tCIUcul+HgYIWv\n7/s5GoMgCKYtXDiXxYsXAdC1aw9mzZqPh4eHhaMSspJI7Olku+U2TjNOEXLSG8nVJmcvrorBc1NB\no01hHU7pl+8pZCyzN8T0UTa/r/4miYkadu68o193crIWiV0QcqGBAwfz778HmTTpf7Rq9YGlwxGy\ngXjdLZ2cZugSqXu77bh6/43j1BM5c2F1bIqkHtrxP/3yA7mMhu6O+vXrIdEW+bamea2HQBSnEYTc\n4fDhf+nTp5e+vLOnpyf79x8WSf0tJlrs6SB/EWNYDk9AHp6A1ZUQomc3zr6LShKuez/AKviMfpPa\ntSJhHc4Y7VY/nyGp+0XF42GhekMaja5O/AeDP6ZYlRKAoRiD8LpQSwfwDhD3OD46lDPbpnLnzGYA\nuo9ZQ6Wm/bPwCuIe51YisaeDJJcT/N8nuHXdhTJA93rZy/1dsu+Cb6giF1d+INH1/Yy2zXAwdL8P\njE3k03jLJVJnZxtWr/6IM4klLBaDIAi6V9ju/fcHp7dOIT76JQorW2q1G0+Fxp9ZOjQBKF8g+3sz\nRWJPB8lT9wqZuqqHPrFri2TDtHtaNdaBf+NyxPh/wJCeAUg2bkbbvnSyZbutYdz7nJgELMnWVknH\njuU5Y0bJxHeNKMWZ/d7VexwY+Ijx40dx+N8DADRp0pRFi5ZQqlSZLL/Wu3qP8wKR2DNAXdkddt0n\n5KR3lp/b9taPOJ0ZY3w9t6qEfZzyWf4jucwoqd8OiUI80RYE4ejRw/z77wFcXFyZOXMOvXp9Kl5h\neweJxG6KRqt/vS2uTyXi+lbO2vNLEk7Hv8D2wRajzfHFuxDV7Gejbb/aWjHGydZo2/3gKBwRBOFd\nFRERjouLKwCffPIZz58/49NPPxfFZt5hIrGnQnEjFPeuO3n5T1e0hZNSZxZ+85VH3cf5aH+sXl4w\n2h720THU+aqn2P+uQpYiqY+ITbB4UpckicDASPbtu8eDB+FQ8z0LRyQI74a4uDh8fRewdu2PHDx4\njBIlSiKTyRg7dqKlQxMsTCT2VLh33YmmgAOStUK3QZKyLLErQq/ivivliPrQj0+jcatktO20UkEH\nN3ujbasi4+iYoLbou4qSJLFgwUl+/vkyoaG6mvQ2Ngo+E4ldELLd8eNHGTt2BA8e3Ecmk3H48L98\n/vkAS4cl5BIisb+Bze4HACiex+DRTNdFHvdJeaKnNsjUeWVxQdhf88X+xkr9NpVnA2JqTkVVoGmK\n/f+wUTI4We13gLURcXyUmPMFaF53926YUVIHSEgQ08MKQnYKCwtl5syv2bjxFwAqVKiIr+8y6tat\nb+HIhNxEJPY3cB53NMW2mJG1MnVO170fYhV0ymhbZONVJJTuleoxi+0NNeg3RsTSMlGTI610rVbi\n0qXnnDnzlOPHH1G1qheTJhn3MJQt6865cwP47rvzrFx5nthY8c66IGSno0cP89VXAwgJCcba2prR\no8czfPhorK1zaK4KIc8Qif0NEhsURBajQv48BkVwHECmJnpx2dfOKKlrbT2JaLkZtUedVI+RgNtK\n3WOAcTEJtE7MmdZwREQ83br9weXLL/TbLl16wbhxDVG+NjWtk5MNEyc2onhxF9asuUSrViWJyJEo\nBeHdkz9/ASIiwqlfvyF+fsspW7acpUMScimR2N8gYo2u1KLNngc4jz1KyMmeZp/L5sFWrF8c168H\nfxoKctO3vZ2r4bl63xwsPBMYGEXBgo5cuxakLxMbHBzL4cMPad26VIr9JUmiR49KeHvr3hYQ1eYE\nIWtoNBr27t1Nu3btkclklC9fgd27D1C1anXkclENXEid+O1IgySXEbqzI5Krremd30AR5o/zMUMJ\nx/Qm9TAZnLdS6Nfza7OnTKxWKxEVZVzYpkoVT9av78ilS1/wv/810bfS9+y598ZzyGQy5HLxnqwg\nZKUbN67Tvv0H9OvXm+3bt+q3V69eUyR1wSTRYn9dstHviR+WMPs08thnuO9sqF8P7fhfupI6QHkP\nQ1W7myHZU9np6NFHjBv3D/XrF2b58jYpPs+f35FatQowYkRd6tUrRJMmxbIlDkEQDOLj41myZBHL\nly9GpVKRP38BnJyyocql8FYTiT0ZhwXnsFt/nRD/vuafRNLi+Yur0aaIFr+hcUnf87DkbfMmiWrc\ns6Gx7u29jX//fQiAnV3qvwLvvVeM994TCV0QcsLp0ycZM2Y4d+/qpj/u23cAX389A2dnFwtHJuQ1\nIrEnUdyPwH7ddcMGc95blyScDxmXm42pOo7Eou3SfYovkhWh+TUiLmPXTyeVyjAQ79atUGJjVdjb\ni9rugmApe/b8Td++ujdkypYth6/vMho0aGThqIS8SiT2JO7tdwCgKeQA8Wrs1/gTOyRlBbhUSRKe\nvxh/sw7uE5mhGMY62vBXshrw9mnsa0paydrT00G/rNVK3L0bSrVqovykIFhKixatqFixEu3afcyo\nUeOwsbExfZAgpEKMwniNLEGDR8PfcFhxKUPHOVycabT+suuNDF/7FzvDK3VXQ6IzfPwrf/99h5o1\nVxMW9uYWv5eXLrG7utqwYUMnkdQFIYc9f/6M0aOHER4eBoCtrS0HDhxj4sT/iaQuZJposQOotfpF\n+UtdJbX4TqXTf7wmHvtrhrnSgz+LyHA3fmyy5T1hMeSXMv5wXZIk1qy5zOTJ/wKwffst+vevkWK/\nfv2q06NHJcqVc8faWpHic0EQsodWq2X9+rXMmjWdqKhIbGxsmD/fFwArK/E4TMgaIrEDyCBsQ1vk\nL+OwPvEUu823iZrR0PRxSayfHtIvh3Y6b1ZN+W+TVZmrnuyLRkZs3nydKVP+1a9v2XKdfv2qp5i2\nsWRJ19cPFQQhm92+fYuxY0dw5oyuWFWbNu0YMWKMiaMEIeNEVzyAQo66pheJrYsTM7Imobs7QwZa\nsi6HdAVs1E6l0DiXNSuE+KTcW0yjNfvbVqtWJenRwzCJzIULzzl58rGZZxMEISskJiayaNF8WrZs\nzJkzp/Dyys9PP61n3bpNFCpU2NLhCW8hkdhfI7naoinhnO793bcbBtipCrU0+7rL7XXP1ZplYoIX\nT097OnY0vFY3eHBtGjUqYvb5BEHIvPPnz+HjM5fExEQ+/bQvx4+f5eOPO6XoSROErJKuxmFYWBiP\nHz+matWqaLXat6ryke2GGygeRREzuV6Gj7V6+i+KKN1McJLCjuj6fiaOeLMnySq3tUpnTfjnz6PJ\nn98hxR+Hpk2LU69eIbp0qfDGbnhBELJfYmKifnKWhg0bM2bMBN57rxmNG4tpjYXsZzJD79q1i549\nezJ58mQAZs2axe+//57tgeUEm213cJpzFsXjjFd3U0TcxvVAJ/16SC/zu7zbJKsL385Ei/3Fixjm\nzDlOnTo/cfVqUIrPra0V7NrlTf/+NURSFwQL2Lt3Nw0a1OTs2TP6bZMmTRVJXcgxJhP72rVr+fPP\nP3FzcwNg4sSJbNmyJdsDywnOU08CICU9T5cHxaa1u5486j7ufxpmZouuMxfk5o1o1QIvFLr/DB1N\nTPby00+XqF37R5YuPUtioobdu++adU1BELLeixcvGDCgD336ePP4cSA///yjpUMS3lEmE7uTkxN2\ndnb6dVtb27fjtYx4Q8tYJoFb579wWPif6eO0GvJtN7xCFl13PnGVhpkdRhcXw72dF52Qxp4QGBhB\nYrKuepHYBcHyJEni11/X0aRJXXbu3IG9vQOzZ89n+fLvLR2a8I4y+Yzdzc2N7du3k5CQgL+/P7t3\n78bd3T0nYstW1kcMXec2+wMACP/5Q5PHOfw3Wb8c+d5aEkp2Nev6KuADN3v8lYbR9x4m3l2fOLER\ne/fe4/79cAAePYrg2bMoChYUk0QIgiU8fhzI8OFfceLEMQBat/6ABQv8KFpUzLEgWI7JFvvMmTO5\nevUqMTExTJ06lYSEBObMmZMTsWUrTSkXYvtWMipEI7mkXfHJ9uZq7G8avoWbm9QB1ttaGSX1h8Gm\nn/Pb2VmxZMmH2NgoGDOmPqdO9RNJXRAsyMbGluvXr+Hh4cH33//Ehg2/i6QuWJzJFvuxY8eYNm2a\n0bZNmzbRq1evbAsqJ2jKuhEzsS7EqVHeCCXSx/TAFqezY/XLLztfydT1Jyeb7OVhcJS+LrwkSRw/\nHpjqrGoNGhTm0qVB5Mtn98bPBUHIXleuXKJixcpYWVnh6enJunWbKFeuPO7u+SwdmiAAaST269ev\n4+/vz5o1a4iLM9QcV6vVfPvtt3k+setZyQnb9nGa1eIUEbeNBsuFv/8XWqcSZl9yeLKkPiI2QZ/U\n4+JUtG27ievXQ1ixoo1RsZnkRFIXhJwXHR3N/PmzWL36e6ZMmcbIkbov+mIWNiG3STWx29jY8PLl\nS6Kiojh//rx+u0wmY8KECTkSXI5Qpv00wvlQL2wC/zbapirQ1OzLaYHNyWZw+19MIgD374fx6ac7\nuHtXNynEqFH78fCwp1GjItjaisq/gmBJBw7sY8KEMTx+HIhCoTBq7AhCbpNqxihdujSlS5emQYMG\n1KhhPJHIvn37sj2w7GR98BFK/5fEjqiZ5n4yVZRRUo+tOJSYuvMyde0CnoZn4g+Co3jVT+DsbINC\nYfiSoVZr8fbexq1bg0ViFwQLCQ4OZurUCWzf/gcA1arVYPHi5VStmoEpnQUhh5nMGF5eXvj4+BAW\npmtJJiYmcubMGT780PQI8txIHhCJy/BDvDzYDV69OpZKXXhZYoR+OaTXEySrzA1UG5asC76eSo1D\nss88POzZsKETbdtuIjg4FqVSzvLlH+LmJrrdBcES7t+/S9u2rQgLC8POzo6JE6cyaNBglErxRVvI\n3UyOip8wYQKurq5cunSJKlWqEBYWho+PT07Eli3sNtwEIF+rrXjW+JV8LUxX0dPYF850Uvext2ZL\nsi74XeEpu/KKFXNhw4ZOVKzowYEDvenatWKmrikIgvlKlChFxYqVadasBUeOnGbIkOEiqQt5gsnE\nrlAoGDRoEB4eHvTu3ZuVK1eyYcOGnIgtW9j/egMAKakPXB6WelEYZZh/llwzWgaLHAyv0j0KjkJK\n5Z31GjUKcORIHypV8sySawuCkD66gcHLePhQN/+DXC5n/fpNbNmygxIlSlo4OkFIP5OJPSEhgefP\nnyOTyQgMDESpVPLkyZOciC1baAo78vJgN2RJeTW2d4VU93X5tzsAitjM/bwt3Ayd7leDovj1x4t0\n7vw7KlX6JnwRBCF7XblyiQ8/bMHMmVMZP36U/ou3s7OLmHNByHNM9isNHDiQU6dOMWDAADp27IhC\noaB9+/Y5EVu2CN3XBZLNppbQpsQb95PFBeuXY6pl7i2AgKRBcWUS1FTNb5gBzs/vDBMnildlBMFS\nYmJi8PGZy6pV36LVailatBiDBw8XyVzI00wm9tatW+uXz549S0xMDC4uLtkaVLaSG/8Pq66dP+U+\nkoTTycH61djqU8y6VLBMRmUPR/36suBo2iX73Nf3NM2bF6d+/cJmnV8QBPMdOnSQ8eNH8+jRQ+Ry\nOV9+OZSJE/+Ho6Oj6YMFIRdLtSteq9Xy22+/MWvWLHbt2gWAUqnE2tqamTNn5liA2SVybmOCL36a\nYrsi/Caev7hg82Q/AFobd5CZN//8HAdro/U6Ngq6dTMeEBccnL4Z5QRByDrPnz/js8968ujRQypX\nrsqePQeZNWueSOrCWyHVFvusWbOIiIigRo0a/Pbbb4SFhVGmTBmmTZtm1IrPM+LVKB5FoSmnm342\noVOZFLsoX17E7e9mRtvC399p1uUkYKOdLrFXUGv4N0yXwDt0KMfWrTdwcLBi2bIPad++rFnnFwQh\nY149N5fJZBQoUJAJE6YAMgYPHvZ2zFgpCElSTew3btzgt99+A6Bbt260aNGCwoULs3jxYqpUqZJj\nAWYJScJ14D/E9q+sT+xv2id5Uo8v5U1Ukx/MvuQhK8O78T9HxOlvdIsWxZk3rwW9elXB3l78MRGE\nnPDoUQDjx4/C27s3nTt3A2DEiDEWjkoQskeqiT35N1h7e3tKlizJhg0bUCjeXMzlTebOncvly5eR\nyWRMmTKFatWq6T979uwZY8aMQaVSUalSJb755hszfwTTlFdCsLoQhK2LNYrH0WgKOJD4QXGjfexu\nfKtfjq49m7jKIzJ1za3J3lkvpTW82mZjo2TAgLQr3gmCkDU0Gg2rV69k/vzZxMbGEhj4iI4duyCX\nm/d4TRDyglR/u18fFWptbZ2hpH727FkCAgLYvHkzc+bMSTHV6/z58+nfvz9bt25FoVDw9OnTDIae\nflZnnwNgc+gxjvPP4TLqcIp97K4bEntmk/pVpVyf2JslqjN1LkEQzHPt2lXatm3JtGlTiI2NpVOn\nLuzYsUckdeGtl2qLPSgoiK1bt+rXg4ODjda7deuW5olPnTqlfxZfunRpIiIiiI6OxtHREa1Wy/nz\n5/Hz0736NX369Ez9EKY4Lr6Q5udWzw7r31WPMXMEfHJnk82zPi069QI4giBkvfj4eCZPnsvChQvR\naDQUKlQYHx8/PvigraVDE4QckWpir1mzptGsbjVq1DBaN5XYQ0JCqFy5sn7d3d2d4OBgHB0dCQ0N\nxcHBgXnz5uHv70+dOnUYO3ZsGmfLHE1+exQvDKPPNV6G+usOF2Zif81Xvx6bydY6wMak1vpncYlE\nHHlIB59TyOUyFAoZjRoVZezYBpm+hiAIbyZJEn/88QdarZaBA79kypRpODpmriS0IOQlqSb2efMy\nN4vZ65KXUJUkiRcvXtCnTx8KFy7MoEGDOHz4MM2bN0/zHJ6eZv7Pea2/bsKXfQ+g/14US1sbzpUs\nqdN5F54F3/BeewZIwNWkZYWdNYmJEqdPGyrXFSzobP7PkQMyH1toFp3n7SbuT9YKDQ1FoVAk1dhw\nYt26dchkMho0EF+is5P4Pc6dsm1GAy8vL0JCQvTrQUFBeHrq6p+7ublRqFAhihUrBkDDhg25c+eO\nycQeHByVqZisVGqU4+sQV80NgqOwenES16TPQrrfQbLLD5m8xkkrBbjaA/BpWAx3X5vsRaVSZ/rn\nyC6enk5ZFltu/Rlzg6y8z+86SZL4889tTJkygbZt2+PruxTQ/U0JDo4S9zkbid/jnGHOl6dsG0XS\nuHFj/bzt/v7+eHl56Ys/KJVKihYtysOHD/WflyyZ/ZMsqN4rTFy/ypA0MFAe/VD/mWSXuZa6Bhjp\naEunpKQOUE2tRas1nuxFlKoUhKzx+HEgn37ag0GD+hESEsy9e3dITEy0dFiCYHHZ1mKvVasWlStX\nxtvbG5lMxvTp09m2bRtOTk68//77TJkyhUmTJiFJEuXKlaNly5bZE4gkIQ+JQ+tp/4YPdUk2vlTP\nTF+mhZs9N5MNmlsWqWupN2tWnB07uqPVSmi14OEh5lcXhMzQaDSsXbuaOXO+ISYmGmdnF6ZPn0Xv\n3n3EiHdBIB2J/ebNm0yZontdZO/evXz77bc0adKE6tWrmzz5uHHjjNYrVDDMpFa8eHE2bdpkRsgZ\nIEl4Vl4PQPD1vik+VkTdy5LLXFHKjZL6mZfRlExqqXt62uP5xi8VgiBkVFxcHF26tOf8+XMAtG/f\nkXnzFpI/fwELRyYIuYfJr7fffPMNc+fO1T8fb9euXZYPrMsubp115WA1nkmt5MRk06RqNThc8QFA\nHh/y+qEZ0jr5tKzJkrogCFnLzs6OUqVKU6BAQX7+eSNr1vwikrogvMZki12pVBq1tEuWLIlSmW09\n+FlKeTsMAMnJGnlAJC7D/iVsZycAPDYV1O8XX+Yzs6/hlWxgg19UPPlFUheELHXq1AlsbW2pWbM2\nAHPmLEAul+PsnIdnmRSEbGSyxa5UKgkMDNQP+jpy5IjRq2u5VrIYFY8iydd2O5qiuiRse2cdMk08\nAGrnsiQU72zWJU4rjSvx9Y5XmRmsIAivi4gIZ+zYkXTs2JaRI4foB8a5urqJpC4IaTDZ9J44cSJD\nhgzhwYMH1K5dm8KFC+Pj45MTsWWK8kaoflmm1iX5yKUtAHA6NVz/WViHM/pR8hnVwc3w7PxpcBRi\nvLsgZJ4kSeza9RdTpoznxYvnWFlZ0b59R0uHJQh5hsnEbmVlxc6dOwkNDcXa2jrPzFesruhO8IXe\nyOI1OE0+jtX5F2AlB0mr3ye0w1mQm/dYIUBuSONDYhNTvZFhYXGEhMQhl8uQycDFxZZ8+cTIeEF4\nk2fPnjJx4lj27v0bgLp16+Pnt5zy5SuYOFIQhFdMZrXBgwfj5OREhw4daN++fU7ElDVkMrBVIsll\nqKt5ED25LgCu1kdT9gAAIABJREFUez/U76JxKmX26Qc6G5Lz9JjU68Fv23aTyZMP6df796/O/Pmt\nzL6uILyt1Go1H3/8IY8eBeDo6MTXX8+kb9/+4hU2Qcggk4l93759XLt2jT179uDt7U3JkiXp2LEj\n7dq1y4n4Ms9aQezgpFfztGqsgs8AIMmtQWFt1ikD5DIuJ823XlelSbML/vUCNXK56LAXhDdRKpWM\nHTuRPXv+Zv78RRQqVNjSIQlCnpSur8JVqlRh/PjxbNiwgUKFCjFhwoTsjitbOB/tp19+2fW62edZ\nbG/4QrA+Ii6NPUGrNV4XiV0QdBISEli4cB4rV67Qb/P27s26dRtFUheETDDZYg8KCmL//v3s3buX\n0NBQ2rVrx99//50TsWUtScLm0Z8AaK3dkOy8zD7VI4Xu+1BxjZZ8Jt4QcHGxoUwZt6TKcxL58oli\nNYJw9uwZxowZxu3bt7Czs6N7d288PDxEyWVByAImE3vXrl1p164dEydOpGrVqjkRU6bJguNwmnOG\n6El10RZ4VTzGkIBDO/6XqfMft9bdtiGxputSe3tXxtu7ssn9BOFdEBUVyezZM/j555+QJIlSpUrj\n67sMDw8PS4cmCG+NVBN7UFAQXl5erF+/Xl+QJjAwUP950aJFsz86Mzku/A+b/QHY7A9AslGQ0Koo\nMTNKACAhQ7LzNPvc/grD04syGm0aewqCkNyePX8zadJYnj17ilKpZPjwUYwePQFbW1tLhyYIb5VU\nE/uCBQvw9fVlwIAByGQyo6I0MpmMgwcP5kiA5pBFGkapyxI0WJ17gfPxubp1MldcZ5qjjX65sUqT\nxp6CILwiSRLff7+CZ8+eUqtWbfz8VlCpkujJEoTskGpi9/X1BWD16tWULl3a6LOLFy9mb1SZZHP0\nidF6fJcyOASdAiChSOZG8x9L6oZvmKjOvjlvBeEtoNVqiY6OwtnZBZlMhq/vMg4dOkD//oNQKBSm\nTyAIgllSzU2RkZE8evSIKVOmEBgYqP/n/v37TJo0KSdjzLDYgVVIbFwIVSV3AOQKf8NnVUebfd7b\nybrhx6fj+frp00+YP/+E2dcThLzq7t07dO78EQMG9NH39pUpU5YvvhgskrogZLNUW+wXL15k3bp1\n3Lhxg759DVOeyuVymjRpkiPBmStmjG6yCMXtMFz77UOqdRHu6j5Te9Qz+7wtk5WQbWKiG/6PP24w\nePAeABo1KkrTpsXMvq4g5BWJiYl8++1S/Px8SEhIwMPDg8ePAylaVPz+C0JOSTWxN2vWjGbNmrFp\n0yZ69eqVkzFlHRmE7uiIk/8uAGJq/M/suvCzHKxJTDq2fULak72EhMTqkzrA9OlHOHCgNwqF6LwX\n3l7nz59jzJjh3LihqxHh7d2bGTNm4+6ez8KRCcK7JdXE/scff9C1a1devHjB0qVLU3w+cuTIbA0s\nK2jKuiW9v/4XAJLMvLrw4TJYbm8YNPdTZHya+/v4nDJa9/cPZseOW3TtWtGs6wtCbjdjxlRWrlyO\nJEkUL16CRYuW0qxZC0uHJQjvpFSbkK/qMyuVShQKRYp/8gr7S7P0y6oC5j1COG5l+EJw/mW0yVnc\nFixoycqVbfHw0NWTb9asOF26iEkshLeXnZ0dcrmcYcNGceTIaZHUBcGCZFI6JlePjo7G0dGRkJAQ\nHj58SK1atSwyMUNwcJTJfWTRicgiEtEW1s1C57a9Bsqo+7rjP4swqyveyzNpHndJ4lFIdLqPCwuL\nY9So/Sxb9iEuLrn/XV1PT6d03eO0TPld95hibnerrAjprZQV99nSgoKCePDgPvXrNwB05WHv3LlN\nlSq5o4jV23CPcztxj3OGZ1L+yQiT2XnWrFns2bOH8PBwvL29+fXXX5kxY4Y58WU/rYRHvU04LvwP\nq9PPsDv2iz6pR9VbZPbz9Vc6JagztL9SKWflynZ5IqkLQnpIksSmTb/SpEkd+vXrTVhYKAA2Nja5\nJqkLwrvOZGK/fv063bt3Z8+ePXTu3JklS5YQEBCQE7FlmM1f93T/3h+Aa//9OD4Yqv8svsynZp0z\nONmXgblRKZ+tX70axOnTj994rJOTDfb2ouUqvB3u379Ht24dGTlyCOHh4VSpUpX4+LTHmwiCkPNM\nJvZXPfWHDx+mZcuWgO6VltzIwe+CYcXKUH0uotkvoDRv8pUFDoaZ3Bxe++z8+We0avUr/frtZO/e\ne2adXxByO5VKxbJli2nevCHHjh3G3d2db7/9gc2bt1OwYCFLhycIwmtMDhMvWbIk7dq1w93dnYoV\nK7Jjxw5cXFxyIrYMk9xtISRpGtWuP+i3Jxb72Oxz7rTRtbjbJaiMBs1FRMTzySfbAXj5Mo4+ff6k\nS5cKzJnTgnz57FKcZ90xNbeeZ66cbfYLtXQAQi701VcD2LlzBwDduvXkm2/miUlbBCEXM5nYZ8+e\nze3bt/VlZcuUKYOPj0+2B2aOyCXNkD+PRXn1CY42tw0fyMwf6GcrSYCMNq89X7ezs+KLL2oavdq2\nbdtNli//8I3nyf1JPeuULyCm3nyb9Os3kMuXL+Hj40fLlq0tHY4gCCaYTOzx8fH8+++/LF26FJlM\nRo0aNShTpkxOxJZhmhIuaEq4IFMGgK5GBsGfPDf7fBLwLKmozOszuVlbKxg4sCbffXee6OhEWrYs\nwYoVbbCySvtVwNw8WlyMchUADh06yPnz5xg3Tlc6ukmTppw6dR4rq9z7uysIgoHJpuzXX39NdHQ0\n3t7e9OjRg5CQEKZOnZoTsZlN7ngFALVLebOfrQPcUxhantXUKadodXW1pX//6jRrVpy1az/Gw8P8\nawmCpb18+ZIhQ76gZ8/O+PjM5fz5c/rPRFIXhLzDZIs9JCQEPz8//XqLFi347LPPsjWozLJ+qptS\nVp6QuWfGH7saErV1KvuMHt0Aa2u5yZa6IORWkiSxdetmvv56EqGhodja2jJ+/BSqVath6dAEQTCD\nyRZ7XFwccXFx+vXY2FgSEhLSOMIyZCFxuAzYjzwgEkWM7vWz6NqzzT5fpAxeJhXhKa7RotFo2bz5\nOq/X83FwsBJJXcizAgIe4u3dhaFDBxEaGsp77zXj8OFTDB8+SrTSBSGPMtli79mzJ23btqVKlSoA\n+Pv758o68R5NtwCgiLiNMuwqAJKVo9nn+93G8EftYGA4HXr8wblzT6lduwBlyrhnLlhByCWWLFnE\noUMHcXV15Ztv5tGz5yfIMlnISRAEyzKZ2Lt160bjxo3x9/dHJpPx9ddfkz9//pyILf1Uhuffjre+\nMmwu0NTsU/ole3+9+Xs/8/ixblDZyZOPKV3ajYQEDba25k0qIwiWpFKp9K3xqVNnIpfLmThxKl5e\nXhaOTBCErJBmZjpy5Aj379+ndu3atG6de19zsbocrF9Wxulec4sv3RvJ2vz37YOTuuFLLT3N/ceG\nkeKnTj0mMjKBzz+vbva5BcESYmNjWbRoPocP/8vevf9ibW1Nvnz58PVdZunQBEHIQqk+Y1++fDkr\nV64kKCiIqVOn8tdff+VkXBmivBCUtGRoucdUG2/2+ZKPhv9AJlG0qLN+/Y8/blKjRn4cHVMbTicI\nuc/Ro4dp3rwhK1Yswd//KsePH7V0SIIgZJNUW+zHjx9nw4YNKJVKoqKiGD58OB06dMjJ2NJN62WH\nupwbsoLneTWMTetQzOzzTXUwTNryTa8qTOteid277zJw4C62bu1Gkybmn1sQclJYWCgzZkxl06Zf\nAahYsRJ+fsupXbuuhSMTBCG7pJrYra2tUSp1Hzs5OaHRaHIsqIxK6FSGhE5l8FyfbKIXufnPvw/a\n6I4dGquria9Uyvn447Jcv/6VeFddyDP+/nsn48ePIiQkGBsbG8aMmcDQoSOxtha9TYLwNks1+70+\nMjbXj5RVGeZJj6rna/ZpQpL9nL3jDZPdyGQykdSFPCUiIpyQkGAaNWqCr+9SSpcua+mQBEHIAakm\n9nv37jFhwoRU13NbvXiHy/P0y/EVvjD7PFeVhmEHZTTvTn13Ie/TaDRcu3aF6tVrAtCr16e4ubnz\n4YdtkcvNny9BEIS8JdXEPm7cOKP1hg0bZnswmSEpbABQ5auZqfOcSSo2I1fl3kcPgvC669f9GTNm\nGNev+3P48ClKlSqNTCajbduPLB2aIAg5LNXE3rlz55yMw2yy0HhkGi0kta4Ti2buD5mfg+4LgvZq\nEEfuv6RZs+KZjlEQskt8fDx+fj6sWLEEtVpNwYKFCAp6QalSpS0dmiAIFpLn++ecRx0mX7PfUTyL\n0W3Qmt997i9PNo7gz1uMHr2fgIAItJk4pyBkl5Mnj9O8eUOWLFmEWq2mX7+BHD9+lgYNGlk6NEEQ\nLCjPJ3br/14AYLv7LgDKqyFmn6tFvmQlaOcc5fHjKNasuYRcnssHDgrvnFWrvqVTp3bcv3+PcuXK\ns3PnfhYs8MPJydn0wYIgvNXSldjDwsK4elVXf12rTTl9qaXIQuMNKw0PAKAp5mDWufwVhlvhPmKP\nvmt/+nTzy9IKQnZp2fJ9HB2dGD9+MgcPHqd+/QaWDkkQhFzC5Mveu3btYtmyZVhbW7Nr1y5mzZpF\npUqV6N69e07Elya7zbd0C3LDQDdVCfMGzw11NhSluTS+AWvy2dK4cVHRWhdyhWfPnrJp06+MHj0e\nmUxG2bLluHjRHxcXV0uHJghCLmOyxb527Vr+/PNP3NzcAJg4cSJbtmzJ9sDSI65PJUK3fUzU8kL6\nbYnFO5p1rutK3Wj4cmrd5C5DhtShevVcNtmN8M7RarWsWbOaxo3rMn/+bLZu3az/TCR1QRDexGSL\n3cnJCTs7O/26ra1trpmnWXKwQlPBHYVSC4Hmn2ePteE2/BgZn8aegpBzbt26yZgxwzl37gwAbdu2\np0kT8WhIEIS0mUzsbm5ubN++nYSEBPz9/dm9ezfu7rlrPnJF7AMAEgq/n+Fjo4G+LoYvLhU0uWcM\ngfBuSkhIYMmSRSxb5odKpcLLKz/z5/vSvn3unKtBEITcxWRX/MyZM7l69SoxMTFMnTqVhIQEZs+e\nnROxpZvjpckAKGKeZui4GzdCWHMtSL/+0/2XWRqXIJhj7drV+PouQKVS8dln/Thx4pxI6oIgpJvJ\nFruzszPTpk3LiVjMogi/pV+OL90rQ8euXn2BX79MmuXqVghhu25Bn2pZGZ4gpIskSfr5GPr1+4KT\nJ48zePBwGjZsbOHIBEHIa0wm9mbNmr1xApjDhw9nRzwZ4uB3HmuH+ZDUkx5XaViGjk9QyuHVFKxR\niSgUYgS8kPP27PkbX98F/P77Dtzc3LGxsWH9+t8sHZYgCHmUycS+ceNG/bJKpeLUqVMkJCSk6+Rz\n587l8uXLyGQypkyZQrVqKVvDvr6+XLp0iV9++SUDYYMsLB77H6/BpH0AqK2rgSxj9XZODKlnWOnx\nO4rR9TN0vCBkxosXz5k8eTy7dv0JwM8//8To0eMtHJUgCHmdyUxYuHBh/T8lSpSgV69eHDt2zOSJ\nz549S0BAAJs3b2bOnDnMmTMnxT53797l3LlzZgVuffoZeBqeqSvXtsjwOZ7WLaw7NjyexoUc8fIy\nr7iNIGSEVqtl/fq1NG5cl127/sTBwZG5c30YMWKMpUMTBOEtYLLFfurUKaP158+f8+jRI5MnPnXq\nFK1btwagdOnSREREEB0djaOjoWzr/PnzGT16NCtWrMho3MjCEqDpLsOGwIxNerHfWqFf3qfVUHV7\njwzHIAgZde/eHbp1G83Ro0cBeP/9D1mwwI8iRYpaODJBEN4WJhP7d999p1+WyWQ4Ojoyc+ZMkycO\nCQmhcuXK+nV3d3eCg4P1iX3btm3Uq1ePwoULmxM3isAoiLPXrVxogqaIU4aO/8zZ8IpbVfGKm5BD\nnj59ytGjR/Hw8GTuXB86duzyxjEsgiAI5jKZ2CdNmmSUoM0lSYYZ0sLDw9m2bRtr167lxYsX6T6H\np2ey5D22HmzUFe6gRhMUzWoaf24qnqR/r3r9vNkmFHLsWubL7fHlRQ8fPqREiRIAdOnSnrVr19Kh\nQ4dcVw/ibSN+l7OfuMe5k8nEvmDBAtavX5/hE3t5eRESYphpLSgoCE9PTwBOnz5NaGgovXv3JjEx\nkUePHjF37lymTJmS5jmDg6P0yzJtJB5Jy+FNG6EqURKSfZ6WALkMkmZyaxccRXAGfq7MCk5njJbg\n6emUq+PLa6Kjo5g79xvWrv2R7dv/1k+n+vnnnxMcHCXudTYSv8vZT9zjnGHOlyeTib1QoUJ89tln\nVK9e3aiU7MiRI9M8rnHjxixfvhxvb2/8/f3x8vLSd8O3adOGNm3aAPD48WMmT55sMqm/zs5/qX5Z\nVbIZZKA7c3CybnibDF1VENJn//49TJw4lidPHqNQKLh27YqYJ10QhBxhMrEXKVKEIkWKZPjEtWrV\nonLlynh7eyOTyZg+fTrbtm3DycmJ99/PeOnX1yki7xlWMpDUJeA/K93AuZaJ6kzHIQjJBQUFMXXq\nBHbs2AZAjRo18fVdTtWqovCRIAg5I9XE/tdff9GhQweGDctY0Zfkxo0bZ7ReoUKFFPsUKVIkw++w\nAyDXhR7ZeGWGDttga+h1qPbNEfpeeIZCIUehkDFiRD2qVvXKeCyCAJw8eZzPP/+E8PBw7O3tmTRp\nKgMHfoVSafL7syAIQpZJ9S/O1q1b6dAhF9enfjX6TaZIc7fXLbG31i9fOhHI4cMB+vVevTI/SFB4\nd5UrVwG5XE6LFq1YuHAJxYoVt3RIgiC8gzJWqi2XsP3jDrYPdXPC2264ic32u+k6Lh54pND9yCNi\nE9BqJaPP5fI8eTsEC1GpVKxbt4bExEQAPDw82L//CL/9tk0kdUEQLCbVFvvFixdp3rx5iu2vJquw\nZK14hwVnIekJgfUJNWplKOkpcvs0WS340TGJfPZaYhe14oX0unTpAqNHD8ff/yqhoS/1pWBFQhcE\nwdJSTeyVKlXCz88vJ2NJN9WHasNo9qfFsboQlNbueg3cdaPyi2q0OAATJjSif/9YtFoJjUZL+fL5\nsiVe4e0RExPD/PmzWb16JVqtlmLFSlCzZm1LhyUIgqCXamK3trY2uypcdlPILhhW1NbEdytr8phL\nSkM3e2BSd3yDBrnz5xNyp3///Yfx40cTGPgIuVzOkCEjGD9+Mg4OYo4BQRByj1QT+5tmYsstEj72\nRHkD1NbViVncDFWd/CaPCUz2/PyFKKogZNCJE8fw9u4KQNWq1fHzW0b16jUtHJUgCEJKqSb28eNz\n5/SRdteW4nDDBwBViTokNihh8hgtMMBFV5TmgwQ14km6kFGNGjXh/fc/pGHDJnz11VDxCpsgCLlW\nnhsG7njha/1yfJlP03XMMSvDK3FlxYQvQjoEBDykTx9v7t/XFUKSyWT8+usWhg0bKZK6IAi5Wp77\nCyUhQ4ZEWJv9qD3SN2jpJztDUZrpMekZPy+8q9RqNatWfYePzxzi4uJQKJSsXfsrgJiFTRCEPCHP\nJXYUNqCJR6sxPWDulYCkwXLFXmuth4bGIUm619zkchkODlYoFHmuE0PIIlevXmb06OFcuXIJgC5d\nujFr1gILRyUIgpAxeS+xJ03/KinTn4BvKHVd8f97rbXeqdMWbt58qV8/cqQPFSt6ILxbYmNjWbRo\nPitXLkej0VCkSFF8fPxo3fpDS4cmCIKQYXkvsau0oABHv/NoCnugqpMfVYOCqe6emGy5ulpj9Flc\nnPEkMHK56Gp9Fz158pgffvgOrVbLoEGDmTTpa/1MhIIgCHlNnkvsMi2gANu/H4D6CapqHoT/9lGq\n+4clS9YlNYZKc8+eRREQEGG0r5WV6IZ/V0RGRuDk5IxMJqNs2XIsWOBHhQoVqV27rqVDEwRByJS8\nlcm0GrBSGW1K+Khkmof4Jz0zV0iS0WtuAQGRtGpVgoYNC1O9en6KF3eheHGXrI5YyGUkSWLbtt9p\n0KAmf/yxRb+9d+8+IqkLgvBWyFMtdmXoJcOKRvfcPDGNbniAx0mJ3U0yrgvfoEFhGjToAoBKpeHS\npRdi4NxbLjDwERMmjObgwX8A2LPnb7p162nhqARBELJW3spkku6ZuNY6P/HtyiAp5WhKu6Z5yP1X\nif21CV+SUyrl1K1bKOviFHIVjUbDDz98x3vv1efgwX9wcXHFz285q1f/bOnQBEEQslyearG/onEu\nRuyX1Yia3wTSGPCmAb5Lmn/9PZUm1f3E+8lvr6dPn9C//6dcuHAegA4dOjNnjg/585suQywIgpAX\n5a0WezKaMq5gous8OFnS7x2vezZ/7VpwtsYl5C7u7vmIiIigYMFCrF//Gz/+uE4kdUEQ3mp5NrGn\nxwFrQ4dEVbWWzZuv07r1r+zZc9eCUQnZ7dSpE4SHhwFga2vLunWbOH78LG3atLNwZIIgCNkvTyV2\nq5Ck6VrTeF6e3IukFruDVkKl0jB8+F60WonBg3dz+fILJCl95xHyhvDwMMaMGU7Hjm2ZOdMwp0C5\ncuVxcnK2YGSCIAg5J08lduvrRwGQrOzTtf8CBxsAesWr6N17h357bKya7t23EhISl/VBCjlOkiR2\n7txB48Z1+fXXdVhbW1OkSFHxxU0QhHdSnho8Z33vMhQA+U03HM+dIb5LGdSV8r1x35vJnr+XUmu4\nEGlcTtbZ2RZPz/R9QRByr6dPnzBp0lj27t0NQP36DfH1XUa5cuUtHJkgCIJl5KnEToHHACgPeaK8\nfRNVba9UE/sYJ1v98sAENQP3fsKdO6H89ddtFiw4ycKFrXIkZCH7vHjxgvfeq09UVCROTs58/fVM\n+vTph1yepzqiBEEQslTeSezaZHXdA0sDoK7onuru/yXNwZ5Pa5jRrWxZd0aPrs8nn1SmYEGn7IlT\nyDH58+fno48+JjIyknnzFlKwoKhFIAiCkHcSe+BJw3KcAwCaQm+eqONSspnftoQbP0eXy2UiqedR\nCQkJLFmyiObNW1G/fgMAFi1airW1tYUjEwRByD3yTmKX6QrMaGwKktC/Clbng8Ba8cZdj1ole83t\ntTnYhbzp9OlTjB07nDt3brNr158cPnwKhUIhkrogCMJr8k5if7gPAI1LKTQK5zRrxM921I2G7x2X\nyLVrwVSp4pkjIQpZLzIyglmzZrBu3U8AlClTFh+fxSgUb/5SJwiC8K7LO6OMFFYAyDSxxHcti6pJ\n4VR3rZg077ryXhgtW/7C9eui2lxetHv3Lpo0qce6dT9hZWXFmDET+PffEzRs2NjSoQmCIORaeSex\nJ0ks2g7SqO1+VinnhlLXmlv3yTYA5sw5DoBaLbrl84rIyAjGjBnG8+fPqF27LgcOHGPSpKnY2tqa\nPlgQBOEdlne64p8kDZ7TpF10ZKV9smeujyMB+OefB/z2mz+lS7uJWdxyMa1Wi1arRalU4uzswrx5\niwgNfcnnnw8UXe+CIAjplIcSu+4ddpkUn+Zu4Umtebfv/yMszLDvhAkHePhwRPbFJ2TKnTu3GTt2\nBC1btmbUqHEAdO7czcJRCXnNy5ch/PTTKhYunG/pULLdsGGDiI+PN+rFGjt2EiVLlgJg//69bN68\nAaVSiVqt5rPPPqd5c139DrVazerVKzl79hS2tnZYWVkxcuQ4SpcuY5Gf5ZUdO/4gJiaa3r37WjSO\njRvXc+jQAUBG//5f0LBhE6PPr169zPLli7GysqJatRp8+eVQ/WeSJDFkyADq1KnP558PZPz4kUyY\n8D8KFEh9XFhWyzuJPUoBDqD4Jz+Of54ktn8VtMWN639LwImkiV/mNS/GlSG12bjxGuHhCUyf3gx5\nGlO8CpaRmJjI8uWLWbx4IYmJiTx58pjBg4djY2Nj6dCEPChfPg8mTPifpcPIMVOmTKNUKV0yvnDh\nP5YsWcjSpSu5du0KW7ZsZPHiFTg7uxATE824cSNxdHSiTp16bNy4nujoKNas2YBMJuPq1ctMmTKO\nDRu2olRaJi2EhYXy11/b+fHH9Ra5/itPnz7hwIH9rFq1lujoaIYOHUi9eg2Neg0XLZrPjBlzKFmy\nFPPmfcPVq5epWrU6ADt37kCl0tVdUSgUfPXVMPz8fPDxWZxjP0PeSeyxKnAAmwPP4Yk9cf0qp9jl\nnsKQuD/MZ0eXGc2YMKER06cfoW/fajkZrZAO586dYezYEdy8eQOA3r37MH36LJHU8xDPSuuM1oOv\nv7mlZbvlNk4zTunX47qXJXpmowxfb/funVy6dIHw8HAePLjPoEGDOXBgHw8fPmDatNm4u7szdepE\n/vprB+fOnWbVqu+Qy+W0bv0BPXp8grd3Zxo0aIybmxtt27Zn3rxvUKlUyOVyJk36mkKFjAflnjt3\nhh9//B4rKyucnJz45pv5TJs2mZ49P6FGjVokJMTTu3d3Nm/ewY8/fs+VK5fQajV06dKD999vw5w5\nM1AqrYiMDGfKlOnMnDmVuLg44uPjGT16PJUqVWHv3r/ZuHE9Xl75cXFxpXbtunz4YTt8fObw9OkT\n1Go1Awd+Re3addO8N5UrVyEw8BEAv/++iQEDBuHs7AKAg4MjX345lI0b11OnTj127PiDdet+Q5bU\nw1m1anV+/PGXFEl9796/2bp1MzKZDG/v3rRq9QEffdSKv/8+CMDUqRPo0qUHFy+e5+nTJzx79hQX\nF9d035/k/vxzG23atEMulxMU9IJZs6YBut6FqVNnUrhwEby9O1OuXAXq1atP5crVWLzYB5lMhr29\nPVOmzMDJyYnly/24ft2fxMREOnXqyscfd9JfQ6PRMHLkYKPr5s9fgK+//ka/fuHCfzRo0AgrKyvc\n3NwoUKAgDx8+MOrNePkyRN8zUq9eQ86ePU3VqtUJDw/nn3/20rFjF4KCXgBQrlwFoqIiefw4kCJF\niqb53zCr5J3E/tqzdckmZeixyQbVOST9W6GQsWBBK9Faz0USEhKYMeN/rFmzGkmSKFmyFL6+y2jS\npKmlQxPygMDAR3z33Y/s3LmDX3/9mTVrNrBnz04OHNhHjx69AF13qK/vAlauXIOzszOTJ4+lY8cu\nqNVqGjTZHAQAAAAgAElEQVRoRIMGjZg7dybt23ekVasPOHToAGvW/MDUqTONrhUVFcX06bMpVKgw\ns2ZN48yZUzRr1oITJ45Ro0Ytzp07Q926Dbh27QovXjzn229Xk5iYSP/+n9K0aXMAnJ2dmTjxfzx6\nFED79p1o2rQ558+fY8OGdcyatYBVq77lp59+wc7Onj59elK7dl3++Wcv+fJ5MHnyNMLDwxk58ivW\nrfstzfty6NBBypevAEBAQABly1Yw+rxs2XI8ehRAdHQ01tY2ODkZF+p6fT02Noaff/6Rdes2kZio\nYs6c6bRq9UGq11erVXz33Y/s3ft3uu+PjY3hMcKFC/8xdOgoQJc4+/X7glq16rBr159s2/Y7w4eP\n5unTJ8ydu4hSpUozcuRgxo+fQtGixdi27Xe2bduCt/enFChQiOHDx5CQEE+PHp2MErtCoWDFih/S\nvI+hoS9xdXXTr7u5ufHyZYhRYi9YsBCXLl2gevWa/PffGX1r/rvvljJo0BD9F6xXatSoxYUL/4nE\nnsJrU7VK9ilDv5Q0Gv7V624ANm/4AiBYlrW1NTdv3kAulzNs2CjGjJmAnZ2dpcMS8ogKFSohk8nI\nl8+D0qXLolAocHPLR0zMZf0+oaGhWFtb4+am+wPt47NE/1mlSrrevlu3bvDVV8MAqFWrDj///GOK\na7m6urJgwWw0Gg1Pnz6hdu26vPdeczZuXM/QoSM5duwIrVp9wNWrl/H3v8qwYYMAkCQtISEhRtdz\nd8/HunU/smnTL6hUKmxtbYmICMfBwQF3d92cF69a5deuXeHy5YtcuXIJ0H0ZVqlUWFlZGcU3d+43\n2NraEhISQqFChZgyZQage3FIq9UY7StJ6OdReP2zN3n48AHFipXAxsYWGxtb5s/3S3P/ihV1P2fj\nxk3TfX8KFy6iPz4kJBgvLy/9vVqyZBE//bSKqKhIypevCICtrR2lSulKil+/7s+CBbMBUKlUVKxY\nCRsbGyIjI/jqq/4olUrCw8NM/pymvGmSyMmTv2bpUl/kcjlly5YjJiaGS5cuoFAoqFq1eorE7unp\npW/B54S8k/XKukEMxAyphja+MpJzyopjv9nqfukj03gdTrCMFy9eoFarKFy4CDKZDD+/5cTExFCl\nSlVLhybkMcmfdSZfTj5Nr1wuR6t98xs0SuWr5CjTH6NSqZHJ5Fy7doXvv18BwPTps5k3bxYLFy6h\nRImS+PktAHQtWw8PLx49esi1a1cYP34KDx7co337jnz2Wb9Ur7dly0Y8PLz4+utZ3Lx5nRUrliBJ\nkr47HNAvK5VW9OnTP0V39etePWM/ceIYO3dux8PDA4BixUpw8+YNvLzy6/e9c+cWJUqUwtHREbVa\nTWjoS/0XCoBbt25Srlx5fQxyuQJJSvsVYbXaMIfHqy8dGb0/xnTX/umnVdSv34BOnbpx6NABTp48\nnnQNQ8qytbVl+fJVRvfv4sXzXLjwHytW/IBSqeT9998zOnt6uuI9PDx59ChAvx4cHKS/r6+UKlWG\npUtXAroBf1FRURw7doSbN68zaNDnhIeHoVLp/t61afORiZ856+Wd99hddM9dExsXJr5XhTe+y34t\naarW4qKMbK4hSRIbNqynSZO6jBgxRP+HtGTJUiKpvwWCr/c1+ic18T3KGe1nzvP1jHBzc0Or1RAc\nHIQkSUyYMIqoqCijfSpWrMSFC/8BcOnSeSpUqEiVKtVYseIHVqz4AU9PL2JiosmfvwBRUVFcuHAe\nlUoFQNOmzVm3bg2VK1dFqVRSqVIVTpw4hlarJSEhgcWLfVLEFBERrm+hHjlyCLVajbOzC5GREURG\nRpKQEM/Fi+cBqFSpCsePHwF0g8pWrfo2zZ+3ceP3SExM1CfA7t17sWbND4SF6VqssbEx/PDDd/Ts\n+QkAXbv2YNkyP31ivnLlEnPnziAxMVF/zuLFS/DoUQCxsbEkJCQwatQQ/ReR+Ph44uLiuH371hvj\nMef+eHh4Ehysa9WGh+vulSRJHD9+RH/fkytTpiynT+tegz5wYB///XeWiIhwvLzyo1QqOX78CBqN\n1ujYV13xyf9JntQBatWqy6lTx1GpVISEBBMcHEyJEqWM9pk7dyZ3795Bo9Gwb99uGjduwvDho1m7\ndiM//PAzn38+kPbtO+qTenBwEJ6eXqn958tyeafFbsJzuYz4pOfohVed51xpV2rVKoBCkfZ3l3XH\n1Nx6nva78YJ57t+/y9ixIzlx4hig+7YdExONo6OYhEfIfmPHTmLq1IkAtGzZOsUz5IEDv2LevFns\n3LkDpdKKyZO/TnGOLl26M3jwAIoWLUbv3n1Ys+YHGjduStOmzVmyZCHz5i0CdIPPataszZdf9gMk\nOnfunuJcbdp8xOzZ0zl06ABdu/bgwIH97Nu3m759BzJ06ECKFClG+fIVkcvltGzZmgsXzvHVV/3R\naDT07z/I5M87fPgYpkwZR+3adalSpSqDBg1h7NjhWFlZoVar6d7dm+rVawLwySd9WL9+Df3798bZ\n2QVHR0fmz/czGrhqZ2fHgAFfMWrUEAB69vwEmUxGp07dGDSoL+XLl9N3kb/OnPtTq1YdLl++RLly\nFejYsQuLFy+kQIFCdOvWEx+fOZw9e9po/5Ejx+HjM4cNG9ZhbW3DjBmzkcsVbNiwjmHDBvHee81o\n1KgJixbNY/LkaSbv3ysFChTg4487MXToF8hkMsaNm4RcLuf0/9u787ia8j+O46/bXsrSbjdorMPY\ntyzZjWWMMRTZIkK2EBFFqVC2yD7WmRGm2fxswzBj7NvYiRiU9iztunV/f1xd3WmzlNvN9/l4zOOh\ne86553O/Q597tu/7zCkiI5/w1VcD6dPnS3x8PAHo2rWn4smE/Fy5cpkvvuj7xjW8L4lMltcVhBLo\nh3bw5BRPex5Gat461+KZUhnbKr56/E1rIWTKCA2dQPnyBc9UNmdP7m+CxaWOpYQR7UvudykzMyNi\nYxMLX7EQGRkZBAWtwt/fj/T0dExMTPD2XsyAAd8onTb7WBXVOAv5U6cxPnbsCM2ataBs2XK4uDgz\napSj4tGpkqyoxzg+Pg5X12ls2rS9VP2euHs3lA0b1rB06cp32t7M7O0PhEpul3lLO8xeRbjeioVM\nGebmZQpt6jn5fKNd+EpCoTIzM+nTpxuXL18CYNAgOxYs8MHExKSQLQXh45SWlsbkyePR19ejdu06\natHUi4OJiSl9+37JDz/sYMiQ4aoup0hkZmaydm0gM2bM/qD7VZ/Gnv6swMVZ2RnsG+QNpVevWsVd\nkZAHTU1NunXrSXx8Av7+K+jUqbOqSxKEEq1Xrz706tVH1WWUCP37l67ZJjU1NVm2LPCD71d9bp6L\nvwmA1rV4NKKTlRY9yfGM+nfdatC0qSX29uLGrA/ljz9+5+DB/YqfJ0924c8/T4umLgiCoALq09hf\nMZoaic6fEUqv3c9xg1zX5hXZtq0fjRtb/HdToYjFxcXh5DQaW9uvmTZtIgkJ8YD8OfUyZcoUsrUg\nCIJQHNSrsUu1IFMbmZbyjRVXtV5/DIlEgoWF4Yeu7KMik8kIDv4ea+vmhITsQV9fH2fnaYrpKwVB\nEATVUZ9r7AAS+Q38Gs/SlV72NJTfJGf9UpprE6Fo/fvvA2bOnMqffx4DoEMHG/z95RN4CIIgCKqn\nXo1dUz4N4suOr6chPJ/jaL1tRuHTJArvTiaT4eAwjOvXr1KhQgUWLvRl0CC7UvVoiiCoo82b1/P7\n7wcxNTUDID09DXv7UXTsaAPA7du3CApaSWpqKlJpBu3bd2LEiNGKmfsOHvwfe/bsQkdH/sz7kCHD\nsbHpqrLPA/LHxLZu3ciiRUtVWsf582fZsGENGhqatGnTjpEjxygtv38/TDHhjoaGJrNmzaVSpcq8\nePECT8+5GBjo4+0tX+7hMQdb26GK6XeLi1o19hRLV9L29yezxutTvnNzTCk5NeVlXpsJ7yl7timJ\nRIK3tx/bt2/By8sPMzMzVZcmCMIr33xjy9dfDwbgxYvnjBw5hNat25CZmYmn51y8vPywsvoUmUzG\nypUBfPvtBhwdx3P16j/8+ONuVqwIwsjIiKdPE3BycqBWrdpUq1ZDZZ/H39+XBQt8Vbb/bCtX+hMQ\nEIiZmTnOzmPp2LGzItkN4Ntv12NvP5JWrdpw+PBBvvtuGzNnzsHf35dGjRpz716oYt1Jk1xwc3Nh\nw4ZtxXpApFaNPauSkVJTv3Axkn96fgqAxcF7DPM/ydixTbGxqaGaAkuZlJQUlizxITk5maVL5VnC\nbdta07attYorE0oKc3PlYJCYGJc819u+/SozZhxR/Dxs2GcEBHR76/0VFtvaoEFDAgOXcffubZKT\nUxWxnVFRkXh7e5CVlYWlZUXmzvXEz89LEam6YIGvIib15cuXjBnjRMuWyhNh5RUlevLkCZKTkxg1\nyhGASZPGMWXKDCIiHrNr1040NbWoU6cekyZNY//+3zhz5hRxcbEsWODDrl07c8WL3rt3l0WLPDA0\nNKJu3fo8e/aUuXM9+fHH3Rw5chCJRIP27TthZ2df4DiVLVsOExNT4uLiOH/+DB06dMTKSv67UiKR\nMG7cROzsBjBmjBM//hiMg4OjYma+ChWM2bRpR66Z+kJDbxMQsBgNDQkNGzbG09MdZ+exuLi4UrNm\nbX78MZhnz57RpEkzdu3aSUpKCk2bNkcikbzR+OR05co/GBsbY2lpiVQqZdEiT2JjY0hNTcXBYSzt\n2rXH2XmsIhDGyckZH58FJCYmkpmZydSpM6ld24rDhw+wd28wmpoa1KhRi1mz5irtZ/nyJYSF3VN6\nzcdnqeJ+oYiIcIyMymJhYQlAmzbtuHjxnFJjL1euPM+fPwcgMfEF5cqVB2D2bHdu376l1NhNTU2p\nWrU6Fy6co0WLVgX+P3wfxdrYfXx8uHLlChKJhDlz5tCo0etM9DNnzrBs2TI0NDT45JNPWLRokSJ5\n6E2NnHMMXjX2aNffib4Ww+LFXYr0M3ys/vzzGDNmTOHhw3/R1NRk4sTJ4jq6UCIUFNtau7YVlpaV\nWLjQg/DwWEVs54YNQdjaDsXauiNBQSu5ffsW8DpS9cCBfejo6LB69Qbi4mJxdh7Hrl0hSvvNK0p0\n0CA73N1dGTXKkRcvnvP0aQKVKlXGx8eTdeu2oKOjw7x5sxUpbdHRUaxb9y0vX77MM150y5YNjBzp\nSMeONsybNxs9PT2ePIng+PGjBAVtBmD8+NHY2HTF0tIy3zF69Ohfnj5NwNzcgocPHyoS5rLp6+tj\nbGxMXFzsq4jXOkrL/9vUAVas8GfmzDnUrm2Fl9d8IiIicq2TLSzsHj/8EMLTpwlvPD6NGn2u2P7S\npfOK6W8TE1/QsmVrevXqQ0REOPPmzaZdO3m4S82atejffyBbt26iVau29O3bnwcP7rNypT8rVgSR\nmppKQEAgRkZGTJzoSFjYPaX41WnTXPP9DJB3hOt/P/eYMU6MGTOcrVs3kpWVxcaN2wEwMMj7yaDG\njZtw6dIF9Wzs586d4+HDhwQHBxMWFsacOXMIDg5WLJ8/fz7bt2/H0tKSyZMnc+LECTp27PjG7//s\nWRrx+jnKvxbDJ5+Up1o1cWf2+0hIiMfDYy7Bwd8D8ijG5csDRVMXSoyCYluzYzttbW2RySSK2M7Q\n0NtMmTIdgAkTpgDw8897lSJcmzRpBsjDSHR0tHnx4rnSkx55RYnKj+QkxMXFceHCWdq378SDB/eJ\njo7CxUUeCZucnERUVBQgD56RSCT5xos+fPgvjRrJZ56ztu7AhQvnuHXrBuHhj5k0aRwgD3SJinqS\nq7Hv2bOLY8eOkpKSzMuXGXh4eKOtrf0qwjV3MJZMJkNDQwOJRD5DWmEePXpI7dpWAMybt7DAqU5r\n17ZCR0fnrcYnx3EfcXGxNG0qj7A1MirLrVs3+PXXECQSDV68eK5Yr169hgBcu3aVZ8+ecuiQfD6N\n9PQ0QP7Fzc1t+quxfcDz5wVPdFaYvCZgX79+DePGTaR79178+GMwW7duZNKkvM9cAZibmyu+6BWX\nYmvsp0+fpmtX+c0XtWrV4vnz5yQlJWFoKH8ULSQkRPFnY2NjRQrRmzI01KHnjv78L8drf/wxrEhq\n/1j98MMPTJ48mbi4OHR1dZkxYzYTJkzOlQEtCKpUUGxrdmznd9/t4NmzNEVsZ34xrnlFuII87+Dl\nywxFfviQIcM5fvxonlGiHTp04tSpE5w7d5phwxyQSKBOnXosW7ZaaV/79/+m2F9+8aLy+1nkZy5z\nRri2adMOV1fl08j/lX2NPS4ujilTnKhVS96EsyNce/T4QrFuSkoKL168wMTElGrVanDr1g3F6WaQ\nf8EwMzPHwMBA8VpeZ1RzXifOK8L1bcYnv/f+/feDvHjxgjVrNvHixQvGjHn9ez47xlVbW4tp02bS\nsOHrbwcZGRksW7aErVu/x8TEFFfXqbn2UdipeFNTM8X8HJB3hOu1a1cYP34SAC1atGLx4kUFfq4P\nodieY4+Li6NChdenMIyNjYmNjVX8nN3UY2JiOHny5FsdrQNItDT4X035/OPVX0r59dfBlCkjGtD7\nOHToEHFxcbRr157jx08xZcp00dSFAsXEuCj9l5/hwxsprfcu19ffRHZsp7a2tlJsZ9269bl06TwA\nmzat4/z5s0rb5YxwjY6OQkNDA1NTU0W0Z9u21vlGiXbsaMPp0ycJDw+nTp26VKtWg3//fcDTpwmA\n/I712NiYPOv8b7xo5cpVuH1bPstmdiRpnTr1uHTpImlpachkMlas8FcckebF1NSUnj17s2XLRgC6\nd+/FqVMnFO8LsGFDEH379gdyRrzK642Pj2PevFlER0cpvW+NGp9w48Z1AHx9FxIWFkaZMmWIj48D\n5A0uL+8yPv+NcK1YsRIaGhr8+ecfeUa41q/fkL/+Og7Agwf3X13jT0ZTUxMTE1Oio6O4ffuW0pcP\nkJ+K/2+Ma86zNBUrViI5OZnIyCdIpVJOnfqbFi2U772oXLkqN2/Kx+XWrZtUqVI1z3HIFhsbi7l5\n8U6g9sFunssrRC4+Ph4nJyc8PDyUvgTkx3DdFQzDfoILw9mf4/UQHS2a9q37jpXJ/3K9S4KOusvM\nzCQyMpIqVeSPDwYEBNCpUydGjBghHmErZh/j37eiYGSkh4GBDmZmRpQrp4+enrbSn3v27EJw8E7s\n7e3p2rUrNjadWL3aH1fX6bi5ubFv309UrFiRHj1s+OuvI5Qrp4+ZmRG2tl9z69ZVXFwmkJGRwaJF\n3rn+Hw0fPpTFixdTuXJlhg0bxrx587hz5wrW1tbExETSsWP7V9sYMW+eO7NnT0NHR4f69etTr15N\nbt++oqg9u85p08Yr1TllyiTc3d35+efd1K5dm8TERD77zAoHh5FMmTIOTU1NunbtSpUqyk+klCmj\ni6GhnqJmZ2cn+vbty9Chg7GysmLz5k14enqSnJyMVCrF2tqaqVMnoaWlhY1NW2SyGbi6TkFfXx8t\nLS08PObTsqVyGI2n53w8PT0B+Pzzz6lVqxbDhg3F19eX6tWrU61aNcqU0aV8eQN0dbUVtZiZffZG\n45Pzd07nzh3YunUrZmZGDBjQl/Hjx3P37i2+/vprKlWqSHDwNnR0tKhQoQxmZkaMGzcaNzc3pkwZ\nR1ZWFnPnzqV27aq0b2+Nk9NI6taty9ixjgQFraB795/f6oDF23shixbJb5rs168PzZo1JDY2lsDA\nQBYuXIi7uxuenp7s2fM9Ojo6eHl5YWxswMiRI3nx4gXR0dG4uExgwoQJtGnThtu3r9G/f/9i/R1Q\nbLGtgYGBmJmZYWtrC0CXLl345ZdfFEfqSUlJDB8+nKlTp9KhQ4fC3zBAAn98iexaT+IuDsW9jC4b\nDHQwzJJxPz7pnevMjm392NLdrl+/xvTpk0hOTubo0b/R1dVVq6hLdSbGufip6xhfv34NPT09ate2\nYseOLchkMoYPd1B1WXkq7jEeO3YkXl5+SpcI1F1CQjyurtPYuPHNH3d7ly8AxXYqvl27dhw6dAiA\nGzduYG5urmjqAH5+fowYMeLNmnoOsleBLxv15Y24Zmbum0KE/KWmpuLt7Un37h25fPkSSUlJPHz4\nr4qrEgQBQEdHGz8/LyZOdOTy5Uv07/+1qktSmZkz3QgMXFb4impk1aplTJvmWuxnRIvtiB3A39+f\nCxcuIJFI8PDw4ObNmxgZGWFtbU2LFi1o0qSJYt0+ffowePDg/N/s1RF71s1exJ8bQiVTQ6QSCQuT\n0nBKzX3N5U19TEfsf//9F9OnT+bBg/tIJBIcHByZO1f+zCyo71GOuhHjXPzEGBc/McYfxrscsRfr\nNfYZM2Yo/Vy37uvr4NevX3/r90ue0JhHZXswovcutI6PRGqgTc+0d2/qHxNPT3eCglYBULduPQIC\nVhXrc5SCIAiCaqhVupvMQJvzD55y/p8o0gzkR9gzpxxScVXqoW7deujo6DBr1lyOHDkhmrogCEIp\npVZTyoKM9esvQY5JaCqpsJqSLCIinEuXLtK375cADB48hDZt2lG9eg3VFiYIgiAUK7Vq7OnpmZQt\nqwvl9BSvfdbQXIUVlTyZmZls3boJb+8FSKUZ1K9fn1q1rJBIJKKpC4IgfATUqrHr6mqxffuXBN1/\nhiege/IRtrbFG3+nTm7duomLyyQuXpRPxNG7dz/FjXGCUBpERj5h+HBb6tSR36+TkZFBzZq1mTFj\nNpqamqSlpREYuIzQ0FuAhAoVTJg+fZbikanHjx+xalUAz549JTMzi88+a8TEiVPR0dFR4afK7dKl\nCyxZsoixYyfSuXP+8amRkU9wd5/F5s07inT/f/xxBF/fBaxfv4WaNWvnWp6ZmYmjoyMTJ7pQuXKV\nPN7hw0hKSmLBgrkkJSWhr2+Ap6e30gQzmZmZLF3qw+PHj8jIyGDAgG/o2bM3//xzifXr16ClpYW+\nvj7u7gv5/fcDZGZmMmjQEJV9nqKiVtfYs/3WVH4CvlkTSwwNS9Y/SFVIS0vDz8+brl3bc/HieSws\nLNmy5Tu2bNlZqp4BFQSAatWqK2YJW79+C1JpBr//fhCAwMBlmJqa8fPPP7Nx43bs7UcwffpkpFIp\nmZmZuLu7MmTIcDZu3K5ohtkztJUkV65cZsCAbwps6sXl8uWLnDlzUjElbV5+/nkvzZs3V2lTB9i9\n+3uaNGnG2rWb6djRhp07tyktP3PmFKmpqaxZs5HAwHWsXRtIVlYWgYHLcXObR2Dgeho2bMQvv4Qw\nYMAgjhw5nGsWPHWkVkfsOiciyIwP5+LAegDE62hCsoqLKgFmzXLhhx92AjBixGjmzfNU+tYqCKVZ\n/foNCQ9/TEpKMmfOnCI4+GfFskaNPqd+/QacOHEcfX0DqlWroQh7kUgkTJgwWTE3ezapVIq3twfR\n0ZHo6Oji7r6A8+fPcv9+GM7OU0lJSWH48MHs3fsbtrZf0bp1OypUqMCBA/9TJMIdOLCPe/dCsbMb\nhq+vF1JpBhoaGsyaNS9XeEtQ0EquXbuCVJrJ118PwsqqDv/7369oaWlhYmJKly7dFet+9902jh8/\nikSigZOTMxUrvr7LKK+I0qioKLy85qGhoUFmZibz53sBklyvWVpWVLxPnTp1adKkmWKe/Lzs3RvM\n3r17yMjIe7//jaj966/juWJn84rBzflF4dSpv/n+++1K++3XbwDdu/dU/Hzx4nnc3OTv0a5dh1zz\nwZcrV56kpCSysrJISUnFwMAADQ2N/0StJlKtWnUkEgl9+/bnp5/2MnbshHw/uzpQr8Z+NpJr+q+T\nfVYm5j9f8sdk8uRpXLt2FR+fpbRu3UbV5QgfiSFl9TmiW7S/QrqmS/n+Reobry+VSjlx4k/69/+a\niIhwqlevgZaWck1WVnV49Ogh+vr6ikzybLq6evzXgQP7MDExwdNzEUeOHOLvv/9CV1c33/23bt2W\n1q3bcunSBe7fD6NmzVqcOPEndnb2bNy4FlvbobRo0YrTp/9m27ZNzJrlrtj+n38ucf9+GGvXfktq\naiojRtiydev39OrVh/Llyys19cePH3H8+FHWr9/KkycR7Ny5lREjRiuW5xVRev78GVq0aMXIkWO4\nc+c2cXFxXL9+JddrORt7fnGj2aKiotDR0aF8+fLExibmuV94HVEbGfkkz9jZp0/jc8Xg5sxlb9vW\nmrZtrQusJT7+daxqhQoVFPPWZ2vY8DMsLCz45pt+JCcnK74ETJ7sgrPzWIyMjDAyKsu4cRMBeaTq\n/v2/FrhPdaBWjR3ArefrgPum0o9z1rl9+37l4MH/ERi4DolEQq1aVhw9ekLM7y58FB49eqg4mgwL\nu8fQocPp0KETd++GkpnHTJTyeFJNQJJnfOl/3blzm+bN5ZGhXbv2AOTJbPnJjn7t0MGGkydPULly\nFR48CKNhw0b4+Xnx6NFDtm3bTFZWllK2N8Dt2zf5/POmgDwjvUaNmjx+/DjP/YSG3qF+/YZoaGhQ\npUpVZs+eR2TkE8XyvCJKW7ZszZw5M0lMTMTGpgsNGzbCwEA/12tvIy4uFjOz1zct5xeNmh1Rm1/s\nbMWKlXLF4L6PvOZau3LlMjEx0QQH/8zTpwlMnuxE27bWLF++FB+fpTRq9DmrV6/gp5/28s03tpib\nWxAdHf1edZQEatfYH5vIYwSnpKSruJIPLyoqktmzZyh+yfTp8yU9e8qjGEVTFz60tzmyLkrZ19gB\n3N1dqVq1OgCVK1fm8eOHudK/7t0LpUOHTmhr6/Djj7uVlr18+ZLw8EdKN4hpauaOeM0vnhReR792\n7GjDvHmzqVmzFq1atUEikaClpY2X1+JcUZ853zdnP5Kfss/733JedWXLL6K0Zs3abN36A+fOnWHd\nutX07t2PXr365Pna28gej4KiUbPHJb/YWR+fBXnG4GZ7k1PxpqamJCTEYWhoSFxcLKamyuE4165d\noVmzlmhpaWFmZk7ZsuWIiYkmLOwujRp9DsijVg8fPvBWn7+kU6ub5/6oasAdS/lpoq6h8YWsXXpk\nZd9u2RgAABhtSURBVGWxdetm2rVrwf79v1GmjCG+vkvp1q2HqksTBJWaMGEK69YFkpaWhoFBGdq2\nbc+3325QLL927QqhoXdo08aaFi1aER0dyd9//wXI/12tXRvI0aO/K71nzojXkydPsH37txgYvI4n\nvXr1nzxrMTU1QyKRcOTIITp16gLIr/+fOHEckF8PPnz44H/21YDLly8C8oz0iIhwqlSpluf716lT\n79W1eCkJCfG4ub2e2TO/iNIjRw5x//49OnTohKPjBO7cuZXna2/D1NSMmJiYAvf737rzip3NLwY3\nW9u21rkiVXM2dYCWLVvzxx9HADh+/CitWilfiqxSpSq3bt0AIDk5SZGnbmJiwoMH9wG4desGVavK\nxzw2NqbYI1U/BLU6Yj/w4Kk8BEaaxfWD92g1uknhG6m5u3dDcXGZxNmzpwHo0aMXfn4BKr8bVRBK\ngkqVKtOpUxe2bdvMuHETmTJlOuvWBdKvXz80NDQpX74CXl5+aGpqAhAQsJolSxaxZctGtLW1adGi\nFaNGOSq9Z9euPbhw4RzOzmPR1NTC3d0TAwMDtm//FmfnsbRta53rhrts1tYd2LNnF/PmLQRg9Oix\n+Pgs4MiRQ0gkEubM8VBav3Hjz6lTpy4TJzoilUpxcnJGX18/z/euWLESPXp8gbPzWGQymeK6MMhv\nEmvRohVjxgyndm0rhgwZxqpVy3Bzm8/y5UvQ15ffNDZ16kzS09Px9/dRei2nfft+5uDB/dy7F4qP\nz0KqV6+h+DwAlpaWpKen8/z583z3O2iQndL6gwbZMXGiIxoaGnTo0AldXT2+/HIAy5cvxdKyEgMH\nDmbJkkWcO3eGli2V884LMnCgLV5e85gwYQyGhkavbg6ElSsD+OYbWzp0sOHcuTOMHz+arKwsJkyY\njK6uHjNmuLFkiTeamlqULVtOce39ypVLipsr1VmxhsAUqQAJMw/3xP/QAUhMZ99f/9Ky5fvPO1fS\nQ2CWLvVl6VJfzMzM8fVdSt++/YvttLsIdfgwxDgXPzHGxWvPnl1oacn46iu7wldWI+PGjWLhQt8S\n9ZhwiYptLRaWr2JfJRJq1apQ8Lpq7MWL13f+T57sgovLTE6ePE+/fl+Ja+mCIKjcV18N5Pz580RE\nhKu6lCLz44+7sbHpUqKa+rtSq1Pxn3T9RP4HQx2MjXM/pqLuEhNf4O3tyb59v3LixFmMjU3Q1dVl\n9ux5qi5NEARBQUtLi40bN5aqsyJffz1I1SUUGbU6Yte0MgFgQFpGqTtyPXhwP9bWLdmyZRNPnyZw\n6tRJVZckCIIgqCG1OmLXf/Woh76a3BbwJqKjo5kzZya//SafLatJk6YsW7aaBg0aqrgyQRAEQR2p\nVWN/lCJ/jKJtRqaKKykav/32Cy4uk3j+/BkGBmVwc3NnzBgnxR28giAIgvC21Kqxa746Yi8tx+vG\nxsY8f/6MLl26sWTJcsWzlIIgCILwrtSqsWu/mkL2MzWdSjYjI4O//jqmmP+5Xbv2HDhwlKZNm5e6\newYEoTiI2FZlxRHbmpSUhLf3fEV4iqvrXGrU+ERpHRHbWrKp1c1zabryU9RqVfQrly5doFu3jtjZ\nDVSaOrFZsxaiqQvCWxCxrcUrOPg7PvusMatXb8DefiSbN6/PtY6IbS3Z1OqI/XlZecLSwwfPqFOt\nrIqreTNJSUn4+XmxadN6srKyqF69Bhoa6vjVRBBKJhHbKldUsa329iMVv6PKly+vNK9GNhHbWrKp\nVWPP1khfPco+cuQQrq4uhIc/RlNTk4kTpzBzphsGBgaqLk0Q3lvZowPRjThcpO+ZXrk7L7rsfeP1\nRWxr0ce25vyse/bsypVJIWJbSz716JA5SbOwNCv5jXHbtm+ZOVP+7bFRo89ZvjyQzz5rrOKqBEH9\nidjWDxPbGhS0Cm1tbfr06a/0uohtLfnUr7Frqcdp7L59v2TVqmWMHj2OsWPH5zqKEAR19zZH1kVJ\nxLbmVtSxrZs2rePZs6f5znopYltLNvXokjk0S5cWvpIKPHhwn+nTp5CeLs+JNzY24fTpS0yYMEk0\ndUEoJiK2lVfbFl1s65Ur/3Dz5g1mz56X5/1AIra15FO7jnNTp2RN3iKVSlm3bg1Ll/qQmppKtWrV\nmDJFflqqpD1CIwiljYhtlSvK2NafftpDTEwUkyc7AVC2bDl8fJYqlovY1pJPvWJbOyzB1NIRB72i\na+7vE9t69eo/TJs2iWvXrgDyEAEvL798T7uVdCLq8sMQ41z8xBgXLxHb+uGU/thWoN/5KFWXQEpK\nCp6e7nTv3olr165QtWo1du36kbVrN6ltUxcEQXhTIra1ZFO7U/GVUqW8VHENBw/+j6CgVWhoaDBu\n3ERmzZqLoaGhiqsSBEH4MERsa8mmVo290c04ePsz5kUiIyMDbW35zuXfVs8yaJBdqbgeIwiCIJQe\nanUqPk6ayemnKR90nzKZjL17g2nZsjH378snXpBIJPj6+oumLgiCIJQ4atXYnyRncOFp6gfb36NH\nD7G1HcCECY5ERISzc+f2wjcSBEEQBBVSq1PxiRIDyn+AvJTMzEw2blyLn583KSkplC9fngULfLC1\nHVr8OxcEQRCE96BWjf2ZxIhqxdzYQ0Pv4Ow8ln/+uQxA//4D8PZegrm5eSFbCoJQ3ERsq7LiiG19\n9OghS5f6APJLkbNmuSsmcMmmLrGtABcunGP16hVoaGgwYMBAxRS5q1YF8M8/l9HR0WH+fC8SExPZ\nuXMrXl5+qvgoRUqtTsWb1K5Ks2YVC1/xPWhoaHDr1k0qVarMzp3BbNiwVTR1QShBRGxr8fr5572M\nHj2OwMD19O7dj++/z/2lQV1iW6VSKf7+vixZspygoE2cO3cWgNOn/+bJkwi+/XYnw4aN4ty5M9Sp\nUxcTE1OOHTuiio9SpNTqiN2iTg2aGhZ9Y4979A8yWXMkEgm1a1uxbdsPtGzZCkPDt58YQBCED0vE\ntsoVVWzr5MnTFX+Ojo7K88BGXWJb79y5TZUqVRXTxC5c6AvIpwru1q3Xq+3aK9YfOHAwixZ5YmPz\n4b9QFSW1auyZGkX7rNvTpwn8tWMOoWe+p2vV9YppEFXxLVkQ1M22E1LuRBXtxJV1LCWMaP/mv5ZE\nbGvRx7YC3L17B29vD3R19Vi5cq3SMnWKbY2KeoK2tjbz5s0mLi6GAQMG0a1bTyIjIzEwuMWvv4ag\nq6uLi8ssLC0rUqVKVaKjo0hLS0NPL/ffDXWhVo1d+2wUWibmSBuZFb5yAWQyGb/8EsKcOa7ExcWi\noaXDs2dPi6hKQRCKk4htLf7YViurOmzbtouQkD2sWrWMWbNeJ7OpU2yrTCYjOjqKoKDNpKen4eBg\nT8uWrZHJZBgZlWXlyrUcOrSf1atX4O29GAATExPi4+NUfpnhfahVY9d4kY5E//3S3cLDHzNrlgu/\n/34IAMvabbEesoyxY+sXRYmC8NF4myProiRiW3MrytjWU6f+pmXL1mhpaWFj04WQkN259qcusa3G\nxibUrVsfPT099PT0qFmzFhER4RgbG9OkifwLVcuWbdixY0ue46qu1OrmOb30TMjnL/2buHjxPO3b\nt+L33w9Rtmw5/P1X0nvKL5S3sCrCKgVB+FBEbCuvti262NZffw1RNNkbN64rvjjl/JzqEtvaoMFn\n3Lt3l/T0dF6+fMnjx4+pWLEyrVu35ezZ0wDcuXNL6TMmJCRgYqLemR9qdcRuFpsKn7z79g0bNqJS\npUrUqVMPX9+lWFhYKtLdBEFQPyK2Va4oY1snTXLBz8+L3bu/VzzulpM6xbZWqlSZ4cNHMXGiIxIJ\n2NnZU6FCBWxsuhIQ4Mf48Q5oamopziZERIRjbm6u1tfXQc1iW60jVzKkUnO+GfbZG22SlpbG2rWB\njBw5mgoVjAF49uyp0nWu94ltLW1E1OWHIca5+IkxLl6lNbZ11aoAGjRoRJcu3VRdikKpj21NTcsk\nU/vNSj59+iSdO7fD19cLT8/X3zj/e/OKIAiC8HZKY2zr3bt3iImJKVFN/V2p1al4AD29gkt+/vwZ\nCxd6KG6GsLL6FDu7YR+iNEEQhI9CaYxttbKqo7gzXt2pXWMvUyb/U+b79v2Km9sMoqOj0NbWZvJk\nF6ZOnZHvM6iCIAiCUNqoVWMf8HVdmlXKe+a5Gzeu4+BgD0Dz5i1ZtiyQunXf77lIQRAEQVA3atXY\nq1cvh7HR6ztGZTKZ4nnKBg0aMm7cBD75pBYjR45GQ0Otbh8QBEEQhCKhVt3PMiIRUuR3sYeG3uHL\nL3tx+vRJxXIvLz8cHBxFUxcEQRA+WsXaAX18fBg8eDC2trZcvXpVadmpU6cYOHAggwcPZs2aNW/0\nfpWXXSLzRjRLl/rSuXM7zpw5xeLFi4qjdEEQBEFQS8V2Kv7cuXM8fPiQ4OBgwsLCmDNnDsHBwYrl\n3t7ebN68GQsLC+zt7enRowe1a9cu4B3hcvxDFk74kjsP7wJgbz+C+fMXFtdHEARBEAS1U2yN/fTp\n03TtKk9Jq1WrFs+fPycpKQlDQ0MeP35MuXLlqFhRfiNcx44dOX36dIGN3TkEgk5tQIaMmjVrsWxZ\nYKHJP4IgCILwsSm2xh4XF0eDBg0UPxsbGxMbG4uhoSGxsbEYGxsrLcsv0Sjb5qsmoPGcxt0m0aTX\nDPZF6LFPTAcrCIIgCEo+2F3x7ztzbWpiXOErCe/tXaYvFN6eGOfiJ8a4+IkxLpmK7eY5c3Nz4uJe\nN+OYmBjMzMzyXBYdHY25uXmu9xAEQRAE4e0UW2Nv164dhw7JM89v3LiBubk5hoaGAFSpUoWkpCTC\nw8ORSqUcO3aMdu3aFVcpgiAIgvDRKNZ0N39/fy5cuIBEIsHDw4ObN29iZGREt27dOH/+PP7+/gB0\n796d0aNHF1cZgiAIgvDRUJ/YVkEQBEEQCiWmaBMEQRCEUkQ0dkEQBEEoRUpkYy/qqWiF3Aoa4zNn\nzjBo0CBsbW1xc3MjKytLRVWqt4LGOFtAQADDhg37wJWVHgWNcWRkJHZ2dgwcOJD58+erqMLSoaBx\n/u677xg8eDB2dnYsWiSm+H5XoaGhdO3alZ07d+Za9tZ9T1bCnD17VjZ27FiZTCaT3bt3TzZo0CCl\n5b169ZI9efJElpmZKbOzs5PdvXtXFWWqtcLGuFu3brLIyEiZTCaTTZo0SXb8+PEPXqO6K2yMZTKZ\n7O7du7LBgwfL7O3tP3R5pUJhYzx58mTZ4cOHZTKZTObp6SmLiIj44DWWBgWNc2JioszGxkaWkZEh\nk8lkslGjRskuX76skjrVWXJyssze3l7m7u4u27FjR67lb9v3StwRe35T0QJKU9FqaGgopqIV3k5B\nYwwQEhKCpaUlIJ8V8OnTpyqpU50VNsYAfn5+TJs2TRXllQoFjXFWVhYXL16kc+fOAHh4eFCpUiWV\n1arOChpnbW1ttLW1SUlJQSqVkpqaSrly5VRZrlrS0dFh48aNec7n8i59r8Q19ri4OCpUqKD4OXsq\nWiDPqWizlwlvrqAxBhTzDcTExHDy5Ek6duz4wWtUd4WNcUhICC1btqRy5cqqKK9UKGiMExISKFOm\nDL6+vtjZ2REQEKCqMtVeQeOsq6vLxIkT6dq1KzY2NjRu3JhPPvlEVaWqLS0tLfT09PJc9i59r8Q1\n9v+Siafxil1eYxwfH4+TkxMeHh5K/6iFd5NzjJ89e0ZISAijRo1SYUWlT84xlslkREdHM3z4cHbu\n3MnNmzc5fvy46oorRXKOc1JSEuvXr+fgwYMcPXqUK1eucPv2bRVWJ0AJbOxiKtriV9AYg/wfq6Oj\nI1OnTsXaWiTovYuCxvjMmTMkJCQwdOhQnJ2duXHjBj4+PqoqVW0VNMYVKlSgUqVKVKtWDU1NTdq0\nacPdu3dVVapaK2icw8LCqFq1KsbGxujo6NC8eXOuX7+uqlJLpXfpeyWusYupaItfQWMM8mu/I0aM\noEOHDqoqUe0VNMY9e/Zk//797N69m9WrV9OgQQPmzJmjynLVUkFjrKWlRdWqVfn3338Vy8Up4ndT\n0DhXrlyZsLAw0tLSALh+/To1atRQVaml0rv0vRI585yYirb45TfG1tbWtGjRgiZNmijW7dOnD4MH\nD1ZhteqpoL/H2cLDw3Fzc2PHjh0qrFR9FTTGDx8+ZPbs2chkMj799FM8PT3R0ChxxzJqoaBx3rVr\nFyEhIWhqatKkSRNcXV1VXa7auX79OosXLyYiIgItLS0sLCzo3LkzVapUeae+VyIbuyAIgiAI70Z8\nfRUEQRCEUkQ0dkEQBEEoRURjFwRBEIRSRDR2QRAEQShFRGMXBEEQhFJES9UFCMLHIDw8nJ49eyo9\nRggwZ84c6tWrl+c2gYGBSKXS95pP/uzZs0yYMIH69esDkJ6eTv369Zk7dy7a2tpv9V5//fUXN27c\nYPz48Vy6dAkzMzOqVq3KokWL+PLLL2nYsOE71xkYGEhISAhVqlQBQCqVYmlpycKFCzEyMsp3u+jo\naO7fv0+bNm3eed+CUNqIxi4IH4ixsbFKnlf/9NNPFfuVyWRMmzaN4OBg7O3t3+p9OnTooJi0KCQk\nhC+++IKqVasyd+7cIqmzX79+Sl9ili5dyrp165g5c2a+25w9e5awsDDR2AUhB9HYBUHFwsLC8PDw\nQFNTk6SkJKZOnUr79u0Vy6VSKe7u7jx48ACJREK9evXw8PDg5cuXLFy4kIcPH5KcnEyfPn1wcHAo\ncF8SiYRmzZpx//59AI4fP86aNWvQ09NDX18fLy8vLCws8Pf358yZM+jo6GBhYcHixYvZt28fp06d\nokePHhw8eJCrV6/i5uZGUFAQ48ePJyAggLlz59K0aVMARo4cyahRo7CysmLBggWkpqaSkpKCi4sL\nbdu2LXRcmjRpwu7duwG4cOEC/v7+6OjokJaWhoeHB2XLlmXFihXIZDLKly/P0KFD33o8BKE0Eo1d\nEFQsLi6OKVOm0KJFCy5fvoyXl5dSYw8NDeXKlSscOHAAgN27d5OYmEhwcDDm5uZ4e3uTmZnJoEGD\naNu2LXXr1s13X+np6Rw7doyBAweSmpqKu7s7e/fuxdLSkp07d7JixQpmz57Nd999x4ULF9DU1GT/\n/v1Kc1V369aN7du3M378eNq0aUNQUBAAffv25dChQzRt2pT4+HjCwsKwtrZm/PjxODg40Lp1a2Jj\nYxk8eDCHDx9GSyv/Xz9SqZR9+/bx+eefA/LgHE9PT+rWrcu+fftYv349q1at4quvvkIqlTJq1Cg2\nbdr01uMhCKWRaOyC8IEkJCQwbNgwpddWrlyJmZkZS5YsYfny5WRkZPDs2TOldWrVqkWFChVwdHTE\nxsaGXr16YWRkxNmzZ4mKiuL8+fMAvHz5kkePHuVqZKGhoUr7tbGx4YsvvuDWrVuYmJhgaWkJQMuW\nLdm1axflypWjffv22Nvb061bN7744gvFOgXp3bs3dnZ2uLm5cfDgQXr27ImmpiZnz54lOTmZNWvW\nAPJ53OPj47GwsFDa/tdff+XSpUvIZDJu3rzJ8OHDGTt2LACmpqYsWbKE9PR0EhMT88z8ftPxEITS\nTjR2QfhA8rvGPn36dHr37s3AgQMJDQ3FyclJabmuri7ff/89N27cUBxt//DDD+jo6DBx4kR69uxZ\n4H5zXmPPSSKRKP0sk8kUr61atYqwsDD+/PNP7O3tCQwMLPTzZd9Md/XqVQ4cOMDs2bMB0NHRITAw\nUClTOi85r7E7OTlRuXJlxVG9q6srCxYsoE2bNhw7doxvv/021/ZvOh6CUNqJx90EQcXi4uKwsrIC\nYP/+/bx8+VJp+bVr1/jpp59o0KABzs7ONGjQgH///ZdmzZopTs9nZWXh6+ub62i/IDVq1CA+Pp4n\nT54AcPr0aRo3bszjx4/ZunUrtWrVwsHBgW7duuXK2JZIJGRkZOR6z759+7J3716eP3+uuEs+Z50J\nCQksWrSo0No8PDwIDAwkKipKaYwyMzM5ePCgYowkEglSqTTXft5lPAShtBCNXRBUzMHBAVdXV0aP\nHk2zZs0oV64cfn5+iuXVqlXj0KFD2NraMnz4cMqWLUvTpk0ZOnQoBgYGDB48mEGDBmFkZET58uXf\neL96enosWrSIadOmMWzYME6fPs3UqVOxsLDg5s2bDBw4kBEjRhAREUH37t2Vtm3Xrh0eHh4cPnxY\n6fXu3bvz22+/0bt3b8Vrc+fO5ciRIwwZMoSxY8fSunXrQmurWLEijo6OzJs3DwBHR0dGjBiBk5MT\nX331FZGRkWzdupXmzZsTEhLCihUr3ns8BKG0EOlugiAIglCKiCN2QRAEQShFRGMXBEEQhFJENHZB\nEARBKEVEYxcEQRCEUkQ0dkEQBEEoRURjFwRBEIRSRDR2QRAEQShFRGMXBEEQhFLk/wiaEHxIxg63\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "####################\n",
    "#Plot ROC\n",
    "####################\n",
    "\n",
    "\n",
    "# Plot linewidth.\n",
    "lw = 2\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "n_classes = 3\n",
    "fpr = dict()\n",
    "tpr = dict()\n",
    "roc_auc = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr[i], tpr[i], th = roc_curve(y_test[:, i], y_pred_test[:, i])\n",
    "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr[\"micro\"], tpr[\"micro\"], _ = roc_curve(y_test.ravel(), y_pred_test.ravel())\n",
    "roc_auc[\"micro\"] = auc(fpr[\"micro\"], tpr[\"micro\"])\n",
    "\n",
    "# Compute macro-average ROC curve and ROC area\n",
    "\n",
    "# First aggregate all false positive rates\n",
    "all_fpr = np.unique(np.concatenate([fpr[i] for i in range(n_classes)]))\n",
    "\n",
    "\n",
    "# Then interpolate all ROC curves at this points\n",
    "mean_tpr = np.zeros_like(all_fpr)\n",
    "for i in range(n_classes):\n",
    "    mean_tpr += interp(all_fpr, fpr[i], tpr[i])\n",
    "\n",
    "# Finally average it and compute AUC\n",
    "mean_tpr /= n_classes\n",
    "\n",
    "fpr[\"macro\"] = all_fpr\n",
    "tpr[\"macro\"] = mean_tpr\n",
    "roc_auc[\"macro\"] = auc(fpr[\"macro\"], tpr[\"macro\"])\n",
    "\n",
    "# Plot all ROC curves\n",
    "plt.figure(1)\n",
    "plt.plot(fpr[\"micro\"], tpr[\"micro\"],\n",
    "         label='micro-average ROC curve (area = {0:0.2f})'\n",
    "               ''.format(roc_auc[\"micro\"]),\n",
    "         color='deeppink', linestyle=':', linewidth=4)\n",
    "\n",
    "plt.plot(fpr[\"macro\"], tpr[\"macro\"],\n",
    "         label='macro-average ROC curve (area = {0:0.2f})'\n",
    "               ''.format(roc_auc[\"macro\"]),\n",
    "         color='navy', linestyle=':', linewidth=4)\n",
    "\n",
    "colors = cycle(['aqua', 'darkorange', 'cornflowerblue'])\n",
    "for i, color in zip(range(n_classes), colors):\n",
    "    plt.plot(fpr[i], tpr[i], color=color, lw=lw,\n",
    "             label='ROC curve of class {0} (area = {1:0.2f})'\n",
    "             ''.format(i+1, roc_auc[i]))\n",
    "\n",
    "plt.plot([0, 1], [0, 1], 'k--', lw=lw)\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Some extension of Receiver operating characteristic to multi-class')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "MfllenWblVYP"
   },
   "outputs": [],
   "source": [
    "#Introduing the Metric F1 Score\n",
    "#######################\n",
    "def f1_score(y_true, y_pred):\n",
    "    y_true = tf.cast(y_true, \"int32\")\n",
    "    y_pred = tf.cast(tf.round(y_pred), \"int32\") # implicit 0.5 threshold via tf.round\n",
    "    y_correct = y_true * y_pred\n",
    "    sum_true = tf.reduce_sum(y_true, axis=1)\n",
    "    sum_pred = tf.reduce_sum(y_pred, axis=1)\n",
    "    sum_correct = tf.reduce_sum(y_correct, axis=1)\n",
    "    precision = sum_correct / sum_pred\n",
    "    recall = sum_correct / sum_true\n",
    "    f_score = 2* precision * recall / (precision + recall)\n",
    "    f_score = tf.where(tf.is_nan(f_score), tf.zeros_like(f_score), f_score)\n",
    "    return tf.reduce_mean(f_score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1448
    },
    "colab_type": "code",
    "id": "AzR1oGtcGq_9",
    "outputId": "89f12eb2-07a2-46d3-8b5a-78fc7a2a4881"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/layers/core.py:143: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "zero_padding2d (ZeroPadding2 (None, 122, 122, 3)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_15 (Conv2D)           (None, 120, 120, 64)      1792      \n",
      "_________________________________________________________________\n",
      "zero_padding2d_1 (ZeroPaddin (None, 122, 122, 64)      0         \n",
      "_________________________________________________________________\n",
      "conv2d_16 (Conv2D)           (None, 120, 120, 64)      36928     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2 (None, 60, 60, 64)        0         \n",
      "_________________________________________________________________\n",
      "zero_padding2d_2 (ZeroPaddin (None, 62, 62, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_17 (Conv2D)           (None, 60, 60, 128)       73856     \n",
      "_________________________________________________________________\n",
      "zero_padding2d_3 (ZeroPaddin (None, 62, 62, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_18 (Conv2D)           (None, 60, 60, 128)       147584    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2 (None, 30, 30, 128)       0         \n",
      "_________________________________________________________________\n",
      "zero_padding2d_4 (ZeroPaddin (None, 32, 32, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_19 (Conv2D)           (None, 30, 30, 256)       295168    \n",
      "_________________________________________________________________\n",
      "zero_padding2d_5 (ZeroPaddin (None, 32, 32, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_20 (Conv2D)           (None, 30, 30, 256)       590080    \n",
      "_________________________________________________________________\n",
      "zero_padding2d_6 (ZeroPaddin (None, 32, 32, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_21 (Conv2D)           (None, 30, 30, 256)       590080    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_5 (MaxPooling2 (None, 15, 15, 256)       0         \n",
      "_________________________________________________________________\n",
      "zero_padding2d_7 (ZeroPaddin (None, 17, 17, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_22 (Conv2D)           (None, 15, 15, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "zero_padding2d_8 (ZeroPaddin (None, 17, 17, 512)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_23 (Conv2D)           (None, 15, 15, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "zero_padding2d_9 (ZeroPaddin (None, 17, 17, 512)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_24 (Conv2D)           (None, 15, 15, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "max_pooling2d_6 (MaxPooling2 (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "zero_padding2d_10 (ZeroPaddi (None, 9, 9, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_25 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "zero_padding2d_11 (ZeroPaddi (None, 9, 9, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_26 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "zero_padding2d_12 (ZeroPaddi (None, 9, 9, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_27 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "max_pooling2d_7 (MaxPooling2 (None, 3, 3, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten_7 (Flatten)          (None, 4608)              0         \n",
      "_________________________________________________________________\n",
      "dense_20 (Dense)             (None, 4096)              18878464  \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 4096)              0         \n",
      "_________________________________________________________________\n",
      "dense_21 (Dense)             (None, 4096)              16781312  \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 4096)              0         \n",
      "_________________________________________________________________\n",
      "dense_22 (Dense)             (None, 3)                 12291     \n",
      "=================================================================\n",
      "Total params: 50,386,755\n",
      "Trainable params: 50,386,755\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "## Model Summary but now we included F1 Score as a metric as well\n",
    "\n",
    "cnn = cnn_vgg16()\n",
    "cnn.summary()\n",
    "\n",
    "sgd = tf.keras.optimizers.SGD(lr=0.005, decay=1e-6, momentum=0.9, nesterov=True)\n",
    "cnn.compile(optimizer=sgd, loss='categorical_crossentropy', metrics=['accuracy', f1_score])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1042
    },
    "colab_type": "code",
    "id": "a6d5bF9kmktm",
    "outputId": "e7ece87b-4de2-40ec-e647-639b8dca5cbc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 8500 samples, validate on 1500 samples\n",
      "Epoch 1/30\n",
      "8500/8500 [==============================] - 95s 11ms/sample - loss: 0.8540 - acc: 0.5558 - f1_score: 0.2718 - val_loss: 0.6946 - val_acc: 0.5673 - val_f1_score: 0.5665\n",
      "Epoch 2/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.6967 - acc: 0.5491 - f1_score: 0.5436 - val_loss: 0.7066 - val_acc: 0.4320 - val_f1_score: 0.4318\n",
      "Epoch 3/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.6959 - acc: 0.5468 - f1_score: 0.5398 - val_loss: 0.6870 - val_acc: 0.5673 - val_f1_score: 0.5668\n",
      "Epoch 4/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.6875 - acc: 0.5680 - f1_score: 0.5644 - val_loss: 0.6721 - val_acc: 0.5673 - val_f1_score: 0.5665\n",
      "Epoch 5/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.6685 - acc: 0.6129 - f1_score: 0.6080 - val_loss: 0.6251 - val_acc: 0.6933 - val_f1_score: 0.6902\n",
      "Epoch 6/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.6361 - acc: 0.6602 - f1_score: 0.6553 - val_loss: 0.7540 - val_acc: 0.5760 - val_f1_score: 0.5752\n",
      "Epoch 7/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.6195 - acc: 0.6742 - f1_score: 0.6726 - val_loss: 0.5626 - val_acc: 0.7167 - val_f1_score: 0.7133\n",
      "Epoch 8/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5947 - acc: 0.6974 - f1_score: 0.6961 - val_loss: 0.5259 - val_acc: 0.7480 - val_f1_score: 0.7458\n",
      "Epoch 9/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5693 - acc: 0.7214 - f1_score: 0.7203 - val_loss: 0.4906 - val_acc: 0.7813 - val_f1_score: 0.7797\n",
      "Epoch 10/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5501 - acc: 0.7346 - f1_score: 0.7318 - val_loss: 0.6874 - val_acc: 0.6300 - val_f1_score: 0.6295\n",
      "Epoch 11/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5268 - acc: 0.7551 - f1_score: 0.7546 - val_loss: 0.4964 - val_acc: 0.7727 - val_f1_score: 0.7711\n",
      "Epoch 12/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5292 - acc: 0.7487 - f1_score: 0.7476 - val_loss: 0.4909 - val_acc: 0.7820 - val_f1_score: 0.7830\n",
      "Epoch 13/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5284 - acc: 0.7552 - f1_score: 0.7549 - val_loss: 0.4913 - val_acc: 0.7700 - val_f1_score: 0.7695\n",
      "Epoch 14/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5141 - acc: 0.7661 - f1_score: 0.7661 - val_loss: 0.4654 - val_acc: 0.7967 - val_f1_score: 0.7962\n",
      "Epoch 15/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5111 - acc: 0.7622 - f1_score: 0.7626 - val_loss: 0.5722 - val_acc: 0.7207 - val_f1_score: 0.7197\n",
      "Epoch 16/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5015 - acc: 0.7712 - f1_score: 0.7719 - val_loss: 0.4659 - val_acc: 0.7920 - val_f1_score: 0.7912\n",
      "Epoch 17/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5051 - acc: 0.7740 - f1_score: 0.7722 - val_loss: 0.4776 - val_acc: 0.7867 - val_f1_score: 0.7873\n",
      "Epoch 18/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.5000 - acc: 0.7714 - f1_score: 0.7697 - val_loss: 0.4636 - val_acc: 0.7973 - val_f1_score: 0.7968\n",
      "Epoch 19/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4952 - acc: 0.7762 - f1_score: 0.7752 - val_loss: 0.5677 - val_acc: 0.7253 - val_f1_score: 0.7239\n",
      "Epoch 20/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4864 - acc: 0.7822 - f1_score: 0.7830 - val_loss: 0.4554 - val_acc: 0.8013 - val_f1_score: 0.8001\n",
      "Epoch 21/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4829 - acc: 0.7833 - f1_score: 0.7820 - val_loss: 0.5623 - val_acc: 0.7433 - val_f1_score: 0.7418\n",
      "Epoch 22/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4798 - acc: 0.7862 - f1_score: 0.7855 - val_loss: 0.5275 - val_acc: 0.7280 - val_f1_score: 0.7288\n",
      "Epoch 23/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4821 - acc: 0.7818 - f1_score: 0.7802 - val_loss: 0.5315 - val_acc: 0.7240 - val_f1_score: 0.7206\n",
      "Epoch 24/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4758 - acc: 0.7868 - f1_score: 0.7854 - val_loss: 0.4477 - val_acc: 0.8040 - val_f1_score: 0.8035\n",
      "Epoch 25/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4679 - acc: 0.7933 - f1_score: 0.7919 - val_loss: 0.5348 - val_acc: 0.7627 - val_f1_score: 0.7617\n",
      "Epoch 26/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4644 - acc: 0.7964 - f1_score: 0.7951 - val_loss: 0.4460 - val_acc: 0.8053 - val_f1_score: 0.8048\n",
      "Epoch 27/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4596 - acc: 0.7976 - f1_score: 0.7965 - val_loss: 0.4805 - val_acc: 0.7887 - val_f1_score: 0.7878\n",
      "Epoch 28/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4585 - acc: 0.7971 - f1_score: 0.7962 - val_loss: 0.4376 - val_acc: 0.8080 - val_f1_score: 0.8074\n",
      "Epoch 29/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4583 - acc: 0.7965 - f1_score: 0.7955 - val_loss: 0.4551 - val_acc: 0.7973 - val_f1_score: 0.7984\n",
      "Epoch 30/30\n",
      "8500/8500 [==============================] - 80s 9ms/sample - loss: 0.4564 - acc: 0.7986 - f1_score: 0.7986 - val_loss: 0.4491 - val_acc: 0.8000 - val_f1_score: 0.7998\n"
     ]
    }
   ],
   "source": [
    "history_lenet = cnn.fit(X, y, validation_split=0.15, epochs=30, batch_size=128, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 752
    },
    "colab_type": "code",
    "id": "epLQD9x3oEyV",
    "outputId": "a14148c3-9cee-4998-d419-eaf3816d9c35"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'acc', 'val_loss', 'val_acc'])\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFnCAYAAAC/5tBZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRBmCQZ2IiChD/H8WDAWlXCZbigAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQjgYCW8hhBDCwUh4CyGEEA7m/wNkhCY7pyPHbgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "history = history_lenet\n",
    "# list all data in history\n",
    "print(history.history.keys())\n",
    "# summarize history for accuracy\n",
    "plt.plot(history.history['acc'])\n",
    "plt.plot(history.history['val_acc'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()\n",
    "# summarize history for loss\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "dqctpDWNLvCH"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Galaxy_Zoo2_CNN.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}