{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Introduction to Convolutional Neural Networks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### The MNIST Dataset - Revisited"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import random\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams[\"figure.figsize\"] = (3,2)  # default figure width, height in inches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.datasets import mnist\n",
    "from keras.utils import to_categorical"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# return images of type float32 in the range 0-1\n",
    "def load_normalized_mnist_data():\n",
    "    (train_images, train_labels), (test_images, test_labels) = mnist.load_data()\n",
    "    # scale the pixel values to the range 0-1\n",
    "    train_images = train_images.astype('float32') / 255\n",
    "    test_images = test_images.astype('float32') / 255\n",
    "    return (train_images, train_labels), (test_images, test_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "(train_images,train_labels), (test_images,test_labels) = load_normalized_mnist_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 28, 28)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_images.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 28, 28)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_images.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dtype: float32\n",
      "Min: 0.0\n",
      "Max: 1.0\n"
     ]
    }
   ],
   "source": [
    "print(\"Dtype:\", train_images.dtype)\n",
    "print(\"Min:\", train_images.min())\n",
    "print(\"Max:\", train_images.max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5, 0, 4, ..., 5, 6, 8], dtype=uint8)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# create the one-hot target vectors\n",
    "train_targets, test_targets = to_categorical(train_labels), to_categorical(test_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 10)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_targets.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 0. 0. 1. 0. 0. 0. 0. 0. 0.]\n"
     ]
    }
   ],
   "source": [
    "print(train_targets[7])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n"
     ]
    }
   ],
   "source": [
    "print(train_labels[7])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 300x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(train_images[7], cmap='gray');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### A Basic Feedforward Network for MNIST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Flatten, Input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def build_network():\n",
    "    network = Sequential()\n",
    "    network.add(Input(shape=(28,28)))\n",
    "    network.add(Flatten())\n",
    "    network.add(Dense(30, activation='relu', name='hidden'))\n",
    "    network.add(Dense(10, activation='softmax', name='output'))\n",
    "    \n",
    "    network.compile(loss='categorical_crossentropy',\n",
    "                    optimizer='rmsprop',\n",
    "                    metrics=['accuracy'])\n",
    "    return network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "network = build_network()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
      "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
      "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
      "│ flatten_1 (Flatten)             │ (None, 784)            │             0 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ hidden (Dense)                  │ (None, 30)             │        23,550 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ output (Dense)                  │ (None, 10)             │           310 │\n",
      "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
      " Total params: 23,860 (93.20 KB)\n",
      " Trainable params: 23,860 (93.20 KB)\n",
      " Non-trainable params: 0 (0.00 B)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "network.summary(print_fn=print)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2026-04-09 16:12:47.678893: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 6ms/step - accuracy: 0.8800 - loss: 0.4371\n",
      "Epoch 2/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 6ms/step - accuracy: 0.9161 - loss: 0.3006\n",
      "Epoch 3/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 6ms/step - accuracy: 0.9200 - loss: 0.2884\n",
      "Epoch 4/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 6ms/step - accuracy: 0.9226 - loss: 0.2820\n",
      "Epoch 5/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 6ms/step - accuracy: 0.9237 - loss: 0.2793\n"
     ]
    }
   ],
   "source": [
    "history = network.fit(train_images, train_targets, epochs=5, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 6ms/step - accuracy: 0.9237 - loss: 0.2789\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.27891018986701965, 0.9236833453178406]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "network.evaluate(train_images, train_targets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9203 - loss: 0.2858\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.285776287317276, 0.9203000068664551]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "network.evaluate(test_images, test_targets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "jupyter": {
     "source_hidden": true
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def show_wrong_images(network):\n",
    "    outputs = network.predict(test_images)\n",
    "    predictions = [np.argmax(output) for output in outputs]\n",
    "    wrong = [i for i in range(10000) if predictions[i] != test_labels[i]]\n",
    "    print(f\"Misclassified {len(wrong)} test images out of {len(test_images)}\")\n",
    "    plt.figure(figsize=(12,12))  # (width, height) in inches\n",
    "    rows, columns = 5, 6\n",
    "    for i in range(1, columns*rows+1):\n",
    "        w = random.choice(wrong)\n",
    "        img = test_images[w]\n",
    "        correct_label = test_labels[w]\n",
    "        prediction = predictions[w]\n",
    "        plt.subplot(rows, columns, i)\n",
    "        plt.title(f'\"{prediction}\"  (correct: {correct_label})')\n",
    "        plt.axis('off')\n",
    "        plt.imshow(img, cmap='gray')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "A random sampling of misclassified images:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step\n",
      "Misclassified 797 test images out of 10000\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7YAAAOrCAYAAABjshrdAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAxC9JREFUeJzs3QeUFEX38OFecs5JsohkFQkCgoRXBEElSBYURVFEVKKKSBIQkWhGVFSiks0iIiAgIFFEgoIKCIJEybm/c+t/Zr/tqd6dZnZmdmr295zD69t3a3t6e2qq+3Z33Ymzbdu2AAAAAAAwVJqU3gAAAAAAAJKDxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABgtphLbU6dOWQUKFLCmT5+e0psSM9q3b2+1bds2pTcj6tH3Qo++592VK1esSpUqWSNGjAjDO5E60f+8oe+F3nPPPWfVqFEjDGuOPRx7Q4+xj76XUmrWrGk988wzyVuJHYQPPvjA9v3qkiVL1P//888/E23fsGFD1eaJJ55wxOV3JC7rEPL/Zd3BGj58uF2qVCn70qVLdqx68803k7WPxL59++yOHTvaZcqUsbNly2bnzJnTrl69uv3hhx/aV65ccbTdsGGDnSZNGnvTpk12NAjU90qUKKFibv9Kly4d346+lzJ9b9u2bXa/fv3sm266SfW9QoUK2U2bNrXXrl2rtY22vuel/23fvt3u2bOnXatWLTtjxoxJjo0Jxzvpt4MHDw56u6ZNm2bnyJHDPn78uB2rpk+fbo8fPz7Z65F9JH1QxoNMmTLZxYsXt7t06WLv3r07qvtfoL43b948u1GjRvY111xjZ8iQwS5SpIjdqlUr+5dfftHWRd9Lmb731ltv2a1bt7aLFSum3oPOnTu7tvvnn3/U+PHpp5/a0cLred/HH39s16xZ086SJYs6t5CxcPHixfE/59h79VL7sZecw+y+J37//Xd1PMqVK5edOXNmu3bt2vb3339v+5PjmIwdMgYGK+x3bOfNm2etWrUq3C9jXbx40Xr11VetRx55xEqbNq0Vq9566y3rww8/TNY6Dh8+bP39999W69atrTFjxljDhw+3rrnmGuvBBx+0BgwY4Gh78803W9WqVbPGjh1rmWDChAnW1KlTHf/k7xONGjUKy2vS97x77733rHfffTe+T/Xu3dvasWOHukr33XffGd33hIx1r732mnXy5EmrfPnyEXvd0aNHq6vsOXPmtGLVjBkz1Oc7uXcX77jjDjWOtmzZ0nr99detDh06WLNnz7ZuvfVW9b6Z2v9++eUXK3fu3NbTTz+t/r7HH3/c2rhxo3XLLbdYP//8c9hel77n3ahRo6zvv//eqlixopUuXbpE2xUqVMhq3ry5Oj6bZMiQIerzVKxYMWvcuHHq2HvjjTda+/btC8vrcez1LtaPvT7kHNGXc+zdu9eqVauWtWLFCqtfv37WyJEj1ZMWck7+ww8/ONrKuJcjRw71ukEL59WTs2fP2iVLlrRffPHFsN+xlSxffn/nzp12JJ06dSrRn50+fTrkr1exYkW7Xr16djjcfffddtasWbU73mPGjFHxkydP2qZduRPDhg1T7VauXBkfo++lTN9bt26d1o8OHz5s58+fX13B8xdNfc9L/zty5Ih94sQJ9f9Hjx4dkTu2cnVd1vXdd9/ZkSTj++XLl696XAzWXXfdpfZTcsgYIPvqjTfecMQnT56s4nIciaWx78CBA3a6dOnsxx57zBGn70W+74m//vor/qko6VeJ3bEVc+bMsePi4uxdu3bZ0SBQ/1u1apXa3nHjxiW5Ho69Vy+1H3vJOczOObp3766OQ/JEW8JtlSdXqlSporXv0aOHGm/9nyCNiju2r7zyirpC3rdvXyvcFixYYJUsWdK67rrrtJ9t375dzRPNnz+/lTlzZqts2bLanUm5st2kSRN1pSBbtmzW7bffbq1evdrRRq5axMXFWcuWLbO6d++u5vMWLVpU/ax+/fpqjtv69eutunXrWlmyZLGef/559bPz589bgwcPtkqXLm1lzJhRXc2UZ8gl7m/atGnqCrv8vlx9l3V9++236mfy9/3666/q9WU75J+8rs+uXbvUv2DJ+s+cOWNduHDBEZc7HKdPn7YWLVpkmXqn59prr1V3ZMKBvue971WtWlV9vhLKmzevddttt1nbtm3T2pvW9/LkyWNlz549oq8p/S9DhgxqrPAnd0oefvhhq3Dhwmrskc+B3MlL+Bn/448/rDZt2qhtl3FHruB/+eWXjvUsXbpUjTcff/yx9cILL1hFihRRbU+cOKGe9JD3VN7/pk2bqr+/Y8eO6vdk/Je7rHKHKlOmTFbBggWtxx57zDp27Ji2rV9//bVVr1499fsyDlevXl19doWMc7JNu3fvjh/7ZLzy2bNnjxrnA5HtFbIdCckTK0KODyb3P39yjJL36fjx42FZP33Pe98TJUqUUH3Xi4YNG6r/fvrpp5YJ5HMud5rliQG5diJ3ZMKNYy/H3oTIOaIz51i+fLl6CkByLx95vWbNmlkbNmywfv/9d+24K8f6TZs2BTEqWFbiz8Ikkwz2L7/8sjV58mTtZCEcfvzxR6tKlSpafPPmzeqkOX369Najjz6q3ih5Iz7//PP4QivyxkkbOZmSN1/avvPOO+oNlDfUv4iDJLWSJA8aNEid9PgcOXJEJcfySGCnTp3UyZOc2MmbJ7fg5fXl8UR5ZGz8+PHWb7/9pgZmn6FDh6pHeSQBe/HFF9XJ6po1a9SjS3LLXg4cTz75pDqJ9CXmCU/QJBkXf/31l6d9dvbsWbX9cgCSv/ODDz5Qjwv4v18VKlRQsZUrV6rH90wiFywkYfK/kBFK9L2r73v+Dhw4YOXLl0+Lm9z3IkX6n1xUk3Erof3796sDliQ1MvaUK1dOJbpz5sxRF7BkfDl48KAab2T5qaeeUhcZPvroIzVmSTv/fT5s2DD1e3KxUg6S8v/FpUuXrMaNG1t16tRRj0/KQUtIEisXBB966CG1/j///NN644031OdS3lPfNkubLl26qAS4f//+Vq5cuVSbb775xrrvvvvU5/e///5TUyhk7BQJL5A88MADagz7v5uRiZPH67JmzWoNHDhQJfJyoN25c6ca9yWR9iUTJvc/eb/lEU35TMkxQ5J53+cz1Oh73vve1ZJpBXKhXvper169rGi3ePFiNZbIVAx5BFnOhyTRlc9ujx49wvKaHHs59vqQc0RvznH+/HmVNPvznSfIDcHrr7/ecQNEyNgnCfFVs8NECiTceuut8ctujyKHysWLF9UjMH369NF+VrduXTt79uxaYZCEt7hbtGihim0kfORn//796vfk9/0fh6hTp472uK7cqpefTZw40RGfOnWqmoS/fPlyR1zaJXw8ViZWS7uWLVtqj/cl3NakHguQW/dX87jUyJEjHcWVbr/9dnvPnj2ubaXQVJMmTWzTSJ+Qv23r1q1hWT99L7i+l9APP/ygPr8DBw6Mqb4X6FHkUClatKgqyuDvgQceUGOKW3EQ35giha5kGxOOT/Lo2bXXXqumkfjGIt+jh1Kc78yZM451yeOU8rPnnnvOEZd1SlwK7yT0zTffOOJSzEnG2ho1aqjHm922M9DjoL7x14svvvhCFVhKOPY1btw40UfuTOt/ZcuWjf+7pEjMCy+8kOgj48lF37u6vpdQoEeRhRQDK1++vB3tjh49qvZB3rx5VZ+Tse+TTz6x77zzTtfzolDg2Pt/OPb+H3KO6M057rnnHlU0yjdNy0cKy8k2yWPv/iQne/zxx+2oeRR5yZIl1ty5c5Nd6MOro0ePqqul/lcEDh06pCYmy52A4sWLO37mexzo8uXL6rZ7ixYtrFKlSjkeTZM7BXLVw/f4mk/Xrl1dC1TJLX+5M5GQFCWRKyZyt0SKNvn+/e9//4vfV0KuosiVFrkLnCaN823x+uiSXDW5mjtmUuRBHrGTx/3kb/XdxXUj+1a22ySyP+XRSbniE65CPvS94Pqez7///qv6njwim1iJdxP7XiTJnRH/sU/6vowp99xzj7pL6c83pnz11Vfqrq7cafWRq7NypVfez61btzp+r3Pnzok+gSOPOPuPfXLXSR4rSjj2+R5H9419MgZJ0Sb5ihN5XNltOwORR6W93jGTp21kTJAndmQfyRVreVTKf+w2tf/Jkzdyp1uKb8i4J2O6HOfCgb53dX3vapnS93yPHUt/kCJF8kSHTP+S6QPy1IOvgGMocez9Pxx7yTmiPed4/PHH1ZNE7dq1U09iyZ3jnj17WuvWrUs070jO2BfyR5HlkTR55Oz+++9Xj3ZFkv/BReaOCXlMLzGS/MpjeAmf/faRziFvvFT0kkfkfOQk3I3MO/M9mucjz47Lo7ByMpXYib2Qx6Olc8lBIFJkvo/88yW5cjIrj+JJpTz/k1fZt147e7SQx8Pk0ctIPMZF37t68hj83XffrZIauYDkP/fW5L4Xaf79T8Y1uSCX1NgnZB6L2/dl+i4Eyc8TriOxsU8qvPrqDSQc++TxYZnnGWjsE4G2NRTkmNCgQQNrypQpVqtWreKrMMoUFZkrLPN8ZTqJyf1PppP4yLQY33sZrgq79L3wMaXv+c4XZGqBfNuCj5zTyMmszPeTR0X9bzCEAsfe1H3sJeeI/pyjSZMm6hsI5OK1b8qozP+Vi8tyQ8Ot/yWn74U8sZUTBkmMZI6qfyYvHyKJ+QpahIrMlZId4FaQJBwSu2PhFpfE+IYbblCl793IpO5oIQckKQcvd7llvlxCsm8TPgNvgunTp6sPriTt4ULfC44UL7r33nvVHPiFCxcmmdSY2PciSebFpvTYJ0+r+F/1lbFPxnr5HLpJ7MAbTjKX99y5c+qkLiGZk+Sb0+Of2Jrc/+Sqt1ypl/cgHIktfS+8pO+51R6INnIclKctZG68/9Nsvgtb8reEMrHl2BucWDv2knOYkXP06NFDPRUl/U5uAFauXNl6//331c/KlCmjtZc7vMGOfSFPbOWqnBSuqF27tmsHlH/z589Xj/6GitwtkCILUpgkId+jxVu2bEn0d+XkSpJsScb9SaVDOVlLTkeQ7ZLvEJRJ1kldfZB20iHl0T95wxMTzqtnvscB5C6L/xUxuWvtO/kzgUxWl8fhpQCYVIQNF/re1ZN+LgVXpNjIrFmzVCXcxJjY9yJNHjnyH/tkXJNieEmNfUKe2Ehs7PP9PFgypsn3I8qxIKkCgr5K9rKtchU3nGOfFMuSK8H+j+bKMcvX32Kt/8m47j+mhwp9L7zkc33TTTdZ0U7Ok+S8Ze3atSpxSvjkmhSxC8eFLI69Vy8Wj73kHObkHFmzZnU8USTnB3Ju4J8vypOWMo4EO4Uw5HNs5dEnSVz9/wn5Kgj5/26PviWX7Czf89o+MpBK6WqpzCyd3+3xFbm6KNW/pKR+wjvMcgIkc09l7pmcIAZL5pnImyR3QhOrSiwk0ZeDg1Qmk87mtq2+jpHYVzd4Lb0tjym6kasn0on9q0tLx5e7HOH6upxwkLmDsp98XzsSTvS9q/uqKamy98knn6g5gHLlOCkm9r1Ik/4nSWHCUv4ylsiYItXf/cfFhGOKjMk//fSTtWrVqvifyZg0adIk9Xhuch5TkrFPEkippOx20uQbx2T8la/4kS9tl/fabTt9Y19iCZrXr1yRK8OyTjmpS2jmzJnqv/4VGE3qf75HzBKSY5qcxLrNsw4F+t7Vfd3P1ZC+LmOqCX1PyCPH8nmXquo+8tmRpwVkHAnHBWaOvRx7yTmiP+dIrKL5vHnz1NcRSi2OhKRKsgh67LMjJJxVkX1fZi6vsWPHDkd806ZNqkqfVOvr37+/PWnSJPv555+3b7rppvg2W7ZsURUKixQpYo8YMcIeNWqUqv6ZMWNGe/Xq1VpVZLcqo1I1TKqH+ZNqY02bNlVVX9u3b2+//vrr9oQJE+xu3brZefLkcaxLqsLK+qWatFQJk7ZS2TRhtVH5omNZ17Bhw+yZM2faixcvvuoKZU8//bRdrVo1VS1T9sfLL79sV69eXb32k08+qbWXbcmSJYtW0SyaSZVYef+k4mq40fe8973x48erfibV8KR6n/8//y8fN63vSX+Tz6b881UElcrcsiyf53BYt26dep2FCxc64n///bddqFAhtf+k+vE777xjDxkyRI1Tx44dU20OHDhgFyxY0M6ZM6caf+T9qVy5shpj5s2bF78uX1Xk2bNna68vlV1l/HTz2GOPqd+TqsKy7jfeeEONP4ULF3as67333lPtKlWqZL/00kv222+/rcZIGf98XnnlFdWmV69e9owZM+zPPvvsqivTHj58WO0Tqbj41FNPqX0i25g2bVq1X86fP29s/ytQoIDdoUMHdfyScb1fv37qGJMpU6b4SpihRt+7uqrI0md944P0wZtvvjl++eeff3Y9ruzcudM2gVRLl89Q+vTp7b59+9qvvfaaOq+Qz9ZXX30Vltfk2MuxNzHkHHFRk3P89ddf9i233GIPHz5cHevlGJ45c2Y1/rkdW3v06GEXL17cUZ35asRMYisnJPny5VM7358krlLSWspNy0Fevg7B/6tFNmzYoL7yQZJgOZFp0KCB/eOPPzraBJPYigsXLqiTDfm5JFu5c+e2q1atag8dOtT+77//HG0nT56s3mxfO1nvokWL4n8uJ6LytRfy9RiyLQnLcHvtZN9++6199913q5NLOQjJumrXrq3+PreOJF/D0alTJ9sUsk/lfb733nsj8nr0Pe99z/fVMIn98/9qHNP6nmx/Yn9bsF+H5MWNN95oP/zww1pcvuZMDlT58+dXY4pcsJNxOGECJ19zJl+V4Bsf5QAkX4mTULCJrZAkS8Y7OZDJWHPDDTfYzzzzjPpKNf+TfjnASrscOXKo7ZADqY9c9LjvvvvUdvrvz6tJLiTh79Kli/pKI0ku5Kt/unbtah86dEhra1L/Gzx4sLpgKceNdOnSqfFdTmw2b94c1tel73nve0mNf3L8Tahdu3bqqwVNcvDgQfU3ygm0jDfy+ZGv9woXjr0cexNDzjEqanIO+Tqw5s2bx19UlmPvs88+65rUys1AOSbLjbdgRSyxjYQXX3xR7TD/75hF8DZu3Kiu1sh/Qd+LJPqed1OmTFEHHt+dWND/IoW+F3r//POPusi0YMGCMKw9tnDeF3oce+l7KWX+/Pnq4rb/he+rESf/Y8UI+S41KRg1fvz4iMytTA1k/oI8f+8/Jw1O9D36XkqSz+iNN96oKoAPGDAgRbclVjD2eUPfCz35Wozvv/9ezX9H0jj2hh5jnzf0PSss8+Zvu+0265VXXgl6HTGV2AIAAAAAUp+QV0UGAAAAACCSSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgtHReG8bFxYV3S2CcSNUdo+8hpfoe/Q9uGPuQUhj7kJIY+xDtfY87tgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMli6lNwAAEDuyZcumxXr16uVYrlWrltbmzjvv9LT+Z5991rE8evToq95GmKVatWpa7JtvvtFiefPm1WLPP/+8Y3nkyJEh3joAQLTgji0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADBanG3btqeGcXFWJGXMmFGLlS9fXot17txZi2XKlMmxvGnTJq3NiRMntNgNN9ygxS5duuRYnjdvntZm3759WuzQoUNWrPPYdZIt0n0P0S9SfU/Q/xLXqFEjLTZlyhQtlj9/fsfy0aNHtTbbtm3TYjfeeKMW279/v2O5SpUqWptz585Z4cTYF17p0jnrWn7//fdamzp16nhal/+xPl++fAGP89GMsc9dp06dAo5DXv3666+O5Zdffllr8/HHH2uxy5cvW7GOsQ/R3ve4YwsAAAAAMBqJLQAAAADAaCS2AAAAAACjRc0c2549ezqWH3jgAa1N5cqVrWi0ceNGLda0aVMtdvDgQSuWMNcivFq1auVYvvfee7U2d9xxR8D5jG7vldvn+ciRI1qsa9euWmz+/PlWSmOeWeS59bVp06ZpsSxZsmixOXPmOJZHjBihtdm5c6cWe/bZZ7VYqVKlHMuPPfaYFWmMfeH14YcfBjwfCFbbtm0D9s9oltrGvjRp9PsvrVu31mJvv/22YzlXrlxh3a4hQ4ZosWHDhlmxjrEvOrn19/bt2zuWBw4cqLUpVKiQpxpENWrUcCz/9ttvVqQxxxYAAAAAkCrwKDIAAAAAwGgktgAAAAAAo5HYAgAAAACM5vwW9AhZu3atFrv55psdy2nTpg164vDp06cdy2fOnPH0e24Tpq+77rqAv+e/7aJs2bIxXzwKgVWqVEmLde/eXYu1adNGi+XOnTtgEY2LFy9qsWXLlmmxvHnzJrmcWBEBt+2PhuJRCD///vDSSy9pbdzG1h49emix2bNnB7UNW7Zs0WKvvvpqUOtCdOrQoYMWa9euXcjWf/78ecfy9u3bQ7ZuhF/OnDm12MyZMwP+3rp167TYmjVrtFjt2rWDKlR64403BmwDhEPVqlW12PTp07VY6dKlA67LLa9Kl05PDfPkyWOZgju2AAAAAACjkdgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaClSPOr6668PWJDmv//+C6pggDhy5Ihj+fDhw55+799//9Vi586dC/h7J0+e1GJeXxPmKl68uBZ7+umnHcvdunXT2mTOnFmLbd68WYt99tlnAYthLFiwQIsdOHBAi2XKlMmxnDFjRk+FNZB6ZcuWLWCxlF9//TVkhaLcfPnll1osR44cVz1GI/LcxrlHHnlEi40dO9ZT8ZJg+fcPt4JkiM4xR4waNcrT7/7888+O5datW2tt9u7dq8VKlCihxb7++uuAxUCzZMmixbJmzRqwmCmQlJYtWzqW+/bt66lgrds53WG/POS3337T2tSqVUuLXblyRYudPXvWMgV3bAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFSpHjUtddeq8WOHTtmpbRGjRoF9Xvz5s3TYlu3bg3BFiFauBWYcCu2VKBAAcfyqVOntDZPPPGEFps0aZIWu3z5shUq/hP/4+LiwlqwBUiKW6GLe++9V4vdeuutWqx58+aO5fvvv19rs2zZMt6ACKtatapj+aWXXtLa3HHHHRHcIpjG/7MtHn74YS3mVkjRv1iUW6EoN7t37w54jP7uu++0Nnfeeaenc0j/wqhAUsU/b7/99oBF+LwUnRKrV6+2Epo6darlhdsx1a3wVLTiji0AAAAAwGgktgAAAAAAo5HYAgAAAACMliKT6qJhPm2TJk202CeffBKy5+QRW9y+7N3ty9jHjh3rWH7nnXe0Njt37rRS2kMPPaTF3Obdzpo1K0JbhGjjP+farX9cuHAhqPmXQ4YM0do0bdrU07o2bNjgWN62bZun30N450i//fbbjuVq1aoFvX7/eY9ffPGFp1oFMNsrr7yixU6cOKHFmjVrpsUOHjwYtu0CrkaWLFk85Rd33323Frty5Ypj+YMPPtDaPPLII562Y5bf+VvDhg091aP47LPPLJNxxxYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABgtRYpHRVr9+vU9FcVxKwbkr3///jE30RqB7du3z1MBshUrVkTl7syQIUPAYljjxo3TYjt27AjrdiF6nT17NmBxlooVK2qxiRMnarHOnTsn2R8T+/L4Dz/8UIutXbvWsXz69GmtDcLr4sWLWmzmzJmO5WuvvVZrkzdvXi02bdo0LTZo0CDHcp48ebQ2FI+KPddcc40WO3r0qBajUFTydezY0bE8ffr0EKw1dapXr55j+aWXXtLa1KhRw1NhtGeeecaxPGfOHE/b0KhRo4CvucPlfM7/9WIBd2wBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRYrJ4VKVKlRzLw4YNC6pQlFi5cqVj+d1339Xa2LZ91dsIs3z88ceWKdKk0a9XLV26NGBhF7fiPUi9du/e7Vg+efKk1qZIkSJarGvXrgHX1atXL63NF198ocUuX77seXsROVeuXNFi48ePdyzPmzdPa1OsWDEttnr1ai126dIlx3LNmjWD3FKYbuvWrSm9CUYpW7asFmvatKkWO3bsWIS2KLYULFhQi82ePTtgsTs3jRs31mK///67Y/nIkSOe1rV+/fqAuUmZMmU8Ha/dilr99ttvlim4YwsAAAAAMBqJLQAAAADAaCS2AAAAAACjkdgCAAAAAIxmfPGoXLlyabEZM2YkWUwqMefOndNi99xzj2P5+PHjV72NQCT17dtXi/kXX7nrrru0Nv/9919Ytwtm8e8jpUuX9vR7/oXKxP333+9Y3r9/fzK3DtHOv2BYYjEv6tatG4ItQrSLi4vTYiNHjkyRbTGB25jsVrStXLlyWqxTp05h265YUbVqVS02ZMgQLeZfLGrmzJlam6+++spT4bxgueU5R48edSwXLVpUa3PzzTdrsWuuuUaLUTwKAAAAAIAI4VFkAAAAAIDRSGwBAAAAAEYzao6t25ce79q1S4vlzJkz4LpWrlypxZo1a6bFatWq5Vhu1aqVFQ02b96c5Lxicfjw4QhuEVJChw4dtNiwYcO02NChQx3L33zzTVi3C9Erb968Wmz69OlarFGjRkHNiZs1a5YWY04tUopt2+x8g9+rDz74QIsVKlTISo0qVqwY8DheuHBhT+cEbuM0rIC1SNxi+/btCzjPef78+UHt3sqVKwesWSF69eoV1Nh36tQpLbZs2TLLZNyxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARjOqeNSbb74ZVKGoCxcuaLG1a9dqsUWLFmmxKlWqWCZo3769FuvXr5+nolkwQ4YMGTy9x1OmTNFiY8aMcSxTUCV1uPHGG7XYiBEjtFjdunW1WLB9xO3L3YGU8v7777PzDeFWnPPixYsR346ePXsGbLN69WottmbNmqBeL106/VS8RIkSWmz27NkBC0W9++67nsb8y5cvB7GlsatmzZqezq9OnjypxYYPH+5Y/uGHHzy9ZqVKlbTYwIEDHctFixbV2tSoUcPT8dr2i61atcpT0SnTcccWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYLc72WCEkLi7OiqTSpUtrsR07dqT4drk5ePCgFjt69GhE90/69Om1Nlu2bPFUTCZYkSpAFA3vcTQUi/rss888FTerU6eOFvvtt9+sWBLJ4lcm9b8cOXIEHAOKFCkS1LhWoEABrc3ff//tqcjf6NGjrVjC2BdZDRo00GKLFy/29Ls7d+50LJcpU8YyGWNf+O3fv9+xXLBgQa3NvHnztFibNm0CrjtLlixa7IMPPtBirVu3DriuTz75xFPxqCVLllihEqtj30svvaTFnnnmGS12+vRpLTZ+/HjHct68eT2dq5UsWVKLufW1YPeX7fdeufWpBQsWWKbw2ve4YwsAAAAAMBqJLQAAAADAaCS2AAAAAACjkdgCAAAAAIyWLqU3IBKTybdt26bFfv75Zy12+fJlTxPx/e3atUuL7du3zwqnMWPGOJZ79+6ttalUqVJYtwHh9eqrrwYsPnD33XfHfKEoeHf//fc7lnPnzu2psN1zzz2nxdavX+9YXrNmTcBiVWLq1KmetxfwYtOmTVrsr7/+8lSM5dNPP2UnI1GPP/64FnMbN4PVpEkTx/Lzzz+vtbn11ls9revkyZOO5WeffVZrs3fv3qveRrgX4urevbsWy549uxZ74YUXQpbTfPHFF47ljRs3Bv163f22/6uvvrJSA+7YAgAAAACMRmILAAAAADAaiS0AAAAAwGhRO8fWbR7YypUrtVjRokW12KRJk5L88mRx7tw5y2SnTp1K6U1ACBUqVEiLtWrVyrHcsWNHrc1PP/3E+5BK5cqVK+Bc2SxZsmht2rdvr8XWrl2rxT777DPHcrp0+uFi6NChWuzAgQNJbDVw9Y4dO+Ypdu2112oxt88A4FOqVCltZ2TIkCHgDnKrZdG3b18tVr169aDm0x45ckSLdejQwbHMfNrQcau7069fPy02cOBALWbbdsD1b9iwQYsNHz48YO2AhQsXepqbu2PHjoC5UGrBHVsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGC0ONvLrOdEJisjMvLnzx9wortb8aEvv/xSi91zzz0h2y6PXSfZYq3vFS5cWIstWrQoYAG1u+++W2vz33//WalRpPpetPS/9OnTa7G33npLi3Xp0sWx/N577wUsMCUGDx4c8MvdJ0+erLXp1q2blRox9qW89evXa7Gbb75Zi50+fdqxXK5cOa1N6dKltdiyZcusaJTaxr5QcnvvlyxZosUKFCgQcF27d+/WYiVKlAhqu9wKRd1///1azK2QUKQx9oXXJ598kmQR0cRUqFDBU4Ezk3nte9yxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARkuX0hsApzRp9GsNr7/+uhbzLxb1999/a22efvppdm8Uuv322z0Vtbj++usdy6m1UBQsK2PGjJ760cGDBx3Lzz//vNbmxRdfDFgoSmzdutWx3KNHD94KJCpr1qwBCxo2adJEixUpUiTgXq1Tp44Wu/HGGz0VO/Lfrm3btmltMmXKpMVGjx6txQYNGuRYvnz5chJbjWiTJUuWoApFuQm2UJRbEb6PP/5Yiy1evDio9cMcbufoDRo0CPh7M2fOjPlCUcnBHVsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGC0ONu2bU8NXYoypAYVKlQIWKTpxIkTntaVOXNmx3Lt2rW1No888ogWa9u2bcB1z549W4u1a9fOCiePXSfZYq3vufWXV199VYsNGTIkYKGSypUra7Fz585pse3bt1uxJFJ9L1r6X+PGjT0VHBk+fLhjuW7dup4K+OzatUuLNWzY0LG8b98+z9sb61L72FexYkUtNnLkSMfy3XffbZnCbT+7vcf+hYYOHz5sRVpqG/tCadmyZZ4KkwVrx44dWmzu3LmO5REjRng6Zker1D72Batr165abMyYMQGL3Q0bNkxrM2HCBC2WGoqL2h77HndsAQAAAABGI7EFAAAAABiNxBYAAAAAYLR0Kb0B0SR9+vRabPny5Y7l06dPa21ef/11LXblypWAczmaN2/uabvc1jVp0iTH8rPPPutpXYis9u3bB5xDIcaNG6fF8ubN61ju1q2b1ua5557z9MXu99xzj6ftRXRymz+TLVs2LTZw4EDHcvbs2bU2K1eu9NQ/UsOcHQSnWbNmWiyUc2r959e51bY4evSop3Vt3rzZsfzTTz95+r3du3drsSNHjnj6XaQ8t1oCbnPDg7Vp0yYt1qZNGy32xx9/hOw1YYaHHnrI03xat+PzRx99lOS5vuDYnDTu2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKPF2R6/8TbWvizZK/8CT24TwK+77rqQvd769eu12HvvvafF3nnnHSul8UXdgQuQub2flSpV0mK//PKLFitQoIBjuWDBglqbffv2eSpgsXr1aiuWRKrvRcvYV61aNS22Zs0aLXby5EnH8rvvvqu16d+/vxa7dOlSsrcxNUntY1+pUqUCFrLr2LGj1iZz5sxabMqUKVps3rx5Afv6wYMHrdQotY19wcqUKZOngk/XX3+9Fnvttdccy0OGDNHaXLx4UYudOXPGinWpfexzU7hwYcfy5MmTtTYNGzb0VACvSpUqAQvnpVa2x77HHVsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ikddpTx58mixdu3aabEbbrgh4Lq++uorLbZw4UJPRQqiAUUErIDFUU6fPh30/r18+bJj+cMPP9Ta9OnTR4udOHHCinUUUEFq6H8mFVBBZDD2Be+uu+7SYjly5NBiCxYscCyfPXs2Ga8aWxj7dMOGDXMsP/nkk1qbbNmyabFbbrlFi23YsCGZ71DsongUAAAAACBV4FFkAAAAAIDRSGwBAAAAAEYjsQUAAAAAGC1dSm+AaY4eParF3n777RTZFkQf/4JPK1as0NrUqVNHi82ZM0eLffDBB47lr7/+OiTbCABAavPll1+m9CYgBpUpUyZgoah3331Xi1EoKjy4YwsAAAAAMBqJLQAAAADAaCS2AAAAAACjxdkev/GWL4qHP76oG7He9wRjH1Kq/9H3kFJ9j/4HN4x9iPa+xx1bAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDR4mzbtlN6IwAAAAAACBZ3bAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFiKrG9cuWKValSJWvEiBEpvSkx47nnnrNq1KiR0psR9U6dOmUVKFDAmj59ekpvSsyoWbOm9cwzz6T0ZhiBsS/02rdvb7Vt2zYMa44tjH2hx9hH/0tJ9D9vOO5Gac5hB+GDDz6wfb+6ZMkS9f///PPP+J/PnTvXbtu2rX3ttdfamTNntsuUKWP37t3bPnbsmLYu+V1ZnyhRooQ9ePBgO1jTpk2zc+TIYR8/ftyOVdOnT7fHjx+frHWcOXPG7tKli12xYkW1v7JmzWrfeOON9oQJE+wLFy442v7zzz92xowZ7U8//dSOBoH6nr+GDRuqNk888YQjLr8jcVmHfz8MxvDhw+1SpUrZly5dsmPVm2++max9lNDOnTvtDh062Pnz57czZcpkly5d2n7++ecdbebNm2dnyZJF9cFowdhn9tgn5PjQr18/1eek7xUvXlyNh7t373a027Bhg50mTRp706ZNtilj36JFi+z69evbefPmtXPmzGlXr17dnjJliqMNY1/KjX2///673apVKztXrlzq3Kh27dr2999/r7Uzcezzx7E3dFJ7/wvU97Zv32737NnTrlWrljpfTapvknPEfs4Rlju2jz76qLVt2zarU6dO1muvvWbdeeed1htvvGHVqlXLOnv2rBUuo0ePVlfZc+bMacWqGTNmWBMmTEjWOuQ9+PXXX62mTZtaI0eOtMaMGWPddNNNVq9evazOnTs72hYqVMhq3ry5amOaefPmWatWrQr761y8eNF69dVXrUceecRKmzatFaveeust68MPP0z2ejZt2mRVrVrV+vnnn60+ffpYr7/+utWhQwdr//79jnbS73LkyKFe1xSMfdE99skV9jvuuEP1qZYtW8b3vdmzZ1u33nqrdfLkyfi2N998s1WtWjVr7Nixlgk+++wzq1GjRtaFCxesIUOGqCeXMmfObD3wwAPW+PHjw/KajH3e7d27V50DrVixwurXr5869srdbnnPfvjhB+PHvoQ49kbfsTeW+5+c50muIeN3+fLlI/a65BzRmXOks8Jgzpw5Vv369R0xOZGVP0Ae1ZQEINQ2btyoTpQjfRJy7tw5K0OGDFaaNPo1gtOnT1tZs2a1ok2ePHms1atXO2LdunVTFwTkAsS4ceNU5/KRx/HatGlj/fHHH1apUqUsE8j7IknTs88+aw0aNCisr/XFF19Yhw4divhji0n1rzNnzlhZsmSxoo0kFvfff79Vrlw5a8mSJerEOzHymWrdurU1ZcoUa+jQoVZcXJwV7Rj7onvsk3Fv7dq1apx74okn4uNly5a1unTpYn333Xcq4fWRz/TgwYPVCV62bNmsaCZ/0zXXXGN9//33VsaMGVXsscceU581OSmWk4hQY+zz7uWXX7aOHz9ubdmyRfU30bVrV/X+yHuzfv16o8c+H4690XnsjeX+16xZM/W3Zc+eXSVEcvE83Mg5ojfnCMsdW/+kVvhOFuRObjgsWLBAJZh169bVfrZv3z7r4YcftgoXLqwO+Ndee631+OOPqyvbPrIDZUfKGyCDkswx+PLLLx3rWbp0qfqAf/zxx9YLL7xgFSlSRLU9ceKE9eCDD6oTn127dqmrEvIB69ixY/zJvNxpqFixopUpUyarYMGC6oTj2LFj2rZ+/fXXVr169dTvyxWz6tWrqzsVvv0q27R79261HfKvZMmS8b+7Z88ea/v27UHvQ9+6ZIBIqGHDhuq/n376qWWKV155Re33vn37hv21pO/Jvrvuuuu0n8n7IR/S/PnzqyRODigDBgzQBsgmTZqo91v60O23364NAnJiKu/3smXLrO7du6v5vEWLFo3vFzK3XA5M0v+lTz7//PPqZ+fPn1cn5qVLl1Z9v1ixYmreqsT9TZs2zbrlllvU7+fOnVut69tvv1U/k79PrrjJ6/v6XsLPufR7+ReIrE8OrLJNsj8kAb98+XKi7eXumvT3SByoQoGxL7rHPhmrhWxHQpIQCv8LLdL/JElftGiRFe3kb5PPrS+pFenSpbPy5cuX5AWk5GDs8z72LV++XD0F4EsqhIy1clK+YcMG6/fffzd67PPh2Budx95Y7n9y3i7HjUgi57CiNucIyx1bNwcOHFD/lYNsOPz444/q5D59+vSOuDzeKAOG7Dh5TFCuTkmiK3dW5KRakuGDBw+qx9Bk+amnnrLy5s1rffTRR+oDL+0SXsEXw4YNU78nSZMMUvL/xaVLl6zGjRtbderUUVeNfFft5EROEpOHHnpIrf/PP/9UVykkoVm5cmX8NksbuWsgJ4H9+/e3cuXKpdp888031n333acSov/++8/6+++/4x8tS3gXQR45k8Hv/6YRBCaJvZwMyWMC69atU9tcokQJNRAnJFdVJGmTbQ3HVf9Qkw+bXJ2cPHly2E7o/PtelSpVtPjmzZut2267Tb2/0vfkQywHoM8//zy+wJkcsKSNnMjLQU/avvPOO+rAJe+l/yR6SWolSZa70HLC7XPkyBGVHMuj+DIFQE7cJamQPiyPHsnryyM6v/zyi+o7v/32mxqYfeSqrDy+KJ+DF198UfXpNWvWqLs/8qiSJCdPPvmk6m++xDxhciDJuPjrr7+S3FdyR0zIgV4e85RkXF5LPmNyV0wOUP5Pegjpe3JQNhFjX/SMfdLn5E7ywIEDVV+Tk7ydO3eqz54k0r4Dqk+FChXUGCLb6n8ciDYyZowaNUr9bfJ0lJwAy4UBGdtnzZoVltdk7PM+9sm5giQt/nznCTIWXn/99UaPfRx7o/fYmxr6XySRc1jRm3PYEfLwww/badOmtX/77bewrL9o0aJqUry/Bx54QBUAWbt2rfazK1euqP/KpHPZFcuXL4//2cmTJ1Xxq5IlS9qXL192TFqXIkEyGTqhzp07q58999xzjrisU+IyATuhb775xhGXgibZs2e3a9SoYZ89e9Z1O8Vdd92limy5qVevXvwEey9mzpyp2vv+VatWzd68ebNr20aNGtnly5e3TdC6dWv71ltvjV92Kx4VKhcvXrTj4uLsPn36aD+rW7euek/9i9IkfD9btGhhZ8iQwd61a1d8bP/+/er35Pf9iyfUqVNHK1Dle98nTpzoiE+dOlX1/YT9Wkg7ab9y5cr4ghLSrmXLlvF93W1bZeK/vJYb6ZOJ9cuEmjVrpl5bitt07NjRnjNnjj1w4EA7Xbp06j1L+Ho+sn8ef/xx21SMfdE19n3xxRf2Nddc4xj7GjdurMZ8N1L8sEmTJna0O3XqlCraKOOR7++SAjALFiwIy+sx9l3d2HfPPfeooj0nTpxwxKXgjbxXY8aMMX7s49gbvcfe1ND/xOjRowMWNgsFcg47anOOiHzdj1w1fv/999Wcx4RXhEJJ7lj5X42Sq2ZyZeyee+5RV+r9+eYNfPXVV+qurtxp9ZGrY3KlTa6Cbd261fF7cjU8sTuB8ohzQlKURK4+yGMdhw8fjv8nV8PkNWSeoZBH3WTiu5S6lkf23LYzEHlU2uuVE9GgQQP1urKN8ry73D1JeBcwIdm3st3RTvbn3Llzk11kxqujR4+qfe7f92TOrRRkkLtQxYsXd30/5RFcedyoRYsWjnkE8lik3KWSq72+Ryd9ZE6MW4EquQMqd8USkvdVrhTLUwoJ+97//vc/9XNf35PPiHxW5C6w/1xxr31PPieBrhgLKVYh5O6YPH7VqlUrdZVanoKQK6CLFy82tu+5YeyLvrFPnniQOxDy1IT0fblbIo/p+X9+TOt/MgaUKVNGzY2bOXOm+nzJcU+e4PCf2hAKjH1XN/bJuYE8OdauXTv1NILcuevZs6e6cyHcCmua0vcEx97oPvbGev+LNHIOK2pzjrA/iiwnDDK/VR7RDff3y/rvYEkuJDGQR5STIvMI3L43yVddTX6ecB0yR9eNzGfyzXv0kXkL8gidzIl08++//6r/+uZIBNrWUJLHWXyPtMjJ0EsvvaROQmWbE07k9u3baC8gII+Cy+OOUpxIEqdI8u97Mmc70Psp/VMef0845yVh35MDnlQylMczA/U9me/teyTeR95HmdMuJ/KB+p4cVOWxy3DzXRCSSrQJSSIvj6BKcuv/OKgJfc8NY1/0jX3yuZSDqxRFkYsqQiowyjQBqZMg83zlkX4T+1+PHj1UAivz5XwnyTK/X8aPp59+Wj3eGA6Mfd5Iv5Iq3HIBxzd1RR7Bk/MieRTerTiZKX2PY2/0H3tjuf+lFHKO6Mw5wprYSpVimWcgJywyV1USv3CRebFuBUnCIbG7tXLF3P+qmyQncmIn1aDdJDbwpQTpaDKHQyZsy7zghGTfhmt+dKjIyeqOHTvUHFX/K5hyR0hi8l6EsmKhzNOTD19K9z23uPS9G264QVWccyPFLCJNCri5Fe/xJT9u+1GuMkd73/PH2BedY5/M5ZWqrXfffbcjLscpIXN6/BNb6ZPhetIoVGTukjwVJSeoCY9BckVc/h6Z1yxt/C9+JQdjX3AXH+TJAKm/IO9F5cqV1fsm5G67qWMfx97oP/bGcv9LCeQc0ZtzhC3TlCtR8v21cmIjj/qG+6sS5JEPKUzif+IkRXmkCmtSZPKyJET+fNW+5OfBkgnQUjCndu3aSRYy8lXUlW31n0idUDivnvkeRZG7LP5k38r3TkV74Qr5XkXZ124HXvk3f/589ehvqMjFGnnv/Pue79HipPqe9E9JshPre3KCmpwDoGyXJFhSXCKpfiPt5EAsj9zLgS6cfU8eQ3333XdVAbeEfN9h65/sSDs5IY/kd9MlF2Nf9I59UihQrgT7V+KWccN35ykhWZanJnyJbzQ/Fifb6lZhXP42+XwnVX08GIx9wZHiZfJ9oj7yGZHPh/9xy6Sxj2Nv9B97Y7n/pQRyjujNOdKEqwqoVHOTE/OFCxdG5Mq8fFDlxChhKXV5fUlipAqtbx6B22ME8vU8P/30k/qSZx957nvSpEnqEbXkPCYij4LJCYXMIfQnJyK+Mteyv6RcuXx5sdxRcNtO36Dk1gmupvS2PLfu9lz8e++9p/7rPx9ZXk9O1qVqXzSTisCSuPr/873H8v/dHjkPRd/z71/S56Vkv1RmlvclId++l7my8r7L1aqEd5jl5FvmZsqcb7kwk5y+JwcnSSTdBhTf3Ab5jMhnRea6ykHWbVt9fc+/LPvVfuWAPPYpTzZ88MEHjtfy9T15LCUh33frRXvf82Hsi+6xT+5KyDr9qwTLnFThX/1TTjhlm6K9/8kFZKkkLWNcwq+xkzntcvyTk7BwVIhn7PM+9rmRqRfz5s1T07WkFoepYx/H3ug/9sZy/0sJ5BxW9OYcdhjcdNNNquLVM888o6rDJfz37bffhuMl7XXr1qnXXLhwoSP+999/24UKFVLVIaX68TvvvGMPGTJEVZk7duyYanPgwAG7YMGCds6cOVWF1vHjx9uVK1dW1SXnzZsXvy5fVeTZs2drry9VkbNmzeq6bY899pj6PamsKet+44037KefftouXLiwY13vvfeealepUiX7pZdest9++227W7duqrKzzyuvvKLa9OrVy54xY4b92WefXXWFMtmGsmXL2s8++6zaH1IN74477lC/K5Xz/EnlWvnZzp07bROFsypywv2zY8cOR3zTpk12tmzZVAXg/v3725MmTbKff/559fnw2bJli+o3RYoUsUeMGGGPGjVKVd3OmDGjvXr1aq0qslt1b3nfpT/7kyqLTZs2Vf24ffv29uuvv25PmDBB9ak8efI41iX9XtYvlYmlP0hb6XcJq3x3795drWvYsGGqut3ixYuvujKjePHFF9VrSZ9788037UcffVStt0OHDlrbHj162MWLF3etlhyNGPuie+w7fPiwOh5Itc+nnnpKjX+yjVKxXz5D58+fd7SXz4IcO/wriUaj4cOHq31w8803q30t2y5VJSU2bdq0sLwmY5/3se+vv/6yb7nlFvU+SX+Xfpw5c2b1frn1L9PGPjcce6Pn2BvL/U8q68u+kX933nmn2p/yTRWyLPszHMg57KjNOcKS2CYs5+z/L7GS5aFw4403qq/W8CdftyIDRf78+VXCIImDJDoJT2Lk61akVL2UQ8+UKZMaAORrIRIKNrEVktRUrVpVDSTy1RY33HCDSvzlq10SkpM1GeCkXY4cOdR2yECW8Csd7rvvPrWdsi0JBzSvnUwG1TZt2qhBS/aHbHeVKlXscePGqa9w8NeuXTv1NTOmCvfBVfpRvnz51CDqTxJXKeXv61fy4ZYDWUIbNmxQXzciSbCcRDdo0MD+8ccfHW2CSWzFhQsXVLIsP5f3Onfu3KofDh061P7vv/8cbSdPnqwOcr52st5FixbF/1wuAMlXrkj/9f8sX01iKwdKOdjI16ikT5/eLlasmP3CCy+obfVPzOVrWeRnpmDsi+6xz3exs0uXLurr3CTBlT7WtWtX+9ChQ1pb+QqiTp062aaQr1CS/Sb7SPajbL+cJIQLY5/3se/o0aN28+bN4y+sSP+TEz23pMLEsc8Nx97oOfbGcv+Tr/ZJ7Njr9bwkGOQc9aIy54jY99hGwpQpU9QH33cnFsn3zz//qIQsXN+FGCvkLqQcKPy/YxbBmz9/vjo590+AoGPsC72NGzequyTyXySOsS/0GPu8o//R/1IKx93ozDki8j22kdKxY0f1naFvvvlmSm9KzJDvg5XqfjI3Eonr1auXms/28ccfs5tCZNSoUaqKo3yvL5LG2Bd6L7/8sqramFRRFzD2hQNjn3cce+l/KYXjbnTmHHGS3YZ0qwAAAAAAiKCYumMLAAAAAEh9SGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDR0nltGBcXF94tgXEiVVCbvoeU6nv0P7hh7ENKYexDSmLsQ7T3Pe7YAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo6WzDJI/f34tVrx48YC/17VrV0/rf/TRR7WYbduO5bi4OK3Ntm3btNjy5cu12LvvvutYXr9+vaftAgAAAGCeli1barHevXtrsccff9yxvGXLlrBuVyziji0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADBanO1fHSmxhi5FkyJdKOqrr77SYlWqVAmq4JPbn+2lXXLWdejQoSQniYv58+dbpvDYdZIt0n0P0S9SfU/Q/5BS/S/W+t7999+vxQYNGqTFrrvuOsfyiBEjtDYvv/yyFjt9+rQV6xj7kBr6X6yNfV5zmjx58jiWq1atqrU5fvy4lRrZHvsed2wBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRjC8e5TaxOtiCT2fOnNFi27dvD7it+fLl02IlSpQI+Jo7duzQ2lSrVs3TdkUDiggg1vteLBaxcNOrV6+AsWLFimltVq1apcUmTJigxWbNmmXFEsa+4Bw8eNDT8dPLZ3Du3LlarFOnTlrs/PnzVixh7Au/nDlzOpaffvpprU2bNm0CFj1zK3zmVgjNJIx94TV79mzHcv369bU23bt3D/h7sYjiUQAAAACAVIFHkQEAAAAARiOxBQAAAAAYLWrn2Lpxm4vjNqembNmyAdc1f/58LbZnz56QzbFdtmxZwO1y26fVq1fXYhs2bLCiEXMtEOt9L1rGvlBym+9as2ZNLTZ+/HjH8r59+7Q2NWrU8DT3rHbt2o7lvXv3WiZj7Ats2LBhWqx///5abM6cOVps0KBBjuU8efJobZYuXarFXn755YCxc+fOWSZj7AutChUqaLGFCxc6losWLRqy1+vXr58WGzNmjGUKxr7wypEjh2P5zTff1No0bNhQi91+++1abOvWrVYsYY4tAAAAACBV4FFkAAAAAIDRSGwBAAAAAEYjsQUAAAAAGM2o4lEmcZu07aV4VLVq1bQYxaPoe3CigIo3bdu29VSoxL+4UywUeAonCqjoihUr5lhet26d1iZnzpxarHHjxp6KL/qbOnWqFrvvvvu0WLt27QIWqzIJY1/wcuXKpcW+/vrrgMX0Ro8eHbDAmWjRooUWe/vttx3L06ZN09o8+eSTlikY+yLLrXDe2rVrAxY8E927d7diCcWjAAAAAACpAo8iAwAAAACMRmILAAAAADAaiS0AAAAAwGgUjwqBAQMGaLEXX3wx4MRnt+JR1atX12IUj4rO4lH169fXYoMHDw7YpkGDBlps6dKlId662EYBFW9mzZqlxdyK57i1Q8r3P5OKNlasWNGxvHnzZq3NkiVLtFjDhg2Der0qVaposeXLlwcsDtS6dWvLZIx9wfvggw+02IMPPhiwmJN/AShx+fJlLVaoUCEt1qhRI8fyyy+/rLUpXLiwZQrGvpTnf54p+vfvr8VKlCjhWD548KBlMopHAQAAAABSBR5FBgAAAAAYjcQWAAAAAGA0ElsAAAAAgNHSpfQGRLv8+fM7lidOnKi1adGiRVBFP44cOaLFHn30US126NAhLbZgwQLH8vr16wO+HoLnVgTKrRAKEE2KFi2qxfbs2RPW12zbtq0Wq1GjhmO5T58+Yd0GRF737t0DHgNHjBgRstdzK6p4+vRpLXbvvfc6litUqKC12bp1a8i2C9Hh8ccf12KdO3fWYh9++KEW8y8W5VYoqk2bNlrszTffDNj/Pv/88yS2GnDq27evtkt69+6txa5cueJpPEwNuGMLAAAAADAaiS0AAAAAwGgktgAAAAAAozHHNoEBAwZoO+iRRx5xLBcvXjzoLw32b5c3b96Ar5fYXCX/L2N2+8LmUM5nSk2GDBmixdz2rxdLly71FAPCoVatWlps9erVYd3ZPXv21GLt2rUL62si5TVr1izgcXHLli1h3YadO3dqMbfjLGJfhw4dtNjevXu12LPPPqvF3ObU+itdunTAmixuY9+3334bcN1IvfzP7YcPH+6pf3bt2lWLnTp1ykqNuGMLAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMlmqLR02dOlWLdezYUYv5F8BwK+Tkxku7UK5r2LBhWuzrr7/29KX2qZ1/sahgC0WJoUOHJrnuaFa/fn1PMS8ompUyatasGdHXGzt2rBabPXu2p6ItiC3RUKSpSJEiKb0JSCHdunVzLNeuXVtr43Y8/vfff4N6vSpVqnhqly6d8zR77ty5Qb0eUgf/c3m38/9Ro0ZpsY8++iis22US7tgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaCS2AAAAAACjpdriUeXKlQtYKCqxmJc2hw8f1mLz5s1zLK9YsUJrs23bNi32/PPPa7EWLVoE3K7y5ctrMYpH6YItFtWgQQNPRZOigX8RqCVLlkR8n7rtG7d9iPAqVqxYUMWd3H6vVq1aWqxdu3bJ2DqY6v3333csd+/eXWtTuHBhLXbo0KGA686aNasW69ChgxYrUKBAwHU1btxYi23dujXg7yG6+Rf/PHPmjNZm+vTpQa3brV/Vq1fP0++eO3cuqNdE6nT69GnH8p49e7Q2r732WgS3yDzcsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEZLtcWj7r//fi02YMCAgEWm3n333YBFoRIrHhWsl19+WYu1bNnSsRwXFxey14M30VooykvBs2goYIXQW716dcA2bgWfvBSPGjt2rBabPXt2UOtC7Pnkk08CFo/69NNPtVjDhg21WKZMmQIWJHMrquhFqVKlgvo9RI/s2bNrsUqVKjmWt2zZorX5448/Qna+mD9//qDWBSSlZ8+ejuX33ntPa7N+/Xotdt9992mxFS4FalMD7tgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaKl2ju327ds9zaOIBv7zfL3Oo9y2bVuYtii258p6nQu6ZMkSLdagQQMrkoYMGRLR14NZxo0bF3AOj5g1a5YWa9u2rWO5Zs2aWps+ffokexsRG/zncz377LNam5EjR2qx33//PeDx7cKFC1qbrVu3arG5c+dqsUGDBiWx1TCR23uaK1cux/JXX30V9PpLly7tWH7iiSeCXhdwNT744APH8qpVqwK2SWzsa968+VXX4IgF3LEFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGS7XFo6JVy5YttdiUKVMCFtc4c+aM1sYtBt3QoUODKh7l1s7/ffFfd6jVq1fPMqEgV0oU1oJ7cac9e/Z4Kh5VtGhRx/L48eO1Nnv37mU3w9WYMWO02MaNG7VY165dtdivv/7qWP7666+1NuvWrdNiGTJk0GJ169blHYox2bNnD9imTJkyWqxw4cJarHbt2lps1KhRjuUcOXJobQ4fPqzF8uXLF3C7gOQWum3atKkW++GHH7RY3759HcsdO3bU2pw/fz7m3hDu2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKPF2f7VbhJrGBdnpbRy5cppsVatWmmxsmXLBpzk7zYh223ytVu7SBeKypIlixbzf9vatGmjtZk/f74VTh67TrJFuu+5FYUaPHiwp3apgX9BrCFDhkR8GyLV96Jl7Aultm3barFPPvkkrK85e/bsgNtgklgd+2KRf2G0EiVKaG1q1KhhmYKxz7IeffRRbb+8/fbbjuU0afT7Nv/9958Wy5kzpxa7ePGiY7lDhw6eilO99NJLWmzu3LmO5datW1smY+yLTm650Gy/465bn925c6cVa32PO7YAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBo6awo4V8YasSIEVqbFi1aeCqu4T/B2EubxNr5b8fAgQMtL7JmzarFnnvuOcfygAEDgt4u/4JY4S4UlZosXbrUU8yteJSXglL16tUL6veCLe7kVUoUgULKF9PxWtBkzpw5ntbvVoind+/eAYtHuW0XkFz+x9Rq1aqxUw03adIkLZYrVy7H8hNPPOFpXcuXLw9YxNO/AJTo37+/p/X7FzMFksttDHvzzTe12OXLlwMWzjOpeJRX3LEFAAAAABiNxBYAAAAAYDQSWwAAAACA0VJkjm3dunW12EcffeRYLl68eNBfzuulndd1+c+jcHu23W1+69NPPx1wroXbNrjF/OfTiiZNmiSx1YimubiAifzn1HqdA+vWbs2aNY7lnj17evo9APDilVdeSXI5pbidvwGJyZw5s2O5UaNGWpvBgwdrsQIFCmix6dOnO5YXL16cKnY8d2wBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRUqR41MSJE7WYf7GoNGn0nPvKlStazEu7uLg4T9vlpV3jxo21mNvkbrd1+ReGcmvjVmigXr16Wmz79u0BtxUAvGjTpo0Wa9u2bch2nn9hqDFjxmhtatasqcVWr14dsm1AeGXKlMmxnDt3bq3NoUOHtNilS5fCul3+x9njx4+H9fWAhPLkycMOiXFZsmRxLJcuXVprU6xYMS1WoUIFLda6dWvHcvXq1T1tw+nTp7XY5MmTrdSIO7YAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoKVI8qmzZsgELK7kVivJvczXtgmkT7nUtX75ca9O7d28tRqEoAKHiVhRq3LhxEd3BbkWh/AsIJtYO0alPnz6O5TvuuENrc/fdd2uxU6dOha2AlShatKhjecaMGSF7PSCQEiVKOJavueYarc0///zDjjSYf8Gnjz76yNPvueUOXorYbtq0SYv169dPiy1ZssRKjbhjCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjJYuGiZaiylTpjiWs2XL5qlQVJo0aQK28zIZ22s7r+s6c+aMFnvppZccyyNHjvS0LgAI5/g7Z86ciO7gvXv3RvT1EH67d+92LNesWVNrM2vWLC02f/58Lfbpp586lk+ePKm1OXv2rBa7/fbbtViNGjUcy2nTptXaAFfLrTCP2zlqzpw5HcvlypXT2lA8KrZ4LSjr5s8//3QsT5w4UWszZswYT30vteKOLQAAAADAaCS2AAAAAACjkdgCAAAAAIyWInNs3ebUPPDAA47l4cOHa23Kli3r6blyL8+3e30GfuvWrY7lFStWaG22b9+uxRYuXOipHQCkNrNnz9ZivXv39jQnE9Fp2rRpjuWMGTNqbQYMGKDF3OaQ+cc2bdqktdm5c6cWq1ChQsDtLFy4cMA2QCCrV6/WYq+++qoW69Wrl2P53nvv9TRfF+Y6duyYFnPLaX744QcttmHDhpDN102tuGMLAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMFmd7nJkcFxcX/q2BUSI1qZ2+h5Tqe7HY/9wKMu3du1eL9enTx4qkPXv2aLHixYtb0YixLzj58+fXYpUrV9ZizZo1cyw/8cQTQb8H+/fvdyw3btw4YJHIaMbYF72eeeYZLTZq1KiARUTr16+vxQ4ePGhFI8Y+RHvf444tAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGsWjEDSKCCClUEAleL169dJi48aNC1i4ya3AVLjf02gt3MXYh1jve9H8+YtWboXQFi9e7FiePXu21mbgwIFa7NChQ1Y0YuxDSqF4FAAAAAAgVeBRZAAAAACA0UhsAQAAAABGI7EFAAAAABiN4lEIGkUEkFIooIKUxNiHWO97guJRSKn+R9+DP4pHAQAAAABSBR5FBgAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARouzbdtO6Y0AAAAAACBY3LEFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGi6nE9tSpU1aBAgWs6dOnp/SmxIyaNWtazzzzTEpvRtS7cuWKValSJWvEiBEpvSkxo3379lbbtm1TejOMwNgXeox99L2UQt/zjrGP/pdSOO+L0vM+OwgffPCB7fvVJUuWqP//559/xv+8RIkSKub2r3Tp0vHt5HckJusQ8v9l3cEaPny4XapUKfvSpUt2rHrzzTeTtY98fv/9d7tVq1Z2rly57MyZM9u1a9e2v//+e63dvHnz7CxZstj//POPHQ0C9T3Z3kaNGtnXXHONnSFDBrtIkSLq7/zll1+0dSXsb9JnBw8eHPR2TZs2zc6RI4d9/PhxO1ZNnz7dHj9+fLLXI/uoX79+aizIlCmTXbx4cbtLly727t27He02bNhgp0mTxt60aZMdLQL1P5+PP/7Yrlmzpvrs5MyZ065Vq5a9ePHi+J8z9qXc2Ld//367a9eudsmSJVX/k2NGr1697MOHDxs99nHcDZ/UftwV9D+z+5/vmOP2b+bMmVHd/zjvM/+878CBA/aDDz5o58+fXx13b775ZnvWrFlau1Cc96WzwmDChAnqKlpCu3fvtl544QWrUaNG4XhJ6+LFi9arr75q9erVy0qbNq0Vq9566y0rX7581oMPPhj0Ovbu3WvVqlVL7ad+/fpZWbNmtT744AP13ixevNiqW7dufNvmzZtbOXLkUK/74osvWtHul19+sXLnzm09/fTTaj8dOHDAmjx5snXLLbdYq1atsm666aawvO7o0aPVlaacOXNasWrGjBnWli1brJ49eybrCucdd9xhbd261erevbtVpkwZa+fOnap/LVy40Nq2bZuVPXt21fbmm2+2qlWrZo0dO9aaMmWKZYohQ4aoz0rr1q3V51TGJtlv+/btC8vrMfZ5J8clGftOnz6t+l+xYsWsn3/+2XrjjTesJUuWWOvXr7fSpElj5NjHcTd8OO7S/0zvfz4dOnSwmjZt6ojJmJiQaWMf533Rfd534sQJq06dOtbBgwfVuXmhQoWsWbNmqTuz8oTtfffdF982FOd9YUlsW7RoocWGDx+u/tuxY8dwvKT1xRdfWIcOHYr4o4tygiSJoZszZ85YWbJksaLNyy+/bB0/flx11rJly6pY165drXLlyqkLA3Jy5yMneXKCLh1s6NChVlxcnBXNBg0apMUeeeQRq2jRotbbb79tTZw4MeSvuXHjRnVyLB/ESDp37pyVIUOG+BNxr/0yJa1evdpau3atSiSeeOKJ+Lj0wy5duljfffed1bJly/i4fJ4HDx6sDrDZsmWzop38fXIiIH1BPkuRwNjn3WeffaYusso+u+uuu+LjefLkUe+bfI7lwGri2Mdx9/9w3KX/hZuJ530+VapUsTp16pRkG9PGPs77ovu875133lE3MOTG2f/+9z8Ve/zxx9WUiz59+qi+JueyoTrvSxPJrP/aa6+1br311rCsf8GCBVbJkiWt6667TvvZ9u3b1Y7Knz+/lTlzZnUSPWDAAC05adKkibpKJTvy9ttvVyepCX344YfqA75s2TJ1tV/m80rCJOrXr6/mWEpSKHc8ZWB7/vnn1c/Onz+v3qTSpUtbGTNmVHcJZN6qxP1NmzZN3V2U35c7j7Kub7/9Vv1M/r5ff/1Vvb5sh/yT1/XZtWuX+hfI8uXL1cmbL6kV8nrNmjWzNmzYYP3++++O9nKHTU4GN23aZJlI3if5+ySZD1ffkw9lwjvdPnKX7uGHH7YKFy6s3nv5DMgH+sKFC/Ft/vjjD6tNmzbq5Fq2Uz7sX375pWM9S5cuVe/3xx9/rJ58KFKkiGorV8LkKq70WXnv5Uqs3PH0XUCSO6RyJ6dixYpWpkyZrIIFC1qPPfaYdezYMW1bv/76a6tevXrq9+VzUL16dfW5FdLPZJukH/j6nvRHnz179qjPWSCyvUK2I6FrrrlG/Vc+n/59TwbrRYsWWSaQfS1XI+WqpDzt7v/kSjgw9nkf+4LpfyaPfRx3Oe7S/zjv8yfH1ITnIG5MH/s475sQNed9knNI/uVLan0XTyQvk6cqJacJ5XlfWO7Y+pOkUR4x9E8mQ+nHH39UV6L8bd682brtttus9OnTW48++qh6U+QE6PPPP48v9CPJorSRN1USTmkrVxjkTZUdXqNGDcc6JamVN0muEsnO9zly5IhKjuWRVLkiJp1JEgtJGFesWKFev3z58uqxifHjx1u//fabOin1kStj8hijJP9y90CSpTVr1ljff/+9ekxYTpqffPJJlcT49mXCEzRJxsVff/2V5L6ShFqSZn++q4ySnF9//fXx8apVq6r/rly5Mv5uRrSTJFYe0ZQPjew3OaH17Z9w9D25qCH9JqH9+/erixSyLfLeyx1xSXTnzJmjrurK+yuPZsj7LctPPfWUlTdvXuujjz5SfUbaJbx7KYYNG6Z+r2/fvup99F3lunTpktW4cWP1uMeYMWPi30sZzOSCzEMPPaTW/+eff6q7pfKZlPfTt83SRu6YykDYv39/K1euXKrNN998ox4Tkf7233//WX///bfquyLhlbQHHnhAfVb+b+py4uQRE7miOHDgQJXIy8UVuZInnzsZUBs2bOhoX6FCBZVsyLb674toJFck5f187bXX1FMqMiZIoiv7r0ePHmF5TcY+72OfXHySA6pceJC76nJhUo4RciyQO57yGU3IxLHPh+Mux136H+d9/uQ8U6agSZIi45uMfW5TBE0c+zjvi87zPjlX9b9o7J9zSDIbsvM+OwL69OmjJqhv3bo1LOu/ePGiHRcXp17HX926de3s2bNrhWmuXLkS//9btGihCg3t2rXLUWBEfk9+338Ce506dbQCVfXq1VM/mzhxoiM+depUNRF6+fLljri0k/YrV66MLyoh7Vq2bGlfvnw50W2tWLGiei03UjxE/gVyzz33qOIVJ06ccMSlwI1s05gxY7Tfkf3z+OOP26YoW7ZsfGGEbNmy2S+88IK2X0OlaNGiqiCIvwceeEC9p2vXrtV+5ntPe/bsqbYxYf84efKkfe2116riNr5t9hWLkUI3Z86ccayrc+fO6mfPPfecIy7rlLhM/k/om2++ccSlmJP09Ro1athnz5513U5x1113Jdq/fP3fiy+++EIV90pYvKJx48bq73ZTpkwZu0mTJna0O3r0qPpb8ubNq/rc6NGj7U8++cS+8847XceGUGDsu7qxT7z33ntq/EvY/+QzJPvSjWljnw/HXY679D/O+3zkHFgKa7799tv2Z599Zk+YMEEVbpRzFDkmx8LYx3lfdJ73Pfnkk6qf/fXXX454+/bt1e/36NEjpOd9YU9s5cRcKtNKBaxwOXjwoNo5UhU5oX///VfFn3766UR/VxJUqf7Wtm1b7WePPfaYejP+++8/R2L70Ucfub7BGTNmtM+fP++IN2vWTCWjhw4dcvz77bffHNssJ8GyvHHjxiT/1qQSW6+++uor9VrSaaQC2Y4dO9Q+Sp8+vYoPGzZM+52CBQvabdq0sU3x448/qg/yW2+9ZVevXl2d5F24cCEsryXVLR955BGt30uV5ObNmyf5u/LhveWWW7T4yJEj1Xvhq+bsS2yHDh2qtfUltv4Xb5566ilVkVc+B/79TxIv3zbPnj1b/f78+fOT3NakBrirsWbNGrtp06b2iBEj7AULFthDhgxRn8HWrVu7tpeBV97DaLdnz574REmqIifsCxUqVFAXQEKNse/qff311+oET07spM/37t3bTpcuneuFURPHPsFxl+Mu/Y/zvkCOHDmixjdJCGNh7OO8LzrP+37++WeVX8i5rtzM27lzp/3SSy+pnEm24eGHHw7peV/YH0WW29Ty+GUkCqn43w6XuYtCHhNNjBScksdAE8439ZHHhuVRYqkiLLfqfWSepBuZ95hwArSQ+aryGLY8uuzm33//Vf+Vx6PlETm5BR9u8rj066+/bj333HPxj2/L/F95JEUeCXWbrC37NtoLCCRW5U8eDZf3UshjupHoe9Kv5PHnpPqekLkL/o+6C9/2ys8TriOxvpcuXbr4+d4J+548RiJzTQL1PRFoW0NBPpMNGjRQRSlatWoVX4FRpgjIXGGZ7yH908S+53vURh7zkWIIPvK5bteunZpnL3NSihcvHvLXZuzzRh5tuvvuu1X9BHksXsgjyDINRR7Rk8ey/MdgU/pfQhx3Oe7S/zjvC0SmA8njqlJQVB439T+HMG3s47wvOs/7brzxRjVvt1u3blbt2rVVTKZoyTRBqTkT6pwj7ImtlHKWEzspMR7OD6fsALeJ0eHg9qx4YnFJjG+44QZr3Lhxrr8jhaRSgsz3kwFN5pdJMl65cmXr/fffVz+Tr2Bxm7sg5eZNJPOJZdK69MVwJLYyLzal+54UpvKvjix9TwY3+bvdJHaxJZxkTodUc5bkIiGZU+xLPPwTW9m3Ced8RysZh6RQg8xT8f/KMd9BRv6WUCa2jH1XR2onSF0CX1KbsP9JfQOZr+yf2Jo49nHc5bhL/wutWDjvS2pbjh49qiW2Jo59Ppz3TY+a8z4hF/vlOCvfPHD58mV1U02KoiaWcyTnvC+sia1MGJ47d64qwiRVYcNF7lZJNWSZIJ1QqVKl1H/la20SI2+yTGDesWOH9jOp9iXJQnIGIdkueSOluElSVx+knQyG8v2ekmQmJpRXz6SIT8IrXPJVKzJI+66o+Mgdd6mg57uLaKKzZ8+qq1jhIAVn/Pue9Cu5C5RU3xMlSpRItO/5fh4s6VPynsr7mdhB2ddOyLbKnftw9j0pliVX4mRgS0gKffmKYCUky/LEhC/xjWYyVshnV77OSD4vCZ/ekEJi4TioMPZdff/z73tJ9T8Txz6Ouxx36X+c93nle7LR/9hk4tjnj/O+zFFx3ucj50RSJNRHzk+Ff9HQ5J73hfXrfr766it1xSdc312bkCRo69atc8TkgypVMCdPnqweAXR7dE/urEhFuE8//dRRUVNOgOTWuVSZlQQlWFLOWgaId9991/VD56uqLI/DyYmxVEOWBNdtW33JaGJfW+P1Ky/cyJ2KefPmqa+myZkzp+Nnvu+1DddXNYWS7zGLhOR9lWq1/ndpQtn3ZHBI+PVN8l7KeyrVt/37ZcL3VL6e56effrJWrVoV/zPpE5MmTVKP5ybn0XTpe3ISL5WU/cnA4etH0v+l1PvIkSPV3VS37fT1vcQuDngt+y5X5mSd8uXcCc2cOVP917/6olzokW0yoe8JeeRY9rlUtvaR7Zc7aPJehuMCH2Of97FP+p+M7b4rxYH6n0ljnw/HXY679D/O+/zJ9Ch/cm4q58fyqKjvK89MHPs474vu8z438sj0xIkT1dN7/ndsk33eZ4eRVIqVycFSfSvc5syZoyYhSyGkhDZt2qQmTEul0v79+9uTJk2yn3/+efumm26Kb7NlyxY7a9asqsiVFLQZNWqUqj4r27569er4dr7iUW5VbqWgkxR2civiIYVypGqzVAB7/fXXVdGSbt262Xny5HGsa+DAgWr9t956q6pMLG2lsm7Carfdu3dX65ICTzNnzrQXL1581ZVBpTKZTOKWwlVSIbRXr16qAJIU+PKvlCykYplUz0tYKS1aFShQwO7QoYN6D+W97tevn9rPmTJliq9AHWrr1q1T79vChQsd8b///tsuVKiQKowk1Y/feecdVShJ+smxY8dUmwMHDqgCDTLZX97/8ePH25UrV1bv8bx58+LX5SseJRP+3YpHSf91IwXQfIXCZN1vvPGGKhRWuHBhx7qkH0i7SpUqqUn9UjlR+qj0P59XXnlFtZH+MmPGDFVZ8Wqr4x0+fFjtE6m2KEUOZJ/INqZNm1btF//ia/I5kP3n1i+jkVSslr9DCiX07dvXfu2111QBBPn7pGhbODD2eR/7tm/frj4rckyQ44FUqpbxQvruHXfcYfTY58Nxl+Mu/Y/zPn8PPvigfdttt6lzEN95sJwXy7FYzi9MHvs474vu8z5Rvnx5e9CgQeo1BwwYoM7L5Zgt58n+knveF7bEVioJSzJx77332pEgJ8T58uVzregriat8jY58xYNsk1SAkyQiIakOLF85Iic8skMbNGigKqwlFExiK6QaryRa8nNJlnPnzm1XrVpVVbj1VVz2mTx5skowfe1kvYsWLYr/uSRCUqVMynTLtiSskOz15E6+lkSq9foSDPlqmWeffda1E0liLl/NIl+XY4LBgwfb1apVU/tOKp3KB1kuKGzevDmsr3vjjTe6VnaTSsUySOTPn1+9p3LB5IknnnAkcPI1U1IR2Nc/5aKDf/n9YBNbIQcx6W9y8UL6zQ033GA/88wz6iutEpIBSy6qSDup6CzbIRdPfE6dOmXfd9998V+VkrCvXc0AJwNZly5dVL+T/if9q2vXrqpqn1tlvE6dOtkmkUrF8p7IwC3vufwNUqE7XBj7ru7rfiS5lc9bsWLF1AUI+T25CHH69Gmjxz7Bcff/47hL/+O87/+TpES+vlLOReTcSM6XZf+sX79e6yimjX2c90X/eV/79u3VMVfO+eS8XBJoOVdyk9zzvoh8j22kvPjii+pk2f87ZhE8KQUuHd7/wwCnKVOmqMHDdycWySdffSV3rgN9BRYY+8KBsc8bjrv0vZRE/6P/pRTO+6LzvC9O/seKEadOnVIFo8aPHx+Reb2pgczfu+2226xXXnklpTclqsm8aJmnItW/BwwYkNKbExPka5pkv/rPx4WOsS/0GPu8oe/R91IS/Y/+l1I474vO876YSmwBAAAAAKlPWKsiAwAAAAAQbiS2AAAAAACjkdgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaOm8NoyLiwvvlsA4kSqoTd9DSvU9+h/cMPYhpTD2ISUx9iHa+x53bAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGC0dCm9AYAJihUrpsXGjh2rxdq0aRPU+vfu3avFxo8fn+QyAAAAolvp0qW1WP/+/bXYQw89FHBdcXFxWsy27YC/d/LkSS324osvejq3NQl3bAEAAAAARiOxBQAAAAAYjcQWAAAAAGC0ONvLg9mJPNOdGuTIkcOxPGrUKK3NTTfdpMV2796txVasWOFYnjhxotbm8uXLlik8dp1ki4a+5/VvnT17tmN5zpw5WpsiRYposVq1agWcr+s2D7dt27ZabPXq1Vasi1Tfi5b+h+iSmsY+RJfUNva1atXK07zAnTt3OpaPHTumtZkxY4anc65//vnHsbx161YrnGrUqKHFKlSooMX8t2PNmjVWpDH2BTefdtGiRVqsePHiVjSa7XceK9q3b2+lNK99jzu2AAAAAACjkdgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaKm2eFSePHm02P/+9z8t9uyzzzqWq1WrFrJtyJIlixY7e/asZYrUVERgz549WqxYsWIR3Va3L83u3bu3p4IEboWnTJbaCqgguqSmsS8apE+fXouVLVtWi40ePVqLNW7c2LH85ptvam2efPJJyxSpbezLnDmzFvv8888Dnr+dPHlSa7Nu3Tot1qBBAy12+vRpx/Lx48dD9j7s2rVLixUsWFCLlStXTov9/fffjuUqVapobQ4fPmyFE2OfrmXLlo7ladOmaW0yZcpkmezo0aOO5fz580d8GygeBQAAAABIFXgUGQAAAABgNBJbAAAAAIDRSGwBAAAAAEZLZ8Ug/2IDO3fu9DSR262glL8LFy5ose7du2uxb775JuC6zp07F7ANokPfvn212CeffKLF2rZt61ieNWtWyLahT58+WmzNmjVabOXKlVqsdu3aMV1MCkDsqFixomN53LhxWps77rhDi+3YsUOLLV26NOAxHNHL7f0qWrRowN/7+uuvtdiSJUs8FY/KmDFjwHPDK1eueCpyliFDhoDnnnPmzNFiadKkCVg8yq0NIm/gwIEhKxR14MABx/KCBQu0NjNmzNBi27dv12KtW7d2LNesWVNr88ADD3jarhw5cjiWq1atqrVZv369FQ34VAAAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKPF2bZte2oYF2eZwn9C/dy5c7U2LVq0CGrdjz/+uBZbtGiRFrvvvvu02HvvvedY/ueffyyTeew6yRatfc+tMFSbNm0cy8WLF9fahLtwk5eCVf5FrkwTqb4Xzf3PTbVq1RzL5cqV09rkz59fiz333HMB27ntc7d9478NYsOGDZYJ3PaXW1GO1D72BStv3rxarFWrVlps2LBhjuWff/5Za9O7d28tdujQoYDFJP/66y/LZLE89qVNm1aLdevWTYuNHTs2YJEmr9z6VteuXR3L69at87Su6667TotNmTLFsVywYEGtTZUqVbTYiRMnrGiU2sc+t/HKv5hTunTe6vLu379fi7Vs2TKovudF7ty5tdiKFSs8HQf9DRo0SIuNGDHCioa+xx1bAAAAAIDRSGwBAAAAAEYjsQUAAAAAGM3bg+CG8f/i7KFDh2ptGjduHHAujpsjR45osU2bNmmxbNmyBZwTVKhQIa3N+fPnA24DooPbPNU9e/Y4lleuXOnp91avXh3W7YrkvCxEhv9cHDFnzpyg5sW6tfOfK+v2em7c5qRGA7d5Q/7zpdzmGmfPnj2s2xWr/OsNuM1dFFWrVg1Yy8KtTsbly5e1WI4cOUI29xKR98ADD3iaT+s2h/HSpUuO5S1btmhtXn31VS02e/ZsLXbmzBkrGLt27dJin376qWP55Zdf9vRZef/994PaBoSXW57gdU6tvx07dmixUM6p9Xfs2DEttmzZsqDm2EYz7tgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaCS2AAAAAACjxdkeq8pE65clB+v+++8P+EXaoeb/Rcj169f3VBAjWqX2L+p2U6xYMcfyJ598orWpVatWwMJiYvz48WF7r9y2IZQFrMItksWwoqH/3XnnnZ4KPPgX4tm6daun4k7z58+3opHb35g1a1bHcosWLbQ2L7zwQsCigmLDhg2O5V69enn6AnvGPitg4bLKlSt76sc7d+60QqVOnTparGjRogELFPXs2VOL/fbbb1Y0iqWxr0mTJo7ld955J+D7l5hnnnnGsTxmzBgrGlSoUCFgcaDDhw9rMbfzwz/++MNKaal97HMbY7744ougCg4uWbJEizVs2NAKl4IFC2qxbdu2abGcOXNqsYsXLzqWixQp4qm4bkr0Pe7YAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo6WzUqmpU6dqsVtuuUWL9ejRI2Sv6V+oxKRCUfBm7969juVbb71VazNr1iwtNm7cOC22atWqsBV3Mr14VGrz5Zdfeiqk8NprrzmW3333XU+FSsItf/78juWWLVsGbCOee+45LZYlS5aA+2HZsmWeCmRNnz49xfeNiZ588kkt9tNPPwUs0Hj27NmQbcM111yjxRo3bhyw+JBbAavatWsbUzwqlhw6dCjJPuR2TBULFizQYpMmTbKikX8BvzNnzngqkFW+fPmoLB6V2rkVE/QvAtWsWTNP63LLOT777DPH8tixYz0d37xIly6dp0JRbvyPs+EuFJUc3LEFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGi7PdKm+4NYyLs2Kd299YokQJx/KmTZuCnnzdtm1bx/Ls2bMtk3nsOsmWGvqeW0GpmjVrOpaLFy8esveqXbt2nrYhtfe9aOl/bsWWpkyZosWyZs3qWN62bZvWZt68eVrsvffe02K7d+8OWNypU6dOWqxFixZa7Lbbbgv4/rntZ7ftX7hwYcC/x63ARyil9rEvW7ZsWuzUqVNWNGrevLljOV++fJ76y44dO6xoFMtjn9vrpUmj338xufCmW4G6PHnyaLEuXbposQ8//NBKaal97HPTpk0bx/LHH38csnW7FRtzOy56kT59ei124403evrdCxcuOJYzZ85sRWvf444tAAAAAMBoJLYAAAAAAKOR2AIAAAAAjMYc2wD85zVu3rzZ0xzbS5cuBZz/5v/MummYaxFevXr1cizPmTPH05fX+8/NFatWrTJ2/kpqm2fmVbly5bTY3LlzHctly5b19PccOnQoYN9ym5voNu/b7b3xn8PoNkfo3Xff1WLbt2/3NOco0hj7wit79uyO5f/9739am0KFCmmxAwcOBGz366+/RnxOdigx9pnN6xxbtxosbrUxIo2xT+d/bJw+fbrWpmHDhpbJtvsdiytWrBjxbWCOLQAAAAAgVeBRZAAAAACA0UhsAQAAAABGI7EFAAAAABgtXUpvQLTLkSNHwEJRXpleLAqhMWvWLE8Fn1avXu1YrlWrltZm3LhxnmKIPW6FlfwLOqxdu1ZrU7VqVS2WP39+LVagQIGAhRtmzJihxe6///4kthqpRcaMGbXY+fPnPfUX/+JiR44c0dp8+umnWqxKlSpa7MMPP3QsN2jQIImtBoDkFQTr2LGj1ubee+/VYq+88krAwnkp4fLly1ps+PDhlim4YwsAAAAAMBqJLQAAAADAaCS2AAAAAACjkdgCAAAAAIxG8SggwubMmRPU7xUtWlSLrVq1KgRbhFgxdepUx3K5cuW0NocOHdJi8+bN02J169Z1LJctW1Zr06JFCy3WsmVLLTZ//vwkthqmcStwcueddzqWf/31V63N1q1btdi3336rxc6ePetYPnHihKftevDBB7XY7NmzHcubNm3ytC4gUuLi4rRYyZIltVi6dM5T9kuXLoV1uxCaYlJi0qRJWuznn3/WYrly5XIsd+rUSWtTpkwZLXbddddpsdy5c1vBmDhxohabOXOmZQru2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKNRPCqAa6+9NjLvBFKNWbNmeYp58eOPP2qxWrVqBfy9PXv2aLHixYsHtQ2IjkJRomPHjo7l3r17a20mTJjgaf358+d3LH/11Vdam6pVq2qxRo0aaTGKR8WWkydPBizS5NXBgweD+r2KFStqsRIlSmixcePGBbV+IFJs29Zimzdv1mIUi4ota9asCdhm4cKFngqLLV++PKjiUZcvXw5qu6IZd2wBAAAAAEYjsQUAAAAAGI3EFgAAAABgNObYBpAmTXC5/59//hnU7wFXw8t8WtGuXTvH8ieffKK1Yd6tWcqVKxdwrlbZsmUjPjcMiIQePXposaJFi2qxZcuW8YYgxTz00EOO5Rw5cnj6vQULFoRpi2CSYsWKabFevXppscKFCwdc16VLl7TYqFGjtNj06dMtk3HHFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3iUQHUrFkzqB27atWqoH4PuJoiAm7GjRunxWbNmhWwf7oVj2rbtm3AdSFl3H///Vps69atjuW6detqbebOnavFWrRoocXi4uICForasGGDFvv222+T2Grg6nXo0EGLdevWTYs1bdqU3YsUkzFjRi02dOhQx3K6dPpp98mTJz2NrYgtbgWf/Mc1/+Jjif2eF//++68WGzRokBVruGMLAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMFme7VQRxa+hXSCS1qFy5smN548aNnn5vypQpWqxz585WLPHYdZIttfY9fz/++KMWq1WrlhYrXry4Ftu7d29Q76fb77mtP1b7nmn9r2fPno7l5557TmuTP39+T/vzyJEjAYtVpdZCUYx94VWiRAnH8pw5c7Q2Z86c0WL16tWzYh1jX/RyK8I3b968gL83bNgwLTZ48GArGjH26QoWLOhYrlSpkqcCeHfddZcWK1CggBUqf/zxh2P5hRde0Np88sknlim89j3u2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOlS+kNiHaXLl0K6vf8C68AyeVWKMqNl0JRXhUrVixk60L4TZgwIcllwARVqlRxLOfNm1dr06ZNmwhuESIhffr0Wqxly5ZazK2Y2JUrV6xIctsut6Kh/j7++GMt9vLLL4dsu1K7LFmyaLHGjRtrsTvuuCNkr9m+fXvHcs6cOa1wunz5shabOHGiFnvttdccyzt37rRSA+7YAgAAAACMRmILAAAAADAaiS0AAAAAwGjMsQ2gefPmQe3YHTt2BPV7wNXMnQ12DqzX31u1ahVvCICwyZgxoxYbPHiwY3n9+vVam7/++ot3JcakSaPfaxk9erQW++6777TY0aNHI3oeOH36dC2WKVOmgOvq0aOHFjt79mwytg4J9enTR9shQ4YMMXYnnTlzRouNGzcu4JiZmnHHFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3iUQFky5YtqB1LYQuE2vjx4z0VEfjxxx+1WNGiRYMqHtWuXbur2kYAuBqFChXSYv3793csb9y4kZ2aCpw/f16L3X///Z7Or+bPn+9Yfvvtt4M+x3vmmWccyw0aNNDapE2bVovt27dPi3Xr1s2xfOzYMU/bheBcunQpKnfdxYsXtdiuXbu02IcffuhYnj17ttaG/CJp3LEFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGo3hUACVLlozMOwEEUTzKTZs2bbSYf7Eot6JTEyZM0GJ79+7lfQEQEnXq1NFiefPm1WKffvopexzKDz/8oO2Jrl27Biw4tnjxYq1N5syZQ1bUaubMmVqsb9++Wuzff/8N6jURnJEjR3pqN2jQIMdyhgwZgt7lW7ZscSzPmjVLazN58mQt9s8//wT9mkgcd2wBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDR4mzbtj01jIuzUqNu3bo5lt9++22tzYULF7RY9erVtdjmzZutWOKx6yRbau17SPm+J+h/SKn+Z3Lfcyu8OHbsWE9F8VasWBG27TIdY583N910k6cCiX/99VfAdW3fvl2LjRo1ykqNGPsQ7X2PO7YAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGnNsETTmWiClMM8MKYmxL7DatWtrsVmzZmmxIkWKhOhdSR0Y+5Aa+p/J9QUQHsyxBQAAAACkCjyKDAAAAAAwGoktAAAAAMBoJLYAAAAAAKNRPApBo4gAUgoFVJCSGPsQ631PUMAHKdX/6HvwR/EoAAAAAECqwKPIAAAAAACjkdgCAAAAAIxGYgsAAAAASB3FowAAAAAAiEbcsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEaLqcT2ypUrVqVKlawRI0ak9KbEjPbt21tt27ZN6c2IevS90KPveXfq1CmrQIEC1vTp08PwTqRONWvWtJ555pmU3oyox9gXeox93jH2hR5jnzeMfVE69tlB+OCDD2zfry5ZskT9/z///DP+53PnzrXbtm1rX3vttXbmzJntMmXK2L1797aPHTumrUt+V9YnSpQoYQ8ePNgO1rRp0+wcOXLYx48ft2PV9OnT7fHjxydrHb73L7F/sh99NmzYYKdJk8betGmTHQ0C9b158+bZjRo1sq+55ho7Q4YMdpEiRexWrVrZv/zyi7Yu+l7k+55vv7v9GzlypKNdtPU9L/1PzJw507755pvtjBkz2vny5bO7dOliHzp0yNFGfkd+V9bh3xeDMXz4cLtUqVL2pUuX7Fj15ptvJmsfCTm+JDX2rVixwjGWZMmSxf7nn39sE/re9u3b7Z49e9q1atVSfc+tb/ow9l2d1H7cFYx9KYexj/O+lDLdwPO+dFYYPProo1bhwoWtTp06WcWLF7d++eUX64033rC++uora8OGDVbmzJnD8bLW6NGjVbafM2dOK1bNmDHD2rJli9WzZ8+g11G3bl1r6tSpWnz8+PHWzz//bN1+++3xsZtvvtmqVq2aNXbsWGvKlClBv2akSF/LnTu39fTTT1v58uWzDhw4YE2ePNm65ZZbrFWrVlk33XRTWF6Xvnd17rjjDuuBBx5wxKSv+S+b1PfE22+/bXXv3l19hsaNG2f9/fff1quvvmqtW7fOWrNmjZUpU6aQv+bFixfVa/Tq1ctKmzatFaveeust9Zl+8MEHg17Hvffea5UuXVqLP//88+rOT/Xq1eNjzZs3t3LkyKFe98UXX7SinYxvr732mlWhQgWrfPny1qZNmyLyuox93sTycVcw9oUPY1/SOO+L7pwj0ud9YUls58yZY9WvX98Rq1q1qtW5c2f1qNwjjzwS8tfcuHGjOjjIzoikc+fOWRkyZLDSpNGf6j59+rSVNWtWK9qUKlVK/Uvo7Nmz6oT8f//7n1WoUCHHz+SxgMGDB6vBNVu2bFY0GzRokBaT/la0aFF14J04cWLIX5O+d/XKlCmjLnwFYlLfu3DhgkqQ5AR20aJFVlxcnIrfeuut1j333GO9++671pNPPhny1/3iiy+sQ4cORXzKQFLj25kzZ6wsWbJY0ebGG29U/xLau3evugAh44SM5T4yprdu3VodXIcOHRr/fkarZs2aWcePH7eyZ89ujRkzJiKJLWOfd7F83GXs+/8Y+yKP877ozjkifd4Xljm2/kmtaNmypfrvtm3bwvGS1oIFC9RJiZxU+tu3b5/18MMPq7vIGTNmtK699lrr8ccfV4Oxzx9//GG1adPGypMnjzohkzkGX375pWM9S5cuVSc3H3/8sfXCCy9YRYoUUW1PnDih7iLIG7Br1y6radOm6uSiY8eO8c/hT5gwwapYsaK6Y1OwYEHrscces44dO6Zt69dff23Vq1dP/b7cLZA7CHLFxLdfZZt2796ttkP+lSxZMv539+zZY23fvj2o/ff5559bJ0+ejN9m/6ss8oGRk3UTydxDeZ/kpC8c6HvB9T05qZMLQ0kxqe/JVU3pY+3atXMkQXfffbcaG2TcCFf/k3Hguuuu034m74kcJPLnz6+elClbtqw1YMAALTlp0qSJGm9kO+XO0erVqx1tPvzwQ/U3LVu2TJ2Iy2dKLhb5xiWpbbB+/Xo1/spnTRJ8cf78eXWAkrukMvYWK1ZMzVuVuL9p06apJyvk9+WpC1nXt99+q34mf9+vv/6qXt839iU8zsi4K/+CMXPmTHm+N9GxT8bbSN39TA45dslxI5IY+zjuCsY+xr5ow3nfhFR73heWO7Zu5JFQIY+ShcOPP/6oTq7Sp0/viO/fv1+dLMkJpzwiXa5cOZXoyl1lubImyfDBgwfVXRVZfuqpp6y8efNaH330kboCLu18SbnPsGHD1O/17dtXnaD5rvJfunTJaty4sVWnTh11xdx3x0I6lJwYPvTQQ2r9f/75p3o0W04oV65cGb/N0qZLly6qM/bv39/KlSuXavPNN99Y9913nzoh/e+//9TdBXl8SSS8miG3+OXE7/8eZ786ciddTnzlUT1/8mib/Ey21X9fRCt5v+URTel3clFBLj4kfNQrlOh7V9/3pK/L1ThpL49NyoUi6eMm9z1fsuY21UJi8lmWi1xuT3ckt/9VqVJFi2/evNm67bbb1PgiY58ckCT5k4tYvgJ7kixKGzmgScIpbd955x11QJP3s0aNGo51SlIrSbJcIZcDj8+RI0dUcixTQeSKrBxI5W+VMXTFihXq9eV9lkfGZOz67bffVFLkI3dEhwwZosZheexXxlR5dPv777+3GjVqpD7DcrdbxjtfYi6v4eP7bP/1119BjX2ScLtdFJUnjYT0P/9HpsDYJzjuMvYx9kUHzvuiO+eI2HmfHSEPP/ywnTZtWvu3334Ly/qLFi2qigT5e+CBB9RE5LVr12o/u3LlivqvFNyQXbF8+fL4n508eVIVvypZsqR9+fJlR8EOKdJy5swZx7o6d+6sfvbcc8854rJOicsE7IS++eYbR1wKXmXPnt2uUaOGffbsWdftFHfddZcqsuWmXr168cVFrsaRI0dUoSUp+JUYKQDWpEkT2xRly5aNn5yeLVs2+4UXXoh/H0ONvnd1fe/WW2+1J0yYYH/66af222+/bVeqVEn97ltvvWV035MCUXFxcWqsS0iK+vj64uHDh0P6mhcvXlSv2adPH+1ndevWVWPK7t27Ex1PWrRooT77u3btio/t379f/Z78vn/hmDp16mgFqnzv/cSJEx3xqVOnqrE34bgqpJ20X7lypVr+/fffVbuWLVtqn9GE21qxYkX1Wm5kTExsXEzKli1b1LY888wzibaR/fP444/bJhk9enSSxaNChbGP465g7GPsiwac90VvzhHJ876IJLayIwOdPCSXVF9+5JFHHDE5SZIqyc2bN0/yd2UH3nLLLVpcqnXJdvsq6voS26FDh2ptfYmt/0nkU089ZefMmdP+999/1eCf8J8kXL5tnj17tvr9+fPnJ7mtSXWyYL3zzjvqtaXDJUY6f/Xq1W1T/Pjjj+qDLB8a2W458b9w4UJYXou+lzznz59Xg1yuXLm0C0am9b127drZ6dKls8eMGaOSxR9++MG+6aab7PTp06vP2N69e0P6egcPHlTrlarICcl4I/Gnn3460d+VBFWq/rpd0HrsscdUsvnff/85EtuPPvrI9eAmVXjlfUyoWbNmKhn1H/fk4mbCbfYlYRs3bkzyb00qsQ1W//791Wv//PPPibYpWLCg3aZNG9skkUpsGfuCF2vHXca+/4+xL2Vw3mdGzhHu876wP4q8fPlyNb9VHtEN9/fL+t8Ol4Iq8giqPKKcFHl+3P+ROyG3yn0/T7gOmaPrJl26dPHzznx+//13dStfnvd38++//6r/+uaHBdrWcJBH8WR+ljxOk9S+jfbiKQnVqlUr/v/L45G+91IeEQ8H+l7w5LHTHj16WN26dVPzNOVRflP7njzGK3NIZJqC/BPyaK7Mf503b17YisD49z+pGRBoPJHxUaZfyLxbf/J5kUeJpbCSPKYUaOyTegMJCy/5xj6pqSCPLgca++TxbHn8KJJkn8lcItlH/gWlTO1/KYGxLzixdtxl7Pv/GPtSBud9ZuQc4T7vC2tiK1WKZY6V7DiZqyqJX7jIvFi3idHhkNjXFUlxFP/5c3JyKB1MDmJuEjvpixSZ/C0XH2QOnP/85IRk315//fWWiaQQjVSdlPcgHIktfS/5ZI6jOHr0qNF9T75q7NNPP1WfK5nvWaJECfVP5o7KZ13msISSnBjL4J/SY59bXMa+G264QX3tUVLveUqR+Tty0XLkyJEB522FqzaE6Rj7ghOLx13Gvv+PsS/lcd43PSpzjkic94Ut05SrAXfeeadK6uT7a8Ndrl6KQskEaf83UIqiSMW+pMiJ544dO7S4r9qX/DxYcqfmu+++s2rXrp3k9/f6KprKtrp9z6JPqK/eJlUR1EeKYsmdG7lIYSq5iyZXscKBvpd8vjuM/oOuqX1Pvr9b/vkSI7ki2apVq5C/jlwslLHDf+zzfa1IUmOf7GspcJfY2CcX6ZKTfMp2+b6fM6lxS9rJieDWrVutypUrR2zskwtdsk634hU+UmhQquf7nviAE2NfcGL5uMvYx9gXLTjvyxx1OUckzvvC8nU/UolWKlnKidHChQsjcoVAHkGQNyjh10jI67do0UJVAV23bl2ij1DJ1/P89NNP6gvufaTi56RJk1Ql0eQ8IidftXH58mVVSdmfvHm+r6CR/SXltuXugX8p7ISPesl3VCWWoAVTelsexZMDkf9jAAnJCadsk9x1ina+xywSkjtnixcvVl/6HA70Pe99Tx5/9SdfMyVVb+WumK8KrYl9LzFSbVA+67169Qpb//Mf32TMlSq/kydPVu+N23iSNm1aNe7IHeaE1YSlSryMCzImyIXB5Ix9khjK9/e6nXD4qirLGC1jtVRDlgTXbVt9Y19iX9l1tV/3IxXTZ8+erf5G3wUIN3JBQpjc/8KJsY/jblIY+xj7IoHzvujOOQ5F+rzPDgMpluIrFiWVMRP++/bbb8Pxkva6devUay5cuNAR//vvv+1ChQqpIilS/VgKNgwZMkQVIjl27Jhqc+DAAVUgRCZcDxw40B4/frxduXJlVW103rx58evyFY+SSdduxaOyZs3qum1SiEV+Typ8ybrfeOMNVdSlcOHCjnW99957qp1MqH7ppZdU5bBu3bqpys4+r7zyimrTq1cve8aMGfZnn30WdFVkKYrlVsnZnxTCkf134sQJO9oVKFDA7tChgz1q1Ch70qRJdr9+/ew8efLYmTJliq/CGmr0Pe99b/DgwWp8kCrV8v5IITYpTCCftWnTphnd93wF5zp27Gi/9tprqnBZo0aNXIs7hdKcOXPUa+zYscMR37RpkyoWkTdvXlUkSfb3888/r/Z/wqrAMm4VKVLEHjFihPrcSNV3KQa1evXq+Ha+4lFu1eXlvZfx1J8U72vatKl6b9u3b2+//vrrqiqijGnymUy4Lhl3Zf1SOVHec2kr417Csal79+5qXcOGDbNnzpxpL168OOiqyJ9//rlrJWd/PXr0sIsXL+6oEhmtpMql7Bv5d+edd6q/T4rmybLsz3Bg7OO468PY9/8x9kUW533RnXMMjvB5X1gSW99XW7j9C3VVy4RuvPFG7as2hFQqljcqf/786oRNTtyeeOIJRxVPqWDaunVrVaFLkiCpkvzFF1841hNsYivkzaxataqqIikltm+44QaV+MtXayQknUZO7qSdVHSW7ZCTOJ9Tp07Z9913n9pO2ZaEJ3NXm9jKSaO037x5c5LtpDpZp06dbBPIB6hatWp27ty5VXVa+SDLSXWgvzG56Hve+p5c2LrjjjvUxSapFCz9WJK/hEmKqX1PyJghn1n5jMvAXLNmTXvWrFlhfU0Zx/Lly6cSGH+SuMrX6PjGNfk6BEkiE9qwYYPduHFjlQTLNjdo0EBVl0womMRWSCVySZbl5zL2yudSxkE5sPkqLvtMnjzZvvnmm+PbyXoXLVoU/3O5ACkVGmXf+h9LrjaxlTFB+p981Vli5OT0mmuuUQdjE0gF5MSOu+GoaunD2MdxVzD2OTH2RQ7nfdGdc3wb4fO+iH2PbSRMmTJFvYG+O7FIPvkKDrmqEuirOFI7+l7o0fe8e/HFF9X3bvt/xyyCJ1+DIAd7/xMBODH2hR5jn3eMfaHH2OcNY190jn1x8j9WjJD5WfK1DR06dLAGDBiQ0psTE+SrcmS/zpo1K6U3JarR90KPvufdqVOnVMGo8ePHJ1mQBlc3f/S2226zXnnlFXZbEhj7Qo+xzzvGvtBj7POGsS86x76YSmwBAAAAAKlPWKoiAwAAAAAQKSS2AAAAAACjkdgCAAAAAIxGYgsAAAAAMFo6rw3j4uLCuyUwTqTqjtH3kFJ9j/4HN4x9SCmMfUhJjH2I9r7HHVsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNHSpfQGAAAAAEAk1KxZ07G8atUqrc2VK1e0WNq0acO6XUg+7tgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaCS2AAAAAACjUTwKMMRdd92lxVq2bKnFChYsqMVOnjzpWJ44caLWpnnz5lqsT58+QWwpAABAdOrZs2fAQlFuMUQ/7tgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaCS2AAAAAACjxdm2bXtqGBdnpbQSJUposZ9++kmL5c+f37G8bds2rc28efO02IIFC7TY6dOnHcvbt2/3vL2xzmPXSbZo6HspYcqUKY7l+++/P2TrPnbsmBbLnj27Fuvdu7cW++WXXxzLWbNm1dp89dVXYe0vkep7qbn/IXGMfbp8+fI5lr/99lutTfny5bXY6tWrtZj/744dO1Zrc+HChVTZRRn7kBr6X6SPu26Fotz+1tq1a3saw5ByfY87tgAAAAAAo5HYAgAAAACMRmILAAAAADBa1M6xbdmypRbr1KmTFmvevHnAbXX7E93+Hrd2Z8+eDWqOrdu83vnz5ye5bJpYnWsRLZYvX+5YrlOnTtDr2rt3r2N5586dQa9rx44djuWyZctqbXr16qXFfv75ZytUmGfmrmTJko7lBg0aaG1uueUWT/u4YsWKjuVff/1Va/P6669rsa1bt1qxjrFPV65cuYD9Jdgx322Obb9+/azUiLEveEOGDLGiUb169bRY/fr1tdjQoUMdy0uXLtXauMVCKVbHvtGjR2uxnj17arH9+/drsXbt2jmWmXMbHsyxBQAAAACkCjyKDAAAAAAwGoktAAAAAMBoJLYAAAAAAKNFTfGoqlWrOpZ/+uknT9vgpTDUoUOHtDZ79uwJ+AXzokSJElf9el7bubVp3bq1FovWIlOxWkQgJVx33XVazL/4ysWLFz0VKRs0aJAWW7t2rWP5yJEjlslSWwEVtyJQI0aM0GL+hbxy584d1u3yL64nhg0bpsXGjRvnWL5w4YJlMsY+Xfr06QOOac8++6yn/du5c2fH8nfffae1adGihRY7c+aMFetS29jn1ZIlSwIWX0oNwv2exerYV7NmTU/Fo9q2bRtwn9SuXVtrQ0Gp5KN4FAAAAAAgVeBRZAAAAACA0UhsAQAAAABGI7EFAAAAABgtaotHrVmzJmRFmpo0aaK1+fbbbz0VjypevLhjuWXLllqbe++9N2ARF7ftctt2t2Is1atX12Lbt2+3UlqsFhFICeXKlQtYGGrWrFlam4kTJwYsohGLUlsBlTvvvFOLvfbaawF/b/r06VrsxIkTWsytbx07dsyxXL58ea3NkCFDtNjdd9+txfr16+dYHjNmjGUyxr7wunLlSsD97faZWLRokRXrUtvYF437JVosXbrUU6HBUErtY9/o0aMDFpnav3+/1qZdu3ZajIJSV4fiUQAAAACAVIFHkQEAAAAARiOxBQAAAAAYjcQWAAAAAGC0dFaU8C/S5FZEKWvWrJ7WtXfvXsfyhg0bPP3e4cOHA8bc1jVw4EBP6//3338dy3nz5tXauP3dZ86c8bR+mKtatWoB27j1vRUrVoRpixBNvvnmGy1WpkyZiG7D+vXrtVj79u212I8//qjF0qWLmkMNDPDmm286lrt37661ue+++1Jl8Si4Gzp0qGO5Xr16Wptly5aFtXBT/fr1A/6e23Z5+T0vfzPCz78QolvM7Ri4cuVKLeZWJLdt27aO5b///jvILU29uGMLAAAAADAaiS0AAAAAwGgktgAAAAAAo8XZHr/xNtJflrx27VotVqVKFS3mtvkbN250LFevXt2KBhMnTnQsP/LII572s9v2e503HE6p/Yu6Q2n+/PlarEWLFgHnmb399ttWahSpvpda+l8ovfXWW1qsQYMGjuXy5ctbJmPsC6+SJUs6ln/55RetzYkTJ7RYzZo1A9bcMB1jn9mWLFkS9Bxb/zm1Q4YMsSKNsS+wokWLarEaNWposVmzZmkx//m58+bN09qMHz/eSo1sj+d93LEFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGS2cZJNaKuLj9PbH2N0KXI0cOLdawYUMttmPHDsfy8ePH2Z0wUqZMmVJ6E2CQv/76y7E8ZcoUrU23bt20WJYsWcK6XUByi0V5LRS1dOlSLZYSxaJw9f7++29Psdq1a2uxnj17OpbHjh2rtSlcuLAWmzt3rhZbvXq1lRpxxxYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABgtaotHbdu2TYtVqVJFi9m2bZli+/btxm47QidDhgxaLE0a/RpT2bJlHcs///wzbwOiXsWKFQP276xZs2ptTp8+HdbtgrnczgeAaONW3MlrsSh/DRo0CMEWIZq5FXdq3769Y3nv3r0BC0yJtm3barF27doFfL1YxB1bAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgtKgtHlW+fHktFhcX5+l3Dx8+bEWj5cuXB/x73GL58+cP63Yhstz658GDB7XYtdde61ieM2eO1mbYsGFa7LffftNiv/76q2P53LlznrcXqVP27NkDFu+rU6eOFitXrlzAMcy/kJ4YM2aMFnv11Vc9by8ARMqSJUtCVijKjVtx0aVLlzqWly1b5qmAFczVr18/T7Eff/xRi61cudKxXLt2ba1NLBaU4o4tAAAAAMBoJLYAAAAAAKOR2AIAAAAAjBa1c2y9zjlwi82fP98yYd6w27a7adGihRZbuHBhyLYLKc9tzsRrr70WcN75jBkzPK2/devWAT8jV65c8bQuRK906ZxD+v3336+1admypRYrXbq0FsuUKZNjuWTJklaoFClSRIvdddddWow5tgC88J/fmpz5rvXq1QvZuoI1dOjQgHNs/ZeReo0fPz7g+eEnn3yitWnXrl3Mzbvlji0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADBa1BSPKlGihGO5ePHiWpu4uDhP6zp06JAVjQ4fPhzw73GL+f8eYs/cuXO12JkzZxzLL730ktamcuXKntY/Z84cx3Lbtm21NrNnz/a0LkSHMmXKaLFFixY5losVK+ZpXf/++68W27NnT8BxKHPmzFqsYsWKWmz79u2O5bFjx2ptZs6c6Wlbkfq4jX1ej58wm1vhpiVLllim8C/wtGzZMq3NkCFDIrhFiEVu52+z/WJuBUJXrlwZ8HwxsSJT0Yo7tgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGhRUzzKv+CTWzGTfPnyaTHbti1TlCtXzthtR+R9/fXXAQtmlCpVSot16dIlYKxfv35aG7f1U7gsep0+fVqLtWrVKqh1eSke5aZChQpabMuWLVpsypQpjuX333//qrcRqZfbsdJrDLFXPMokDRo0SOlNABIt2tizZ08tVrNmzYCx1atXW9GKO7YAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoUVM86syZM47lNm3aaG2mTp2qxbZt26bF5s+fb5lQPCouLk5r4xYDxLlz57QdsXXrVi3Wt29fLXb33Xc7lqtXr661eeqpp7TYoEGD2PlRat++fZ5i4fTss89G9PWQOlSqVMmxnD59eq3Nrl27tNiJEyfCul2IvCFDhmixevXqBSwy5Va0aenSpZ7WP3jw4CC2lEJRiG4//fSTFkuTRr+/WaxYsYCFKSkeBQAAAABAmPAoMgAAAADAaCS2AAAAAACjRc0cW3/bt2/XYm7zAk1Svnz5oL5M3v/33OYb+8/fFS+99JIx848RnLRp02qxAQMGaLF8+fKxixFyHTp00GJ79uzRYrNnz2bvw7Pvv//esZwxY0atzcqVK7XYP//8w15OBdzmzy5ZsiTgfFo3bvN1g90Gr68JRIsrV654mnfrNV+JBtyxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFIbAEAAAAARova4lHhljVrVi3WsmVLLVanTp2g1u+2rvz58wecjB0XF+dpXf7t3Nb1zjvveCrssn79ei2G6JMrVy4t1qNHDy02dOjQoNb/559/BvV7iE3+BXveeOMNrU26dPoh5JFHHtFiu3btCvHWIZb5F7szqXAJwq9+/fpB/Z5/gSmv63I7plIoCtGuZs2ajuVPPvlEa+M2tu7bt0+LzZs3zzIFd2wBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRUm3xqHLlymmxDz/8MKgiTW4Fn9za+ce8FsTw0s7L613Na6ZmvXr18lR04tixY1osffr0juVOnTppbTJlyhTUdhUqVEiLde7cOah1vfLKK1psypQpQa0LsalZs2aO5Ycfflhrs3fvXi22YcOGsG4XIN5//312RCrldjz2L/DkVhTKa9Ep/8JQQ4YMueptBFKyUJT4+OOPrUDn/1euXNFibdu21WKrV6+2TMEdWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYLRUWzxqz549Wmzjxo1arGrVqgHX5VY8Kth2bm0OHz6sxQoUKOBY/vfff7U23bp102IUdgmsXr16WqxatWqeJuuXKlXKikbz5s1zLI8dO1Zrc/ny5QhuEaJJjx49tNioUaMC/p5bQamjR4+GbLuQOnkp2nj99ddrsRUrVoR1uxB5wRZ8Gjx4cFC/Jxo0aOBx65DatGnTJmCRpjVr1mhtxo8fH9TruZ1n9uzZM6gitvv27Yu5QlFuuGMLAAAAADAaiS0AAAAAwGgktgAAAAAAo8XZbpNXkjGP1GR16tTRYh07dnQsz58/39Mc2PLly2uxKVOmOJbddv2RI0e02KRJkwKuv3fv3p7mEYeSx66TbJHue08//bSnOTd58+bVYvnz5w+4/rJlywa1XefOndNiO3bs0GKdO3fWYr/88kvAL+U2SaT6XiyOfY0bN9Zis2bN0mL+8/afe+45rY3beGh630rNY1+0+Prrrx3LjRo10toUK1ZMi+3fv9+KdbE89rnNp12yZImnebFe5uK6/d7QoUM9tcP/Se1jX9GiRbVYjRo1gpoXmyZNmoDHTy9tRIcOHYKaY7vaoPm0Xvsed2wBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRKB4VQf4Tvt0mQrtNmK9WrZoW27Bhg5XSYrWIQKFChbRYxYoVtdiQIUMCFiCbMWOG1qZ48eJa7KOPPgpY8On06dNamy1btlipUSwXUAmlbNmyabEffvhBi12+fFmL3XvvvY7lvXv3hnjrzBWrY1+0OHv2rGM5Q4YMWhuKR8Ve/3M7pg4ePDhk60+tn6dQYuxDSqF4FAAAAAAgVeBRZAAAAACA0UhsAQAAAABGI7EFAAAAABgtXUpvQGqe+BzJAjjw7sCBA55iixcvZrciqs2ePVuL7dq1S4uNGjVKi1EsCinlxRdfdCwPHz48xbYFZli6dKljuUGDBim2LQBSDndsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0SgeFUFp06aN5MsBSGX69u3rWJ4xY4bWZurUqRHcIuDqjRw5MsllpG7+haLE0KFDU2RbAEQX7tgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaHG2bdueGsbFhX9rYBSPXSfZ6HtIqb5H/4Mbxj6kFMY+pCTGPkR73+OOLQAAAADAaCS2AAAAAACjkdgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaCS2AAAAAACjkdgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaHG2bdspvREAAAAAAASLO7YAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKOR2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoMZXYnjp1yipQoIA1ffr0lN6UmFGzZk3rmWeeSenNiHr0vdBr37691bZt2zCsOfbQ/0KPsY++l1Loe95duXLFqlSpkjVixIgwviOpC8debzjuRunYZwfhgw8+sH2/umTJEvX///zzT0ebRYsW2fXr17fz5s1r58yZ065evbo9ZcoURxv5HfldWYeQ/y/rDtbw4cPtUqVK2ZcuXbJj1ZtvvpmsfZRwv7v9mzlzpqPtvHnz7CxZstj//POPHQ3oe2b3vW3bttn9+vWzb7rpJjtbtmx2oUKF7KZNm9pr167V2m7YsMFOkyaNvWnTJjtaeOl/CTVs2FC1eeKJJxxxxr6U6X/i999/t1u1amXnypXLzpw5s127dm37+++/19ox9nnDcTf2j7texr7t27fbPXv2tGvVqmVnzJgxybEx4bleiRIl7MGDBwe9XdOmTbNz5MhhHz9+3I5V06dPt8ePHx+S9y+xf7Ifo/XYy3E35bxp4HE3LHdsP/vsM6tRo0bWhQsXrCFDhqgraZkzZ7YeeOABa/z48eF4SevixYvWq6++aj3yyCNW2rRprVj11ltvWR9++GFI1tWhQwdr6tSpjn+1atVytGnevLmVI0cO9bomoO9Fd9977733rHfffdeqVq2aNXbsWKt3797Wjh071FW67777ztH25ptvjm9nonnz5lmrVq0K++sw9nm3d+9eNcatWLHC6tevnzVy5Eh11V2OVz/88IOjLWMffc+H425gMta99tpr1smTJ63y5ctbkTJ69Gh1hzFnzpxWrJoxY4Y1YcKEZK2jbt262vme/KtSpYo6Z7799ttj4tjLcTf6xr5IH3fTWWHwxhtvWNdcc431/fffWxkzZlSxxx57zCpXrpzaQb169Qr5a37xxRfWoUOHIv7o4unTp62sWbO6/uzMmTNWlixZrGglA1qnTp2SbJMmTRqrdevW1pQpU6yhQ4dacXFxVjSj70V335OLKXKxK1u2bPGxLl26qBMhiTds2NDRXj7PgwcPVoNcwt+JdufOnbP69OljPfvss9agQYPC+lqMfd69/PLL1vHjx60tW7ZYZcuWVbGuXbuqY5Mcl9avXx/flrGPvhcOsXjcFc2aNVOfrezZs1tjxoyxNm3aFPbX3Lhxo/Xzzz9HPAGT8T1Dhgzqfbqac8KUVKpUKfUvobNnz1rdu3e3/ve//1mFChUy/tjLcTc6z/sifdwNyx3bEydOWLlz545PakW6dOmsfPnyqTu34bBgwQKrZMmS1nXXXaf9bPv27epDmj9/fvX6smMHDBigDZBNmjRRVwrkQyxXr1avXu1oI0m57ORly5apwUDm8xYtWlT9rH79+mqeh7xBcmVMOtfzzz+vfnb+/Hk1QJQuXVrtk2LFiqlnyCXub9q0adYtt9yifl/2oazr22+/VT+Tv+/XX39Vry/bIf/kdX127dql/l0NGYTlznpS7rjjDmv37t0ROVAlF30vuvte1apVtYNk3rx5rdtuu83atm2ba9+TPrpo0SLLJK+88oqa+9W3b9+wvxZjn/f+t3z5cnU3wndwFdLf5aR8w4YN1u+//+5oz9hH3+O4602ePHlUUhtJMvZJginHKn/79u2zHn74Yatw4cLq2Hfttddajz/+uON8548//rDatGmjtl3GAXly6Msvv3SsZ+nSpep49/HHH1svvPCCVaRIEdVWzjUefPBBdTyTsadp06bq7+/YsaP6PRn/5S5rxYoVrUyZMlkFCxZUN3iOHTumbevXX39t1atXT/2+nINWr15d3aUVcpyVbZJzMN+xV/qkz549e9Q5bjA+//xzdYfdt82mH3s57kbneV+kj7thuWMrf/ioUaOsgQMHWp07d1Y7Qz6k69ats2bNmhWOl7R+/PFHdSXU3+bNm9VJc/r06a1HH31UvVHyRsgH2ldsQN44aSMDirz50vadd95Rf4e8oTVq1HCsU5JaSZLlTox88H2OHDmikmN5LEauyMpAJoObvHlyC15eX+5M/fLLL+qR7N9++00NzD5ydULuWt16663Wiy++qAbsNWvWqDvfcsteBsknn3xSDaS+xFxew8f3KMlff/3laZ/J68ljAfL+SMIh+0Nex5/8TKxcuVJ1zmhG3zOj7/k7cOCAuvDlr0KFCupilPS9li1bWiaQEw25Qjl58uSwXchLiLHPe/+TA7scvP35rnLLhcnrr78+Ps7YR9/juBu9ZOyTGwpyzpbQ/v37/197dwIkRXU/cLwXVg45hIAhHAEUWEAEESJEwyFyB1HkCuEKKAlgQEBRQQRBEATlCkQQRINHUG5I5AqIoBxlLUhBsBBEjkJgl2Mj93Js/+vX/xpqu18v3fTOzM6b/X6qtvT99k1P0/P2vX7T/X5tnazLVSIZ++TKkEx0Fy9ebN3NJONbSkqKNd5J+bnnnrO+YJ0/f741Zko953gzduxY63XyZaX0I/L/4vr160bLli2NBg0aWFeqQ32JTGLlYkjv3r2t7R86dMi6o0wuosh4FtpnqSN3LckEePjw4UaxYsWsOmvWrDG6du1qjbc///yzcezYsZtL+TJ/OSxL/OQ89f+XLt8eSbQqY1T79u21H3sZd2P3vC892uOuGQEXLlwwO3fubCYkJNxcmC6LgZcvXx6JtzOvXbtmvdcLL7yg/K5Ro0ZmkSJFzCNHjtjiGRkZN/+/Xbt2Zr58+cyDBw/ejB0/ftx6nbzeuYC9QYMGSoKqxo0bW7+bPXu2Lf7RRx9Zi/C/+uorW1zqSf0tW7bcXFgt9Z566inzxo0bWe5rjRo1rPdyI0kY5MeLHIsWLVqYs2bNMleuXGlOmzbNLF++vPX+//73v11fI8enf//+Zqyj7cV223OzefNm6+935MiRrr9PSkoyW7dubeqiY8eO5iOPPHKz7JY8Klzo+26v/bVt29ZKXnHu3DlbXBLeyOf09ttvK6+h76PtMe7enrfeesszsV44lCtXzkpI49SzZ09rTHNLShga0yTRlexj5vHx/Pnz5j333GNWrFjx5lgYSpQliUkvXbpk29af/vQn63fDhg2zxWWbEpekT5mtWbPGFpeEV3KeWb9+ffPy5cuu+ynatGmTZf8WOve8XWfOnLH6NjlXz4pOYy/jbuye97WN8rgbkVuR5dJ3UlKSdZ/0ggULrEvdshBdriQ5b+8Nh7Nnz1rfVjm/EZA1t7IwWb4NK1++vO13ofu2b9y4YV12b9eunW39gawRlm/L5FsPueUkM7k33C1Blfy75du5zBYtWmR9YyLfGJ4+ffrmj6xpEBs3brT+K9+iyDctchXYuW7D7z3m8q2JnytmcizWrl1r9OvXz2jbtq0xaNAg6xtCuQot6wLdyLGV/Y51tL3YbntOqamp1t+Z3CaWVYp3Xdpe6JguWbIk24k+/KLvu732J7ciylWcP/zhD1afJ99gDx482LqbKLTmTNf2R98X231fPI+7OUXuknOe98nnKZ+rHGM573QKfa6rVq2yrurKldYQuTIlV7nk8/zuu+9sr5O7D7O6A0f6FWf7k2RWcktl5vYXWooTan9ym6/cCjxs2DDrdmW3/fQit0oHuVorV6Xltmy325B1a3+Mu7Hd9/WP8rgbkVuRBwwYYE1g5d7p0AGTNa5yq4V05nKpOxKcf9yyfkLIrSpZkcmv3IqS+d7vEGkc8sFLRi/Z9xA5CXcjay9Ct6eEyL3jsnZQBq+sTuyF3B4tx0pu/8gJssZEJuVyC6Xc8hJaO5z52OqQwIK2p0/bk9v4H3/8cWtgly+QskpQoUvbk1vS5JazHj16WGukoom+zx9ZKjJjxgzrRDK0dEXWIckyDPlixa0N6tL+6Pv06fvibdzNSc6+T87p5GLErc77hKzhcy4zE6GMzvL7zNvI6rxP8sc4Pzdpf3L7sORh8Wp/wmtfI0FuQ5b2J31iVnRof4y7sT/naB3lcTfsE1v5BmjevHnWzmb+FkDWE8g/TtYYSB3nBDA75I9TDoDbovxIyOpbO7e4TIxr1qxpTJkyxfU1sqg7VoT2Ra4COTtq+bbFbQ1kLKHt6dP25LOSdT2yBl6uYtxqYJe/68zrL2KVZPGTRxfJ+nznt5gyeZeYnOiEM2shfV+wCaBMJqTtyThUu3Zta8wScqeRE30fbS+SdB93c5Ksi83p8z65U8J5xUvGXunrZfLoJqtJRzTXo0pCH7k67VyfrNvYy7irx3nfgCiOu4mRuDVEvkGRW3zdnrcoB93td9kh35hJNmRZnJ9Z6NZiSTGdFelg5ERTTkidJNOcdFjZaQiyX5KOXhZZ3+rbB6knx0Zuf5EPPCuR/PYsdIXb2elK0gWZiETz2XRB0Pb0aHvyXpLwYsOGDVYyOckGmRXpS+SOCUmGEOvkZEH6uN/97neug6/8LFu2zFr2EC70fcHI4zgyP7NbnqEsJ67Oz46+j7YnGHdjk9xu6Tzvk/MXSQR6q/M+UaFChSzP+0K/D0rGVOlTpD+5VQLB0FM8ZF/lCla02p8sEZQrYre6DVmXsZdxV4/zvmiOu2FfYyvfUklWNzmBy5xWXR7GK5mIpSOKRKZQOVih+7Uzd3CSulqyk0rjd7t9RdbKSvavFStW2K6ySMY8yeQs6y+kkwxKbsGWD2nu3LnK7+S+8lBWZTnZlUm0ZCaTxua2r6GGId9kuPGbeltu1XGSfZTjVKtWLWt9cWahZ0xJ5rRYRtuL/bYnJMveZ599Zj0fzy0bY2bS6cqz6WK97QnJhi79nvNHyKMg5P/dbn3LLvq+YI86y5xZdenSpdajQWRdXGb0fbQ9wbgbm6Tvk0lh5seYyFgmY5qcbzrPCTOPadInf/PNN8a2bdtu/k7GxDlz5lhPz8jOLZoy9soFHMmk7DZhDI2jcu4pj/iZMGGCNc657Weo/cmtzW6CPO5Hzm1lzXfm9cW6jr2Mu3qc90V13DUjYNy4cVamqwcffNCcOnWqlfGqevXqVuzjjz+OxFuaixcvtrb//fff2+K7du0yCxcubJYoUcIcPny4OWfOHPOVV14xH3jggZt1/vvf/5qFChUyy5Yta77xxhvmxIkTrQx4+fPnN7dv365kRXbLtCdZwyR7mJNkG/v9739vZX3t0qWLOWPGDCsLcb9+/cxf/OIXtm1JVljZvmRUlWMmdSW7X+aMe88++6y1rbFjx5oLFiwwN2zYcNsZynr16mU2bNjQHD169M3jIcdHspBJBkCnAQMGWFmTM2dKi1W0vdhue9IfyPtINjzJ3uf8kazWmcm+SEZ1ZzY9nUQyK7Kg7/Pf/g4fPmzWq1fP6ifee+89c8iQIWbBggWtscqtjdH30fYE4643yfArY4P8tGrVyur35EkVUpbxJBKSk5Ot91m7dq0tfuzYMfNXv/qVNXZI9uN3333XOt+Rc7S0tDSrzsmTJ81SpUqZd911lzX+ydhUu3Zta4xbunTpzW2FsiIvWrRIeX/Jiiznjm769u1rvU6yCsu2Z86caQ4aNMgsU6aMbVvSD0m9+++/3xw/frz1tAoZo2X8DZk0aZJVR/qrf/7zn9bTLIJmRd6zZ49rJmcn3cdext2EXDvuRmRiKySdufxDJMWz/AMknbmcgEVKenq6WbJkSevgO8nEVVJay74UKFDArFq1qvJokZ07d5otW7a0JsHyx9ykSRNz69attjpBJrbi6tWr1mRZfi+T5eLFi5t169Y1x4wZY/7888+2uu+//771YYfqyXb/85//3Py9dMaS+l1SxMu+ZE7D7beRSccojzG6++67zcTEROu4yfHZsWOH6+SodOnS5quvvmrqgrYXu20v9HiErH6cj4eQfqN79+6mziI9wNL3+W9/Z8+eNZ988knrpFe+yJNHe7z88suugyt9H20vhHHXm/TdWfXrQR8F50etWrXMZ555xvWxhnKSLuc5MqbJxQrph6W/DJFHPMpjYkLnhnLO6nzkYdCJrZALBzLeyjmwjJs1a9Y0X3rpJetxkpnJRFUmF1KvaNGi1n7IJCJEvvDt2rWrtZ/O43m7E1uZtEj93bt337Ke7mMv4+7EXDvuRmximxNef/1164A5nzGL4JYtW2Z1ts6OGLS9SPv222+tbwrlv7g1+r7wo+/zh7ZH28tJH374oXXSHboSi+xj7PWHvi82x92IPMc2pwwZMsRay/vpp5/m9K7EjYkTJ1rZzJzrbmFH2ws/eQSGPAv7VokNQPuLFPo+f+j7aHs5SRIgyXrRv//97zm6H/GEsdcf+r7YHHcTZHYb1r0CAAAAACCK4uqKLQAAAAAg92FiCwAAAADQGhNbAAAAAIDWmNgCAAAAALSW6LdiQkJCZPcE2olW3jHaHnKq7dH+4Ia+DzmFvg85ib4Psd72uGILAAAAANAaE1sAAAAAgNaY2AIAAAAAtMbEFgAAAACgNSa2AAAAAACtMbEFAAAAAGiNiS0AAAAAQGtMbAEAAAAAWmNiCwAAAADQGhNbAAAAAIDWmNgCAAAAALTGxBYAAAAAoDUmtgAAAAAArTGxBQAAAABojYktAAAAAEBrTGwBAAAAAFpjYgsAAAAA0BoTWwAAAACA1pjYAgAAAAC0xsQWAAAAAKA1JrYAAAAAAK0xsQUAAAAAaI2JLQAAAABAa4k5vQMA/MmfP78S6927txIrUKCAErty5YqtfPz4caVOnTp1lNjJkyeV2AcffGArp6en32KvAQCIrtmzZyuxvn37KrF9+/YpsQMHDtjKlStXVurMnTtXiTVp0kSJlS5d2lZu1aqVUufMmTNKDEAwXLEFAAAAAGiNiS0AAAAAQGtMbAEAAAAAWmONLRBBSUlJSuy+++7zXIcjmjZtaivff//9Sp0qVap4rqfNat2tHwkJCUps165dtvL27dsDbRuxs1Z70KBBSmzkyJG2cufOnZU6q1evDvPeAUD2Xb58WYllZGT4GkMrVqzouf0XX3xRiZUqVcrzdcuXL1diTzzxhBJLS0vz3Bbi3x133KHEevXqpcRGjRqlxMqVKxe2/UhOTraVx4wZo9T5/PPPlZhpmka0ccUWAAAAAKA1JrYAAAAAAK0xsQUAAAAAaI2JLQAAAABAawmmz5W9bklkEDsJYPLmzesriZBb8oSgorUoPNptz+1YPvTQQ0rspZde8txWy5YtfSVycvs3pqam2sp79uxR6uzcudPXttzqOT344INKrE+fPkosX758tnLdunWVOvv37zciKZoJCXTu+9w+mwkTJiixxx57zHNb6enpSmzKlCmeSafiUbz2fYh99H3+uCVy+uSTT5TY+vXrldjixYs9t9+8eXMlNnPmTCOIr7/+Wok1btzYiEX0feFTsGBBJda9e3dbefjw4UqdChUqGLGoffv2SmzFihVRb3tcsQUAAAAAaI2JLQAAAABAa0xsAQAAAABaY2ILAAAAANBaopELVKtWTYk9/fTTvhY5b9myxYimGjVqKLGBAwcqsbZt29rKpUuX9pUooUePHtnex3hTvHhxW3nEiBFKnSFDhvhK7HL27FlbedGiRUqdL774QomdOHFCiW3atMlWvnr1qhFJn376qRLbsGGDElu9erWt3KZNm6gnj4K72rVr28rLly9X6rj1FX64JT3r0qWLEps2bZoSO3PmTKD3BETRokWVA3HHHXcosbS0NM/XuiUHdPPII4941tm9e7cSO336tBK7ePGir/dE+KSkpCixZs2ahW37znOwrM4J/CS8ifZ5JiLLrY9xm3O4JSC99957w7YfVx3njBcuXPD1umLFiimxPHm8r4P+9re/jWjyKL+4YgsAAAAA0BoTWwAAAACA1pjYAgAAAAC0xsQWAAAAAKC1uEweVbZsWVt5/PjxSp127dopscGDByuxc+fO2coHDx5U6lSqVMkIlzvvvNNX0hY//vjHPyqxQ4cOKbFRo0YZudnf/vY3W7lbt25KnR07dni+TixevNhWvnz5sqGLBg0aKLFZs2Z5JsgImowI4de1a9ewfTbDhg3zTIwyYcIEJTZ06FAlNmnSJM8kP0BIrVq1bAdj3bp1ysH55S9/qcTWr1+vxB5++GHPMdZN0ERAbonz+vTpo8RIGKQPt3O8kSNHBmof//rXv5TYuHHjsrF3yGkVKlTw/DydY3N2uCVjTE9PV2KvvvqqrTx//nxf23/77beV2KBBg2zlU6dOeSYWzSlcsQUAAAAAaI2JLQAAAABAa0xsAQAAAABaSzD9LArIYr1JrHLe592yZcsc25dY43YffsGCBQNty2fTybZwtr1HH31Uia1cudJW7tu3r1JnyZIlng+/1kn58uWV2Jo1a5RYtWrVlNhPP/1kK3fv3l2ps2nTJiOSotX2Yrnv+81vfqPENm7cGOhvu1mzZkps8+bNtnKhQoWUOlu3blVi1atXV2Jnz561la9fv67UKVOmjKELHfu+WFCsWDEl9sILLyixESNGRGwNrF/h3L7betqGDRsG2hZ9X/TNnj1bif35z38OdM7l1m9/9913hi5ye99XokQJJTZz5kxbuXPnzoG3n5qa6pmnpUuXLkrsm2++McKlSJEinnmJFi5c6Gt+kRNtjyu2AAAAAACtMbEFAAAAAGiNiS0AAAAAQGtMbAEAAAAAWks0NDd8+HAl1qRJkxzZl1hz8eJFJdapUycjN8uXL58Smz59uq28YMECI94MGTLEVn7++ed9Je9xSwbwzjvvRDVRFNy1adNGiRUuXNhWzsjIUOosX75ciX355Zeeh/n8+fOefzvi3Xff9ZVww2no0KG+HhQPfT388MNK7JVXXglbgho/r9u7d68Su3DhghKbPHmy599AUlKSr/1yS8yH2OVMINmnT5/A27py5Yq2iaJyuwoVKvj6W/bTD7glGz1y5Ihn24v0+VXFihWV2KxZszz7vmXLlkU9eZRfXLEFAAAAAGiNiS0AAAAAQGtMbAEAAAAAWmNiCwAAAADQWqLuiX+efvppX/X8cEuOsn//flu5bt26RizYt2+frbx69WqlzsSJE5VYampqRPdLRzt37jTiXY8ePTwTRbn5xz/+ocQmTJgQtv1CcM4kE27Joo4eParUGTBgQNgO++bNm331o86kVoh/lStXVmLvv/++r9du3LjRMyHk/PnzlVhycrISW7t2ra187NgxzwQ/fq1bty7Q6xDbTp48GbZt/fWvfw3bthA5efPmVWJjx44NlChq9+7dvhLdRjqpXNGiRW3l1q1bK3XmzJkTaLzu16+fEau4YgsAAAAA0BoTWwAAAACA1pjYAgAAAAC0xsQWAAAAAKA1rZJHXb9+XYkdOHBAiVWqVCnQ9hMT1cPx3nvvRXWxt1vyi7179yqxEydO2MoXL16M6H7Fi3hL9tGpUycltnDhQiVmmqatfPjwYaXO7NmzldikSZOyvY/IOWlpaZ59R3Y4k+uJ7du3K7FmzZp5bsutTUJfFStWVGIlS5b09Vq3ZFFO3377rRKbN29e2BJDIXe46667PMe9hIQEX9tySxq0YsWKbOwdoqVnz55KrFu3boG2NWXKFCUWzrmDMylUVomhBg8ebCvXq1cv8HuePn3acz4WK7hiCwAAAADQGhNbAAAAAIDWmNgCAAAAALSm1RrbjIwMJbZr1y4l1qpVq0DbL1iwoBKbOnWqrfzRRx8pdfr27Rvo/YBbee6552zlYcOG+VpD6VxP67YGe8iQIUod1gLFnw8//DBs22rUqJESmz59uhK75557Am1/8eLFgV6H2LR+/Xpf43WdOnUCbX/atGlKrEqVKkps4MCBgbaP3MF5jicqV64caFuHDh1SYpcuXQq0LURWnjx5bntdf1bS09Nt5fz58yt17rzzTiVWq1atQO/3/PPPK7EOHToY4ZKamqrEJk+ebCufO3fOiFVcsQUAAAAAaI2JLQAAAABAa0xsAQAAAABaY2ILAAAAANBagumWacatos8HVAf12muv2colSpRQ6owcOdLXg4rnzJljK7ds2dIIF7fj4Jb0pEePHnH/oHifTSfbIt32os0tSdm2bduUWPXq1W3lxER/ud7cErS8/vrrcZUoKlptL5bb3/Hjx5VYqVKlbOUffvhBqTNixAglVqhQIc8Hvrdp08ZXWw7Kb/uOBfR9wZQsWVKJDR06VIn95S9/sZWLFSvm6+/S7XNJSkry/JvQCX1feG3dulWJ1a9fP9C2evXqpcTcEo7qLF76PmeCp+wk+XKe27uNzadOnQpbOwunlJQUJda+fXsltn37dkOXtscVWwAAAACA1pjYAgAAAAC0xsQWAAAAAKA1JrYAAAAAAK3FTPKojIwMzzonTpxQYr1791ZiP/74o63cqVMnpU7Tpk2V2KOPPqrE8uQJNvf//vvvlZgzidXRo0cNncVLEoFIKl68uBL78ssvlVjNmjU9j69bcoPJkycrsdGjRxvxjgQqhnHw4EHluNx777233a/65dYXhnP7efPmNXRB3xdZzqSQzZs3V+osWrTI1+dStWpVW5nkUblj7HVz9913K7Hp06crsS5duni2q2vXrimxAgUKGPEuXvo+53g2f/58pU7Xrl2NeJKiUaIoNySPAgAAAADkCtyKDAAAAADQGhNbAAAAAIDWmNgCAAAAALQWM8mjnAvxs5NIJDk52VbevXu3Zx0xcOBAJVa9enUjXN544w1beeTIkYbO4iWJQDjVr1/fVl61apWvhFJu/8alS5fayn369FHqpKWlGbkRyaMM47777lOOyxdffGErlyxZ0tfxdGtHRYoU8WyjkyZNUmJPPPGEEqtRo4bnPiQmJhq6iJe+z5mk6dy5c0a0lShRQok1atTIVn7zzTeVOlWqVFFie/fuVWKNGze2lc+ePWvojL4vuM8++0yJdezYMdC2OnTooMSWL19uxLt46fucKleurMTWrl2rxCpWrGjo4uTJk55tNlYTRbkheRQAAAAAIFfgVmQAAAAAgNaY2AIAAAAAtBYza2x79+5tK8+bN8+IN/v377eVq1WrZugsXtda+OVcg+i2xqts2bK+tvXaa68psfHjx9vKGRkZt72P8Yp1Zu6SkpJs5bp16yp1ChUq5GudjdvaR6dNmzYpsTVr1iixZs2aeW6LNbbR7/uc/ZVzjbZYsmSJEtuzZ48Sa9Kkia1cpkwZpU6dOnWU2GOPPabE/PSbFy9e9LX9H374wYgn9H3BufVNzZs393zdoUOHlNhDDz2UK3Ne5KbzvnLlyikxt7wVbdu2tZV//etfK3U++eQTJTZr1iwlVrVq1bCspxXTp083vHJi6IQ1tgAAAACAXIFbkQEAAAAAWmNiCwAAAADQGhNbAAAAAIDWEo0Y8fHHH9vKvXr1Uuo0bNjQ0NnEiRNzehcQRs8884xn0hO3hE8LFixQYuPGjeOzQdgT1DnLwK0SQ3Xt2lU5QM8++2zYkr8ETTzjlrynUaNGSuz48eOBto/4U7lyZV9txk/b/d///pcrE0XldseOHfMV27Vrl63cunVrpc7cuXOVWKVKlQLtV0pKihLr0KGDr6SQuQFXbAEAAAAAWmNiCwAAAADQGhNbAAAAAIDWmNgCAAAAALQWM8mjrl27Ziu3aNFCqfPiiy8qsaFDhyqxokWLGtF0/fp1JbZixQoltnDhwijtEaLBT/KIU6dOKbGePXsaOqtbt66t3LRpU6XOpEmTorhHAIIYNWqUrbxy5UqlzrBhw5RYqVKllNi5c+dumUgvq3Fxx44dSuz8+fO28saNG5U6wK306NFDieXPn9/XQXMmOQua9Ay5U5cuXSKaKKp9+/ZKLLcminLDFVsAAAAAgNaY2AIAAAAAtMbEFgAAAACgNSa2AAAAAACtxUzyKKf09HQlNm7cOCU2d+5cJda/f39buXr16kqdBg0aKLHSpUsrsYsXL9rKq1atUupMnTpVibGQO/6dOHFCiSUkJNjKb731lhELypQpYyvny5dPqdOqVSsl1qFDByVWr149zwRuyL2cfwNusUuXLkVxj+BXcnKyryQ8zmSPbkkU3RL1uI3rQCS4JRF165vcEkM56/30009h3jvEk3LlytnKTz31VOBtpaam2sokirp9XLEFAAAAAGiNiS0AAAAAQGtMbAEAAAAAWovZNbbZeXjx6NGjPV/ntv4nT548nusvrly5ctv7iPi0f/9+JXb06FFb+cknn/S1lnvbtm1K7Mcff7SVH3/8cV9rYN22X6BAAc+2XrhwYSV24MABJdamTRtb+euvv1bqIPdyW7PmjL3zzjtR3CNkx+XLlwO9jvW0yEkPPPCAr77JjXPsdRtnkTsVKlRIifXq1cuzjl8LFy60lcnXc/u4YgsAAAAA0BoTWwAAAACA1pjYAgAAAAC0xsQWAAAAAKA17ZNHBUViC2TX4cOHlVi3bt1s5c8//1yp07Bhw8APjvfDbVtXr171fOD8yy+/rMTWrVvnmVgDuF0VKlTgoAGImGrVqgUeZ7/66itb+caNG2HeO+iqaNGiSmzMmDFh2/6aNWvCtq3ciiu2AAAAAACtMbEFAAAAAGiNiS0AAAAAQGtMbAEAAAAAWsu1yaOASNiyZYutXLt2baVO3759lViZMmWUWJ06dTyTNqWlpflKanXkyBFb+YMPPlDqANHSsWNHDjaAiHFLCuU3IWPQxI3A7UhJSVFihw4d4iBmE1dsAQAAAABaY2ILAAAAANAaE1sAAAAAgNaY2AIAAAAAtEbyKCCC3BI5DR8+nGOOXC05OTmndwFAHCtbtmxO7wLi0IULF5TYjBkzbOWBAwcqdU6dOqXEWrRoocT27duX7X3M7bhiCwAAAADQGhNbAAAAAIDWmNgCAAAAALSWYPp8EnVCQkLk9wZaidZDzGl7yKm2R/uDG/o+5BT6PuQk+j7Eetvjii0AAAAAQGtMbAEAAAAAWmNiCwAAAADQGhNbAAAAAIDWmNgCAAAAALTGxBYAAAAAoDUmtgAAAAAArTGxBQAAAABojYktAAAAAEBrCaZpmjm9EwAAAAAABMUVWwAAAACA1pjYAgAAAAC0xsQWAAAAAKA1JrYAAAAAAK0xsQUAAAAAaI2JLQAAAABAa0xsAQAAAABaY2ILAAAAANAaE1sAAAAAgKGz/wOKJjdw/S7zzwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1200x1200 with 30 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_wrong_images(network)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### A Convolutional Neural Network for MNIST"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "To feed images to a ConvNet, they must be of shape (height, width, channels), even if the images are grayscale."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "train_images = train_images.reshape((60000, 28, 28, 1))\n",
    "test_images = test_images.reshape((10000, 28, 28, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 28, 28, 1)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_images.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from keras.layers import Conv2D, MaxPooling2D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def build_convnet():\n",
    "    convnet = Sequential()\n",
    "    convnet.add(Input(shape=(28,28,1)))\n",
    "    convnet.add(Conv2D(64, (3,3), activation='relu', name='conv1'))\n",
    "    convnet.add(MaxPooling2D((2,2), name='pool1'))\n",
    "    convnet.add(Conv2D(64, (3,3), activation='relu', name='conv2'))\n",
    "    convnet.add(MaxPooling2D((2,2), name='pool2'))\n",
    "    convnet.add(Conv2D(64, (3,3), activation='relu', name='conv3'))\n",
    "    convnet.add(Flatten())\n",
    "    convnet.add(Dense(64, activation='relu', name='hidden'))\n",
    "    convnet.add(Dense(10, activation='softmax', name='output'))\n",
    "    \n",
    "    convnet.compile(loss='categorical_crossentropy',\n",
    "               optimizer='rmsprop',\n",
    "               metrics=['accuracy'])\n",
    "    return convnet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "convnet = build_convnet()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
      "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
      "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
      "│ conv1 (Conv2D)                  │ (None, 26, 26, 64)     │           640 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ pool1 (MaxPooling2D)            │ (None, 13, 13, 64)     │             0 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ conv2 (Conv2D)                  │ (None, 11, 11, 64)     │        36,928 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ pool2 (MaxPooling2D)            │ (None, 5, 5, 64)       │             0 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ conv3 (Conv2D)                  │ (None, 3, 3, 64)       │        36,928 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ flatten_2 (Flatten)             │ (None, 576)            │             0 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ hidden (Dense)                  │ (None, 64)             │        36,928 │\n",
      "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
      "│ output (Dense)                  │ (None, 10)             │           650 │\n",
      "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
      " Total params: 112,074 (437.79 KB)\n",
      " Trainable params: 112,074 (437.79 KB)\n",
      " Non-trainable params: 0 (0.00 B)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "convnet.summary(print_fn=print)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 11ms/step - accuracy: 0.9435 - loss: 0.1789\n",
      "Epoch 2/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 11ms/step - accuracy: 0.9844 - loss: 0.0516\n",
      "Epoch 3/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 11ms/step - accuracy: 0.9884 - loss: 0.0387\n",
      "Epoch 4/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 11ms/step - accuracy: 0.9907 - loss: 0.0312\n",
      "Epoch 5/5\n",
      "\u001b[1m938/938\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 11ms/step - accuracy: 0.9923 - loss: 0.0269\n"
     ]
    }
   ],
   "source": [
    "history = convnet.fit(train_images, train_targets, epochs=5, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 6ms/step - accuracy: 0.9948 - loss: 0.0178\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.017846228554844856, 0.9947666525840759]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "convnet.evaluate(train_images, train_targets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9911 - loss: 0.0349\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.03485921770334244, 0.991100013256073]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "convnet.evaluate(test_images, test_targets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "jupyter": {
     "source_hidden": true
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def plot_history(history):\n",
    "    loss_values = history.history['loss']\n",
    "    accuracy_values = history.history['accuracy']\n",
    "    epoch_nums = range(1, len(loss_values)+1)\n",
    "    plt.figure(figsize=(12,4)) # width, height in inches\n",
    "    plt.subplot(1, 2, 1)\n",
    "    plt.plot(epoch_nums, loss_values, 'r')\n",
    "    plt.title(\"Training loss\")\n",
    "    plt.xlabel(\"Epochs\")\n",
    "    plt.ylabel(\"Loss\")\n",
    "    plt.subplot(1, 2, 2)\n",
    "    plt.plot(epoch_nums, accuracy_values, 'b')\n",
    "    plt.title(\"Training accuracy\")\n",
    "    plt.xlabel(\"Epochs\")\n",
    "    plt.ylabel(\"Accuracy\")\n",
    "    plt.ylim(0, 1)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_history(history)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "A random sampling of misclassified images:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1ms/step\n",
      "Misclassified 89 test images out of 10000\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7YAAAOrCAYAAABjshrdAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAA06tJREFUeJzs3QeU1FTbwPEsvRfpSBOR3pEiUkSlqkhHRJoKCqKCigrCiwICIgqiSBWkWxBERUSKSlGQqrQVpPdepLd858l3Zs9kbnYnOzuzO3f2/zuH1zfP3p3JZp65yU1yn0SZpmkaAAAAAABoKkVSrwAAAAAAAAnBwBYAAAAAoDUGtgAAAAAArTGwBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0FlED24sXLxq5c+c2Zs2aldSrEjFq1KhhvP7660m9GmGP3As+cs+927dvG2XLljXefffdEHwSydMTTzxhtGnTJqlXI+yRe8FH7rnHvjf42PeSe1rnnhmAqVOnmp5f/eWXX6z/v3fv3pifFy5c2Io5/StWrFhMO/kdiclrCPn/8tqBGjJkiFm0aFHz5s2bZqQaO3ZsgraRx61bt8z33nvPLFKkiJk2bVqzXLly5uzZs5V28+bNMzNkyGAePXrUDAf+cs/jiy++MGvUqGGte9asWc377rvPXLZsWczPyb34S+655yb/ZJ0bNGhg5suXz0yTJo155513mi1btjS3bNmivJZ3fyd95sCBAwNer5kzZ5pZsmQxz507Z0aqWbNmmaNGjQrK5xfbP9mOHhs3bjRTpEhhbt682QwH5J7euSdiy7thw4bZ2oVb7rnd9x46dMhs3bq1tc/NnDmz2bRpU3P37t22Nux7k2bf69nuTv/mzJkT1vtexhz6H/ft2rXLOhbKli2bmT59evP+++83ly9frrQLRu6lMkJg9OjR1lk0b/v37zf69+9vNGjQIBRvady4ccP46KOPjN69exspU6Y0ItWnn35q5MyZ0+jcuXOCXuett94yhg8fbnTt2tWoWrWqsWDBAuPJJ580oqKirLPFHo8//riRJUsW630HDRpk6ODtt9+21rVVq1bWdpLc2Lp1q3H48OGQvB+5Fz+RnHtbtmwxsmfPbrz88svW9/TYsWPGlClTjGrVqhl//PGHUaFChZC87/vvv29tu6xZsxqRavbs2db3uFevXgG/Rp06dYwZM2Yo8VGjRhl//fWX8dBDD8XEKlWqZNx7773GBx98YEyfPt0Id+ReeOeeR/369Y2OHTvaYpJrvss65Z6QY7569eoZ58+fN/r162ekTp3a+l7VrVvX2Lx5s5EjR46gvyf73vhr166d0aRJE1vsvvvusy3rtu9lzBHeY46DBw9aOSZjsz59+hgZM2Y0pk6dao0Hly1bZu2Xg5p7obxq5m3w4MFWu9WrV4fkzJ2M8uX3//33XzMxXbx4MdafXbp0KejvV6ZMGbNu3boJeg05q5o6dWrzhRdeiIndvn3brF27tlmgQAHlinfPnj2tK0rSJqn5y70//vjDjIqKMj/88MM4X4fci7/knnuB9n3Hjh0zU6VKZT733HMhuWIrV3fktZYuXWompitXrlhX3+PbLwbqkUcesbZTsF2+fNm6ulS/fn3lZyNHjjQzZsxo/vfff2ZSI/f0zz35nnr3fXEJp9xzk39yF47E/vzzz5jYjh07zJQpU5p9+/aNibHvTZp9r2e7v//++67ah9O+lzGH3mOOHj16WMdA0dHRtnUtWLCgWbly5aDnXorEPON51113GTVr1gzJ63/77bdGkSJFjLvvvlv5WXR0tDVXKleuXEb69OmNEiVKWFeNvG3atMlo3LixdaYgU6ZM1pn7NWvW2Np8/vnn1lWl3377zejRo4c1n7dAgQLWzx544AFrjtuGDRussw8ZMmSwzlqKa9euGQMHDjSKFStmpE2b1ihYsKB1D7nEfc2cOdO6uiO/L1d+5LV+/vln62fy923bts16f1kP+Sfv67F7927rnz9yhUzOdMrf4CGv1b17d+PQoUPWlSXfM8xyxV3Ouupw5i5v3rzWFTM5jvC9cyAUyD1yLy7ST8j3+dy5cyHLvzRp0tjOenrIXQrPPPOMkT9/fqvvkT5YvufXr1+PabNnzx6jdevWxh133GGtp8xxWbhwoe11fv31V6uP+OKLL6w7b+68806r7YULF6wzudJnSt8jVwIyZ85stG/fPmb+pXwny5QpY6RLl87IkyeP8dxzzxlnz55V1nXRokXW1R35femH5Wq+7DeE9HOyTtIPefo+6Q89Dhw4YPXzgfj++++N//77L2adffu+S5cuGUuWLDF0RO6FZ+5duXLFuHr1apxtdMu9uXPnWttN/nmULFnSOpb66quvQvKe7Hvd73u9SV557wOc6HTc54Qxx8CwGXOsXLnSugtFxl4e8n5NmzY1Nm7caOzatSuouReSW5F9yaBxx44dymAymH7//XejcuXKSvzvv/82ateubd0W061bN+uDkg9CDmY8hVbkg5M2skOTD1/aTpgwwfoA5QOtXr267TVlQCiD5P/9739WB+Fx+vRpa3AstwQ+9dRT1o5UDuzkw1u1apX1/qVKlbJuGZNbdHbu3Gl1zB7vvPOOdRutDP7lErwcrK5du9ZYvny5dcleDhBffPFF6yDSsy3lPTw8t9Ht27fP7+chtwLIuniT5Pb8vFatWjHxKlWqWP9dvXq1cstUuJHbGmT7jRkzxhgyZIj1mchAV7ZXz549Q/Ke5B6550sGsXLySG5Flu+tDAC9b3MNdv7JSTXpt7wdOXLE+k7LukjfIweZMtCVA9DLly9b/cvx48et74ssv/TSS9btgtOmTbP6LGnXvHlz22sOHjzY+r3XXnvN2knK/xc3b940GjZsaPUbI0eOtHZaQgYSckKwS5cu1uvv3bvX+OSTT6w+RvoTzzpLm6efftoahPTt29fIli2b1eann36yblOX76/c4ign3qTvFNIPesitndJX//9FsfiRYoNywrNFixbKz0qXLm39TNbVd1uEK3IvvHNP3k9us5P2sg+WE0XyPjrnnhznyLGWbEdf0gfJgbKcPJITB8HEvtf9vtf7OFNuB5VBihzbyXGw0xRBnY77fDHmaBpWYw45VpBBsy/PcYJcELznnnuCl3tmInj11VetWyC2b98ekte/ceOGdfupvI+vOnXqWLeZ7d+/3xb3vsTdrFkzq9CLd5GDI0eOWL8nv+97O0StWrWUWyblUr38bPz48bb4jBkzrCIQK1eutMWlnfet2TKxWto1b95cucXKe13jui1ALt27uV1KbquSIlu+5NYAWac333xT+Zlsn+7du5vh7MyZM9b658iRw8yUKZN1y82XX35pNmrUyPGzCQZy7/8l99zzVaJEiZjCHJKL/fv3j/XWyYSSW7ilKIOvjh07Wn3KunXrlJ95+pRevXpZ6+jdP8mtj3fddZdV3Muzzp5bD+Wzk1t3vXXq1Mnxs5PXlLgU3vH2008/2eJS8Er62urVq1u3mDqtp7/bQT39b3ydPn3ayq82bdrE2qZ48eJm48aNTV2Qe+GbezVr1jRHjx5tLliwwBw3bpxZtmxZ63c//fRTrXPv5MmT1t8xaNAgx+Iz8jPv2xCDgX1v/Pa9cgwshQ0l77777jsrDwsVKmTtI3744QfH39Fx3ysYc6QIqzHHY489ZhWNunDhgi0uRV1lnWTaRTBzL0VinMmT29dk1O17hTBYzpw5Y5399D0jcPLkSWPFihXWWcRChQrZfiZnq8StW7ess4nNmjUzihYtGvPzfPnyWWdR5ayHXG3xJkVvnApUySV/OTvs7euvv7b+brlacurUqZh/Dz74oPXzX375xfqvnEWRbSVXgVOkSOG4rv7IWRM3Z+3kNihZV19yu5bn575k28p6hzPPbcdylXby5MnWVSW5BV1uI5Oz33IFN9jIvf+X3HPPlxRGkCs+cmVGvv/yd0lfEwqS7759n/Ql0qc89thjVhEaX54+5ccff7SuqHjfoSFnZ+VMr3ye27dvt/1ep06drKtITuQWZ9++T4pZyW1F3n2fnI2V9/D0fXKrpVzNefPNN2PywHc9/ZFbpQO5WitXpeWWPKfbkHXNP3IvfHNPrkDINBm5I+L555+3rlTI3RYybUnnvs+z7vHt2xOCfW/89r1yDLx48WIr72S/IHkoVzbl7sNXX33V8Xd0yT9vjDm+DrsxhxwbyJ1Ebdu2tXJOrhxLIb7169dbPw923xfyW5HlFh25/U2qFYea785F5o4J2XHERga/chue973fHpIc8sFLRS+5TclD5qk5kXlnnlvzPOTecbkNWzoPJydOnLD+K7dHS3LJACzU5MDU6V57z5wfpwNX2bZukz2peNZbbjGTisgesl3lCyXznGU+lO9JjmAg95J37vnyrjIpUxM8J/XkNt1Q8M0/6dfkhFxcfZ+QeSy+Uy2EZ33l596vEVvflypVqph6A959n9zCKfM8/fV9wt+6hoLchixzi2UKSWx0yz9yT4/cE3K8IFNkPINc7xNMOuWep9+Ob98eDOx7Ayd9n1yMkacUyK32vn24LvnnjTHHrrAbc8j+9eOPP7ZOIHqmjMr8X7kNXqZ/ek/tCEbupUqMAwfZeFJiPJRfTtkATkUhQiG2DtopLgPjcuXKGR9++KHj78ik7sQmV6PlrI1v4hw9etT6rxSa8SVnW6TkdziTPJCzwzJHyveKuucAR3IkmANbci9+IjX34iJnHuVsqfSFoRjYyrzYpO775EqN71lf6fvkeyd/t5PYdryJRU5ySVELuTrtOz/Zm2xb7/k/OiH3wjP3nI4B5Aqkrrkn+0HpAzz9uLe4+vaEvifHfcHNP9+BrY77XsYct8NuzCHkBJ6cRJG5+HJCr2LFisZnn31m/ax48eJK+4TkXkgHtnL27ptvvrGKMAW7U/O9WiDVkKU4hDfPrcXy/LnYyA5OJjD/888/ys+k0qEcrCUkEWS9PM9HjOvsg7STA0G59U8+8NgE4+yZvL7cqitndbzP1sikcc/PvckVd7ldL1S3kgeLfFay7uvWrbPW1/vquRTSCcUBDbkXP5Gae/7IrTZyBSkU5JYj375P8lyK4cXV94nChQvH2vd5fh4o6dOWLl1q3H///XFerfFUspd1lbO4sQn2lYM5c+ZYJ1jiug1ZimLJHTty66iuyL3wyz2nO8t890065Z7se+Vg2nNroTfp2+VYLNiFo9j3hjb/dNz3MuYIzzGHhxSt9b6jSI4P5NhAjhGCmXshnWMr87dk1B3XgUOwyMby7VTliyqlq6dMmWKdnXe6fUWu7En1L3kEjve94lItVMqFy61BcoAYKJnjKR/SpEmTHA84PFWVZY6v7BykMpkkm9O6ehIjtseGuC29LQ9AlisUMv/P+z3Gjx9v3U7t+0gmuUVKhOpRTcEktxzLXEap7Op9K5ScxZOBVChOsJB75J7vbT7epF+Rat1Oc12DlX9yYO59G6D0JdKnSPV3p4NNT58ij+f5888/bY/4kj5p4sSJVgX5hNymJH2ffBelkrIvOWj39GPS/8pB77Bhw5RHoPj2fbGdHAjkkSvSv8vdG763f3qTnb6skw59H7kX3rkn0wN8yfxeqTwqVyY8lUB1zD0h03/kpLJ3fyMnzaTCqjxOLBTY97rf9zrlnxybyvFx+fLlrbupdD3u82DMEZ5jjtgqms+bN896HKHUQwhq7pkhJJU606ZNa1UeDLW5c+da1bX++ecfW3zz5s1WVVKplCsPCZ84caLZr18/s0KFCjFttm7daj0I/c477zTfffdd60HjUv1T1n3NmjVKVWSnKqNSNUyqh/mSamNNmjSxqjY/8cQT5scff2xVo3v++efNO+64w/ZaAwYMsF5fKidKlTBpK5VNvauNyoOO5bUGDx5szpkzx1y2bFm8K5SJPn36WO/VrVs3c9KkSVbVR6cqkp6HJUv1vHB4ULc/UrFVPofUqVObr732mjlmzBizatWq1kPif/zxx5C8J7lH7nnkzp3bbNeundWHSF8j3zP5nqdLly6mGmGwrV+/3vruLl682BY/dOiQmTdvXjNDhgxW9eMJEyaYb7/9tvX9OHv2rNXm2LFjZp48ecysWbNa/c+oUaPMihUrWn3MvHnzYl7LUxX566+/Vt5fqiJL/+nkueees35PKrvKa3/yySfmyy+/bObPn9/2WpMnT7baSZXYoUOHWpU7pY+U/s9jxIgRVpvevXubs2fPtip7BloVecuWLbFW4fYm/bBsP99qjuGI3Avv3Bs4cKB13CEV0qVveOedd6z9tXzXZs6cqXXuCVnPu+++28pD2V6yzQsWLGht7xMnToTkPdn3ut/3du7c2axdu7a1D/AcB8txsVSflf5d5+M+D8Yc4Tnm2Ldvn1mtWjVzyJAhVn8r/Wj69OnNSpUqOfZvCc29kA1sz58/bx3MtWjRwkwM165dM3PmzGltfF8ycJWS1lJuWtZJHocgH6i3jRs3mg0bNrQGwbIzqVevnvn777/b2gQysBXXr1+3DnTl5zJYzp49u1mlShVrxybbyduUKVOsD9vTTl53yZIlMT+XA1EZhMojCmRdvMtwx2dgK8kvO3FpLx2brJvTzlXa5cuXz9oZ6+L48ePWwbZ8iWU7yqMc5DEPoULukXveB6/33nuv9d1NlSqVdVAnO5e///7bDKXy5cubzzzzjOMjHmRHlStXLuu7ICfsXnjhBStnPeQxZ61atYrpH2UH5Pv4h0AHtkIOoqS/kx2Z9FvlypUzX3/9deuRat5ksCA7WGmXJUsWaz1kR+px8eJF88knn7TWU9bFu6+L78BWdtzS3t/nIn3HU089ZeqA3Avv3Pv555/N+vXrWyeb5MSrvJY8fsX7QFHX3PM4ePCg1ZfINpRjqUcffdR6rEiosO91v++VEzLy+ErZF8i+SY6X5bh4w4YNEXHcx5gjfMccZ86cMR9//HGr75PxhjxO8I033nAc1AYj9xLlObaJRZ6hJhvM9xmzCNz8+fOtnb3vgQDIvVAj99ybPn26tePxXIlFwm3atMk6Uy3/BbmXmMg99zjuCz72veSezrkXJf9jRAh5jqkUKRg1alSizOtNDmQOS+3atY0RI0Yk9aqENXIv+Mg992SOjMyTkurzb731Vgg+jeRHHtMk2/Wrr75K6lUJa+Re8JF77rHvDT72veSezrkXUQNbAAAAAEDyE9KqyAAAAAAAhBoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKC1VG4bRkVFhXZNoJ3EqjtG7iGpco/8gxP6PiQV+j4kJfo+hHvuccUWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWkuV1CsAILjat2+vxLp27Wpb7tGjh9Jm+/btrl4/Z86ctuXDhw8rberVq6fEfv/9d1evDwAAAMQXV2wBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0RvEoQBOFCxdWYq+99poScyoMFRUVZVsuU6ZMwMWjfKVOnVqJ3X///UqM4lEAgEiQIoV6XahatWqJWhgyffr0SpszZ84osW+++UaJrV27NshrB4QHrtgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABojeJRQJgqUqSIbfmnn35S2hQvXtzVa23bts22PH/+/ASuHQDET5UqVZRYs2bNlFjp0qX9tvEtiCd27NihxHLlymVbnjdvntJm9uzZSmzFihVKDPBo06aNqzwKB/Xr11dilSpVSpJ1AUKNK7YAAAAAAK0xsAUAAAAAaI2BLQAAAABAaxE5xzZjxoy25QIFCihtunbtqsS6d++uxNavXx/nfB1RqlQpJbZ9+3YltmjRItvyiBEjlDYnTpxQYoh8jRs3VmJffPGFbTlz5syuXmvhwoVK7N1337Ut37x5M97riOStZMmSSmzTpk1KbMuWLUqsRo0atuXbt28Hee2Q1Jo3b25b7tevn9KmcuXKSsw0Tb/zZ53anDp1SomdPHnSb94+++yzSpuGDRsqsZUrVyqxjh07KjEkT2XKlAno986ePavEMmTIoMT++ecfJXb69Gnb8o8//qi0OXLkiBL7+eefA1hTQE9csQUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANCa9sWjUqdOrcTGjx9vW86XL5/S5sEHH3T1+nXq1PFbxMIp5lRQyjf2wgsvKG0efvhhJbZ69WpX6wo9tGrVSokNGjRIibkpFuWb66J3795K7Nq1a/FaR8C33/zyyy+VjXL9+nUl5lQUj2JRkWXGjBlKrH379vEuChWby5cv25ajo6OVNt9++60S++abb/zuw+fNm6e0KV26tBL74IMPlFi3bt1syxMnTlTaIPI888wzSqxPnz5K7NChQ0qsWbNmtuWjR48qbdKkSeOqEJrv9wKRr169eq6OF++//36/r3XlyhUl9sorryixCRMmGDrjii0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWokynCg9ODV0WfUhsefPmVWJ//PGHbTlTpkxKG6cJ/KdPn1ZiP//8s205d+7crgpPOBWsKlu2rN9t+t133ymxxx9/3AhHLlMnwcI199woWrSoElu1apWrPPY1btw4Jfbqq68qsatXrxqhlDNnTtvy3r17lTYFChRQYufPn9cu93TPP7fq1q2rxN58803bcsOGDZU2rVu3dlXAJ9Ik975v3bp1Sqxy5cp+t5FT37djxw4l9tFHH/ktHhVqvv2ck1OnThmJjb4v9Dp16mRbnjJliqvv5pkzZ5TY8uXL/RYlW7t2raGL5N73hdqoUaNsyz169FDapEqVytX2Ml18VkeOHFFiBQsWNHTOPa7YAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaE2dgawZp4JPX331ld/iTocPH1ZiAwcOVGLHjx8PaL2yZ8+uxBYsWGBbrl27ttKmfv36SqxGjRpKbM2aNQGtFxJXt27dAioU5VQs6rXXXkv0QlFOfIug3b59O6SFouAsW7ZsSuzcuXN+C0q89NJLrvq+MWPG2JZLlCihtMmfPz8fTzLUoUMHJeZbWHH79u0BF1tq1KhRnMsiR44cSmzYsGFK7PLly0YgkqIwFBJfq1atlNjgwYMDKmR0xx13+H39pk2bKm26dOmixObMmePqPaGHKlWqKLHPPvtMiZUqVcq2fODAAVfHkBkzZvS7DmfPnlViTsU/s2bNqsQuXrxoW75165YRrrhiCwAAAADQGgNbAAAAAIDWGNgCAAAAALSm/Rzb/v37K7G7777b7wPg169fr8QuXboUtPVyupf96NGjfn8vXbp0SuyNN95QYs2bN0/A2iFU+vbta1vu3bt3QPNpnebUXrlyxQgHbdu2TepVSHZy5sypxEaOHKnEZs6caVseMGCA0uauu+5yNWdy4cKFtuWHHnrI9foiskVHR7uKuTFjxgwl9uSTT9qWTdN0VV9j0qRJSsxpjhqSpzp16iixadOmKbH06dP7fa2lS5e6Ou7zrU1Qvnx5pc3UqVOV2O7du5XYn3/+6Xe9kPh8a+r06dNHafPss8+62q+fOXPGtrx69WpX+bl8+XIltmfPHr91D3r27KnEnNr5zvl+9913XeV/UuCKLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNZS6fQw43bt2imxli1bKrFHHnnE70TopPDrr7/6fciyU3EDNw9eRuJr3bq1Env77bdty6lTpw6oUFS4FIvKnTu33+/crl27EnGNkqcXXnhBiT311FNKrFOnTrblH374QWnz4IMPKrF///3X7zqsWLHCxZoC/69kyZLKppg3b57f4joiKirKtjx//nylzaxZs5QYhaIQV7Eep31vypQpldiFCxdsy+PHj1favPXWW0rs1q1bSixz5sy25ZUrVyptnApK+RZBFRSPSnq+x3hOxaLcFB8TW7duVWKDBg2yLc+dO9cIlrRp07rKs/z58yuxV155xbY8ffp0pQ3FowAAAAAACAJuRQYAAAAAaI2BLQAAAABAawxsAQAAAABaC9viUU6TkJ0K8dx1112GLr755hu/k7Zr166txEzTDOl6ITB9+/b1m6PHjx9X2gwfPjwsC0U56dGjh9+/sWnTpom4RsmTU/GcixcvKrGFCxf6LezlVOAkUClScG4UzsWi1q1bp2yaDBkyuNq/DR061LY8bNgwpc3ly5fZ9IhRs2ZN29YYO3ZsQIWixKOPPmpbXrVqVcBb+r///rMt79y501XxKKfilHPmzAl4PRC3Bx54QIm1b99eid1xxx1+i39ev35daZMtWzYl9umnn4b0WDC9TxGrfv36uSrA69QnHz582LZ89OhRI1xxVAIAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKC1sC0eVbRoUSVWqlQpQ2cdO3a0LX/99ddKm1deeSUR1whu5c+f31WOnjhxwu/E/IMHD4blhk+XLp0Sq1KlihK7efOmbfnIkSMhXS8Yxt9//+0qFkpr1671WzRDjBkzRolRAC/yNW7c2G+hqKioKCUWHR2txAYMGBDktUOk8+2LnIqNnjt3Tok1atRIif35559BXjuEm0KFCtmWCxQooLT54YcflNiCBQsMXdStW9e23L17d1e/d+3aNSXmeyx78uRJI1xxxRYAAAAAoDUGtgAAAAAArTGwBQAAAABoLWzn2Do9pPjYsWNGJOnRo4erdtu2bQv5uiBuTnMtsmTJosTmzp1rW/7999/DctP6PrhbvPfee0rskUceUWIffvhhyNYL4evXX39VYpMnT1ZiWbNmdTW3DZFl+/btAc2rLlGihBJbtGiRbXn+/PlKm4kTJ8Z7HRG5duzYYVvOnDmz0qZr165KbN++fSFdL9/1ePjhh1393rfffhuiNYJTTRGnubPt2rVTYt26dQtaX1SrVi0ldubMmTj71dhyaOnSpUrswQcftC1nz57d1Xp9/vnnWs8754otAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1sK2eNTq1atdxXR26dIlV+18C2kgtHwfRC3KlSvnd5J/bJPuw1H16tWVWM+ePZVYVFSUEvv3339Dtl4IXxcuXHD1IPd69eopMafiP4gsixcvti23bt1aaZMrVy4l1rx5cyVWpUoV23KDBg2UNr169VJiHTp0UGIbNmyIY62RlHwLK1WoUMFVf7J+/Xq/x0nvvPOO0ub69etGYsufP79tOVu2bK5+7+TJkyFaIzgV0HQ6/lmxYkXQxiFFixZ11e7q1atx9oWiSZMmSuzs2bNK7KGHHrItb9y4UWmTN29eJda9e3dDZ1yxBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjYAsAAAAA0FrYFo/q2LGjqwI+ffr0McJRzpw5lViWLFlsy+vWrVPaOBXJ2LVrV5DXDnEpVqyYEkuZMqUSa9GihRJbtWpV2BVFEH379rUtd+nSxdVrORXIWrJkSQLWDpHk66+/VmL9+/dXYgsWLLAt3759O6TrhaTntmDYxIkTlVihQoX87k9nzJihxNauXavExowZY1seOnSo0ubUqVOu1hXBNXfuXNty/fr1g/baf/31lxIbNGiQEvvxxx9dFcVzo0CBAkrs+++/9/t7ToV/9u7dG9A6wJ3t27f7LdbVtm1bJZY7d24l9sMPP/gtRLVjxw4lduDAASW2Z88eI77rLq5cuaLEli1bZlt+9dVXlTbNmjUzIg1XbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWGNgCAAAAALQWtsWjfAstiTlz5hi68J207VT86r///lPaVKpUSYnt27cvyGuHYNi2bVuSb8gaNWoosYEDByqxhg0b+n2tmzdvKrFPP/1Uie3evTte64jwV7FiRSXWvHlz23L27NmVNnfeeaerPmz//v1+iwP+8ssvrtcXkc23qIpTkZUyZcr4zVkxYcIE2/KlS5f8FpiKrZgMgsu3WKZpmkqbLVu2uCrw6Fs0sUKFCkqbb775RomtWLHCbx45FXdKkUK9LjRgwABX6+qmEGV0dLTf30PgfPdnX375pdKmSJEiSuzDDz9UYr55u3DhQqXN0aNHlZjTGMANp0JRadOmVWJNmza1La9cuVJp8+uvvxqRhiu2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGthO8f2+PHjruY0hIMqVaoosbvuusvv7w0ePDighzMjtEL9wGrfuUDijjvuUGL33HOPbblDhw5Km8aNGyuxvHnz+l0HpzlDO3fudDVnCHp76qmnlNjUqVOVWMqUKf2+1tKlS13N1fZ9qP2uXbtcrCkQP/Pnz1divvNznea/NWrUSIn17t1bia1atYqPJJENHz5cif32229+60i0bdvW75xeUadOHb/zt9euXau0ef7555VY165dDX+2bt2qxDZv3uz39xBcNWvWtC0fPnxYadO9e3dXn5+bY6lgKlq0qBL76quvlFjOnDn91rYIdJ5vOAvPkSIAAAAAAC4xsAUAAAAAaI2BLQAAAABAawxsAQAAAABaizKdnojt1DAqKqQr4vsg5FKlSiltFi1aZCQ1p4fCOxUbeuKJJ/w+CPyFF14wdOYydRIs1Lnn69q1a0osderUSuzzzz939btuJv7Xr1/fCCXfYj1NmjRR2uzevdvQRWLlXlLkX6i5eZC7kyVLliix8+fPK7GBAwcqsTfeeMO2nClTJqXNrVu3DF1Eat+XHJQsWdLvvlmcPHlSiT3wwAN+24RaJPV9ixcv9rsfnD59uhLr3Lmz39cuVKiQEtu3b5+r9Xrrrbf8FgN99tlnXb3Wtm3b/BawOnr0qKEL+r7E5Ts2EiNGjFBiDz74oN8CpGcdiobqxG3uccUWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAa2FTPCpcPfLII7blOXPmKG1u3rypxJYuXarEnn76advyxYsXDZ0l9+JR4cCp4I5voSjx6KOP2pb37Nlj6CySCqgkBzdu3LAtt2rVSmmzYMECQxeR2vclV0899ZQSGzlypBLbtGmTbblx48ZGYoukvq969eq25d9++01pkyKFev3l448/VmKffvqpbTlfvnyuioQF6sKFC0ps7NixSmzo0KG25UuXLhk6o+9L3AK1Trlet25dJfbee+8psX79+hmRhOJRAAAAAIBkgVuRAQAAAABaY2ALAAAAANAaA1sAAAAAgNYoHuXH+vXrbcsVKlRQ2pw+fVqJvfHGG0ps2rRpRiSJ1CIC3333nd/iS0lh586dSuydd95RYk4FziJNJBVQSQ4mTJhgW65Tp47SplSpUoYuIrXvc1K4cGFX7fbv329EkkWLFimxRo0a2ZZ79+6ttBk9enRI1yuS+z6ngl2vvPKKkdTWrVunxF5++WUltmbNGiPSJae+L9SyZs3q9/itVq1aSpu//vpLifXq1UuJbdiwwYgkFI8CAAAAACQL3IoMAAAAANAaA1sAAAAAgNZSGWGiaNGituWjR48qba5cuZKoD0YWVapU8Xtve8WKFUO6Xkhcq1evDukc261btyqxr776yu/vOc3RPnjwYNDWCwiVgQMH2pZ37dqltClbtqyr7woSV/PmzZXYm2++GZZ90bx585RYdHS039/LlSuXEsuZM6cSu337tm25RIkS8V5HxK5Pnz5KbNmyZUqsX79+Sqx8+fK25XTp0rnKhfPnzyuxoUOH2paXLl2qtLl586YSA+JjzJgxSixlypS25ddff11pM378eDZ0HLhiCwAAAADQGgNbAAAAAIDWGNgCAAAAALTGwBYAAAAAoLUo0+UTb0P9sORhw4bZlidMmODq9/bt26fEihQpYlvOkSOH0qZ27dpKbPjw4X4LVjVu3DhZPpTbCQ/qRqTnXnJ5UHyopUplr1P422+/KW0uXryoxBo2bGiEI/o+Vbdu3fxuN6f9bsmSJf0WbXTa3k7fSzftEvJaKVKk8FvEpXv37kYo0fc5K168uG05Q4YMSpvNmzeH6FNJPuj7/EubNq2roqSVK1dWYj/99JNtuUmTJvH8hCKX29zjii0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWwqZ4lK/HHntMiTkVlDp16pQS8/2TfAs+iDJlyiixbdu2KbGRI0falqdNmxbHWicvFBFApOeeoHhU8OXJk0eJ/fnnn0qsVq1aSuzgwYNGUqPvC6327dvblnfs2KG06dq1a9Der1SpUkrM6T19jzcmTZqktDlw4IARSvR9SEr0far06dPblpcvX660ueeee5TY6NGjldioUaNsy5cuXQroc4pEFI8CAAAAACQL3IoMAAAAANAaA1sAAAAAgNYY2AIAAAAAtJbKCFPZsmVTYnnz5lVi+fLlU2InTpywLW/fvl1pU65cOSW2f/9+JXbx4kVX6wsAcOf48eNK7LvvvlNiFM5InmbNmuW3Tffu3RNlXQDAI23atMrG+OSTT/wWp+3QoYMSW7BgARs2BLhiCwAAAADQGgNbAAAAAIDWGNgCAAAAALQWZbp84m1UVFTo1wZa4UHdiPTcE/R9SKr8I/eQVLlH/sFJcu/7GjRooMRq1qxpW/7jjz+UNosXLw7peiUHpsvc44otAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1igehYAl9yICSDoUUEFSou9DpOeeYN+LpMo/cg++KB4FAAAAAEgWuBUZAAAAAKA1BrYAAAAAAK0xsAUAAAAAJI/iUQAAAAAAhCOu2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWGNgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWGNgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABoLaIGtrdv3zbKli1rvPvuu0m9KhHjiSeeMNq0aZPUqxH2Ll68aOTOnduYNWtWUq9KxKhRo4bx+uuvJ/VqaIH8Cz7yj9xLKux33aPvI/+SCrkXpvtdMwBTp041Pb/6yy+/WP9/7969MT+fN2+e2aBBAzNfvnxmmjRpzDvvvNNs2bKluWXLFuW15Hfl9UThwoXNgQMHmoGaOXOmmSVLFvPcuXNmpJo1a5Y5atSoBL+ObHenf8OGDbO127hxo5kiRQpz8+bNZjjwl3uSP05/V9q0aW2vI78jcXkN3zwMxJAhQ8yiRYuaN2/eNCPV2LFjE7SNPHbt2mX1B9myZTPTp09v3n///eby5cuVdtKPZMiQwTx69KgZLvzln6+HH37YavPCCy/Y4uRf/CX3/HOTe3PmzDErVapk9Xc5c+Y0n376afPkyZO2NuRe0uWe7Ccee+wxM3fu3NbnF9vxTrjtdwX5l3SSe/653e9+8cUXZo0aNax+O2vWrOZ9991nLlu2LObn9H1Jk3ue7e70T/ZZwd7vpjJCYMuWLUb27NmNl19+2ciZM6dx7NgxY8qUKUa1atWMP/74w6hQoUIo3tZ4//33rTOdWbNmNSLV7Nmzja1btxq9evVK8GvVr1/f6Nixoy1WqVIlZfnee+81PvjgA2P69OmGLsaNG2dkypQpZjllypQhe68bN24YH330kdG7d++Qvk9S+/TTT63vc+fOnQN+jYMHDxr33XeftZ369OljZMyY0Zg6darRoEEDY9myZUadOnVi2j7++ONGlixZrPcdNGiQoZt58+ZZ/V2okX/uRXL+SZ/Xo0cP46GHHjI+/PBD49ChQ1a/tH79emPt2rVGunTpgv6e5F789O/f38ibN6+1X128eHGs7XTc75J/4b3vjfT8e/vtt61+ulWrVtZ2kr5JjpUPHz4ckvej74u/du3aGU2aNLHFZH/sLRj73ZAMbP/3v/8psWeffdYoUKCA1fmNHz8+6O+5adMm46+//rK+iInp6tWrRpo0aYwUKdS7ui9dumQdOIWr4sWLG0899ZTfdnIr8sCBA61E8x4shjPp3GRHkBh++OEH4+TJk4l+y3Zc+XX58mUjQ4YMRrgZPny4ce7cOWuHU6JECSvWtWtXo2TJktaJgQ0bNsS0le+UfI6yY33nnXeMqKgoQxfSL7z66qvGG2+84dgfBhP5516k5t/169eNfv36WQPzJUuWxKxrzZo1jccee8yYNGmS8eKLLwb9fcm9+Nm7d69RpEgR49SpU0auXLkiZr9L/oX/vjeS82/NmjXWIEiO/6UfTwz0ffFXuXJlv2OOYOx3E22Orcw/lC+7HFSEwrfffmsNML3PuHvIGZtnnnnGyJ8/v5E2bVrjrrvuMrp37251xh579uwxWrdubdxxxx3Wesp93gsXLrS9zq+//mpt5C+++MI683XnnXdabS9cuGCdIZIv/+7du60zEpkzZzbat28fM/d39OjRRpkyZayz5nny5DGee+454+zZs8q6Llq0yKhbt671+3LWomrVqtZVWvHAAw9Y67R//35rPeSfdFIeBw4cMKKjo+O13a5cuWIdhPu7siuDKDlg0oXcXSyfy//fZRxaknvyOdx9993Kz+TzkB2E7ETSp09vHUy/9dZbykmZxo0bW5+35JBccZGO2tvnn39ufd6//fabdVVGvk9yosiTFzK3XA7KJf8lJ+UgV1y7ds3aORUrVszK/YIFC1rzFyTua+bMmdZdFfL7cseFvNbPP/9s/Uz+vm3btlnv78k9eV8PyXv558/KlSuts8GeQYWQ92vatKmxceNGY9euXUruSb5v3rzZ0MmIESOs7/1rr70W8vci/8g/GajLvrVt27a2A4FHH33U6lNkn0XuJW3f53ktt3Ta75J/4b/v9bxWJOafHF/LlWi5S1SO+WTua6ix3zXilXseklPeYy8nCT3uC8kVWw/Z0crlerkVWRJPBhpy0B4Kv//+u3Vwnzp1alv8yJEjVoch69KtWzfrzLwMdOfOnWudWZPB8PHjx60z27L80ksvGTly5DCmTZtmHWxLu+bNm9tec/DgwdbvyUGrdFLy/8XNmzeNhg0bGrVq1TJGjhwZc9ZOBrEyMOnSpYv1+nLW7JNPPrEGNKtXr45ZZ2nz9NNPWwPgvn37GtmyZbPa/PTTT8aTTz5pDYjOnz9v3WI2atQo63e8z6TJbcXS+bkdzMn7ydk4aV+qVClrsC7v46t06dLWoEzW1XdbhKuiRYtanZtc0WzWrJl1Jk9OKIQq9+RMlK+///7bqF27tvX5Su7JTkU6ge+//z6mwJnssKSNDGplpydtJ0yYYO245LOsXr267TVlUCuDZLkKKB2Ex+nTp63BsdyKL2fE5G+VgZXk8KpVq6z3l89YpglI7uzcudPqmD3kzJjcyiPfAznzKTktty8uX77cuk1Tvr9yxUfyzTMw996enu/1vn374txW8n2RHbcvz3dFBuf33HNPTLxKlSrWfyX3fG+TD1dygkmuDMr0C/nehBr5R/55Dpad8k1ish+R/sDpziJyL3H6vvjSab9L/oX/vjeS80+mkcj2GzNmjDFkyBDreEgGurK9evbsGZL3ZL9rxDv3JNdlCpCcnJFjOzkOlhz3leDjPjOESpQoETNBOFOmTGb//v3NW7duheS9ChQoYBUE8dWxY0drEvy6deuUn92+fdv6b69evax1XLlyZczP/vvvP/Ouu+4yixQpErPOnknrUiTo8uXLttfq1KmT9bM333zTFpfXlLgUffL2008/2eJS8Cpz5sxm9erVzStXrjiup3jkkUesIltO6tatGzPB3p+aNWuao0ePNhcsWGCOGzfOLFu2rPW7n376qWP74sWLm40bNzbDnfxNPXv2tLbr3LlzzZdfftlMlSqVec8995jnz58P+vvduHHDjIqKMl999VXlZ3Xq1LE+0/3798f6eTZr1swqsLZ79+6Y2JEjR6zfk9/3LZ5Qq1YtpUCV53MfP368LT5jxgwr973zWkg7ab969eqYYjrSrnnz5sr303tdy5QpY72XE8nJ2PLSmxSukKI9Fy5csMWlyIOs08iRI5Xfke3TvXt3UxetWrWyvl8eTsWjgoX8+3/JPf+kQJT0Q88884wtHh0dHbMPPnXqVFDfk9yLX+75fl5xFe/Rbb9L/oX/vjdS8+/MmTPW35IjRw5rnPH++++bX375pdmoUSPH46JgoO+LX+7JMbAUFJaxxnfffWcdpxcqVMjK/R9++MHxdxKy3w3pwPb333+3BnAyWKpatap18H/9+vWQvJdUt3z22WdtMekopEry448/7vfLW61aNSUuFYLli+Gp5uwZ2L7zzjtKW8/A1ncQ89JLL1nV2U6cOGF1Jt7/5EvoWeevv/7a+v358+fHua5xDWwT4tq1a9bgVg76fAftQgbc8hnqSAa5ThWfg+H48ePWa0u1QW/yeUtcBtaxkQGqVH9r06aN8rPnnnvO+tJ7BuOege20adOUtrLDkyqo8hl6a9q0qbVD9M27nTt32tZZdgSyvGnTpjj/1rh2rm79+OOP1nvJzlIqL/7zzz/WNkqdOrUVHzx4sPI7efLkMVu3bm3qQKrrygDjzz//TJSBLfkXP5Gcf23btrVO4sngXE6UrVixwqxQoULM33bw4MGgvh+5Fzi3Awud9rvkX3jveyM1/w4cOBBz8k6qInsf/5cuXdq66BVs9H0Jd/r0aWvfKhdAnSRkvxvSObZS7UpuzZX5rFKBTeYRyC22oeJ7C64U9JHbn+UW5bjIvdzec/485PYRz8+9yRxdJ6lSpYqZ9+ghcwbl9mGZEym3kHr/k1tlT5w4YbXz3Kfub11DRW5/kVs25JZt7wIq3ts2nIunxEVur5bbUpYuXZpouSdztv19npKfcvt7bLkntzNJFVc3uSfzvT23xHvnntzq7Jt3UjRMeOee3KIotx6Fmtwu/fHHHxsrVqywbt+Wv13mjXtuzXYqUqFL7slUBJlq0KFDB2tufGIi/9yJ5PyTKQxS30GmyMh8f5mnV65cOat4lAhVARhyL3R0yT1B/oX3vjdS888z/UKmcUnRIQ/ZrlJzQKbuyfSgUKDvC5zUM5Lpmf/884/1GQUz90I6x9abzKt78MEHjVmzZlnzT4NN5sU6FWMKhdjmzUmBAN85TDI4kUGt/N1O/FWmS0xS3ECcOXNG+ZlsW++5j7qRv83p7wrGl1O+fEmde05xyT05sJVHf8T1eSc2OYEiHZrMQZbBeMWKFY3PPvvM+plnx+9NTrYkVoXrhJAqftJJywGe75yT//77z4p5iugFC/kXf5Gaf/KYuwULFlgHcZJrhQsXtv7J3DPZz0jNhmAi90JPp/0u+Rf++95IzD/ph6Qoq/Rvvo9blP2t5+8oVKhQUN+T477gjjl8LwomZL+baANbTwVeuXoZClIUSooyeZOduRTlkYp9cZGdvxyQ+vJUGJafB0rOnMuVwvvvvz/OQjKeirqyrlJFLzahPHvmucroO9iWK1Fy5VCKIehIzvzIgV4oig/JVXr57HxzT4pXibhyT7azDHJiyz05SZKQHaCslzwCSyb4x5U30k52xNu3b7cO8hMj96Sol/fzy+Q7It8P+Z54k0JvUkHPc/dEOJMBhRTL8/0bPINe+Td//nyrmFmwkH+BicT885ADOM9BnOcOnJYtWwb9fci90NJ1v0v+hfe+N9LyT46TZNutW7fO6qu971yT4rGhuIBE3xfaMUdC97shuRXZc5uFNxlYSOUyeehzKMhBigwivEupS8LLQaRUoZWH1Md2G4HcvvXnn38af/zxR8zPpOLsxIkTrUq2CblNRB71cuvWLauSslPH4Xn8kVQGk0f8DBs2THn8jvftDnJAFtvJAbeP+5FbYH3JFSWpvidnSDwVyTyk05V1kjP/4c7pb5NnJ0u8UaNGIcs93/ySL6rcCiiVcX1vg/F8nnJ2UT53ucrifYVPqnTLI56kuracmElI7kkHIc+wdDrJ5KmqLN8R+a5IRUbZyTqtqyf3YntcVyBl372rC86bN896JJec9ffmuS1eh9yTqpgycPX95+lj5P/7VrkOBvKP/IuNTP2R/Uyonu1I7iUs9+Ki0343NuRfeO97IyX/5JZjOc6Wp5l4yLrLnZJy/C6P+gw2+j7Dde45HZfLsakcH5cvX97Ily9fcI/7zBDInTu32a5dO/O9994zJ06caPbp08e84447zHTp0sVUgwu29evXW5PHFy9ebIsfOnTIzJs3r1WkR6ofT5gwwXz77betyfhnz5612hw7dsyaqCxFngYMGGCOGjXKrFixolUEZt68eTGv5SkeJYWenIpHZcyY0XHdpBCQp2CJvPYnn3xiFSzJnz+/7bUmT55stZMiTkOHDrUqiD3//PNWZWePESNGWG169+5tzp4926owFt+qyFIwQIqKSJVq+XykGJYUpJK/d+bMmUp7KUYi28+3kmg4kiJinTt3Nj/44ANz7NixVh7K3yWf56VLl0LynlJ9Wba7FKLxtnnzZqtAmFTr69u3r7Wt+/XrZ217j61bt1p5c+edd5rvvvuu9Z2RqttSDGrNmjUx7TzFo5yqe8vnLvnsS4onNGnSxPr7n3jiCfPjjz+2qtFJTsn30fu1JO/l9aWar3ze0lbyzrvKd48ePazXkgI7c+bMMZctWxbv6nj79u2zCrVJ8QzJd8lj+cwqVarkmF9S4Vqq53lXiNRNKItHCfKP/BNSHK99+/bmmDFjrIKNUoXSqbAduZc0fZ+YPn269RqyP5D3rFevnrUs/6Rv1HW/K8i/8N73RnL+ScFTOQaSQnmvvfaa1QdK0auUKVNaBQNDgf2u6Tr35Ji8du3a1tjLcxwsx8VS+VjGVcE+7gvJwFYGTvfee6+ZPXt2q0qjDODky/3333+boVS+fHnlcQdCKhVLR5ErVy5rwCADBznQ9K4iK1Uk5TEdUhVYBuBy8O1bhjrQga2QD7NKlSrWQbw8yqVcuXLm66+/bj3axZsMVKWDk3ZS0VnWQzoyj4sXL5pPPvmktZ6yLt5J5XZg+/PPP5v169e3BvzSEchryUGQd2fpWxnvqaeeMnUgVaalEp5sY/nbihUrZr7xxhsh7Zwlj3LmzOlYUVUGrlLK35NXUgFOdmTepDprw4YNrUGw7EhkZyMVxb0FMrAVUoVcBsvyc8l9+U5KHsrJDN/HH02ZMsUaYHrayesuWbIk5udyAkiqcsu2lXXxrtLotoOT0vxSpVxyTzo1eaRWbJ+PHBzky5fPOgGjs1APbMk/8k/I/kr2F/L9lH6kRo0a5ldffRWyvCP34j+w8Oyjnf75HuDptN8V5F9473sjPf+kUrEch8uJA9mOsv7yVJZQYb9rus49uQgnj6+UMZiMCeV4WY6LN2zYEJLjvpA+7iexydko+eJ7rsQi4aQMvZwp9FeOPrkbNGiQNUjzfcYsAiePvpITPL4nf0D+JQbyzx36vuBjv+se+Uf+JRVyLzz3u1HyP0aEkDkKcr92u3btjLfeeiupVyciyLxB2a5fffVVUq9KWJNHN0nBqFGjRhnt27dP6tWJCDKHpXbt2saIESOSelXCHvkXfOQfuZdU2O+6R99H/iUVci8897sRNbAFAAAAACQ/IamKDAAAAABAYmFgCwAAAADQGgNbAAAAAIDWGNgCAAAAALTGwBYAAAAAoLVUbhtGRUWFdk2gncQqqE3uIalyj/yDE/o+JBX6PiQl+j6Ee+5xxRYAAAAAoDUGtgAAAAAArTGwBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAa6mSegUAAAAAAIknQ4YMSqxSpUpKrHbt2rblq1evKm3WrVunxPbt26fEDh8+bIQSV2wBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0RvEoAACgvb///ttvm3feeUeJzZs3T4mZphm09ULyQP4hnKVNm1aJ9e3bV4n169dPiUVFRQXUP548eVKJrV271rb8+OOPG8HEFVsAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtRZkuZwD7ThzWXZUqVZRYs2bNlFjp0qX9tnHaNjt27FBiuXLl8luwYvbs2UpsxYoVRjhKrOIakZZ7SLjELOyic/6lSKGeu2zatKkSmz9/vhJbuXKlbfnRRx9V2ly4cMFIjuj7wlPJkiX95vGBAweUNg0aNFBip0+fNsIRfV/4Iv+CR+f9brho3Lixbfl///uf0qZatWquXuu7777z2w8tW7ZMiR08eDDer53Qvo8rtgAAAAAArTGwBQAAAABojYEtAAAAAEBrETnHtnnz5n4fNly5cmUl5rQp3DyU2GkujtMc29q1a/t9Laf70X3nCImOHTsaSY15Zv59+OGHSqxXr15KbOPGjUps3759tuU777xTabN69WoltmnTJiX266+/2paPHj2qtLl9+7ahC+aZuVO2bFkl9tdffwW0zTt16qTEZs6caSRH9H362Llzp225WLFiSpu///5bidWsWVOJXb582Uhq9H16If8Co9OYIxxUqFBBiS1dutS2fMcddyht1q9f72pfHx0dbSQ15tgCAAAAAJIFbkUGAAAAAGiNgS0AAAAAQGsMbAEAAAAAWktlaG7GjBlKrH379vEuChUb32IRThOov/32WyX2zTffKLE6derYlufNm6e0KV26tBL74IMPlFi3bt1syxMnTlTaIPE9/PDDtuVmzZopbVq2bKnErl69qsR8fzdz5sxKmy5duriK+f7umjVr/H5v3D5cG+Hrsccec9WuTZs2fvsip34OCHeLFi2yLb/44otKm7vvvluJpUmTJiyLR0Ev5B+CrXjx4krMqV/LkCGDbfmdd95R2gwbNkyJ3bhxw9AZV2wBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0FmU6VVZyauiy2FJiW7dunRKrXLmybdnpT1y1apUS27FjhxL76KOP/BaPCrWcOXP6bXPq1CkjsblMnQQL19xzMnv2bNtywYIFlTa1a9c2Eturr75qW96zZ4/S5ueff1Zily5dMsJRYuWebvmXJUsW2/LOnTtdFSqrUaOGEjt27FiQ1y5y0Pfpo1ixYrblxYsXuzoeePrpp5XYrVu3jKRG36cX8i/y97uh5FsASkybNk2JtWrVSonNnz/fttyiRQtDZ277Pq7YAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaE374lElS5ZUYrlz57Ytb9++PeBiS40aNfL7fjly5FBiw4YNU2KXL182IgkFVFTbtm2zLf/5559Kmy5duoTwU0keKKDirHnz5rbluXPnKm0GDhyoxIYMGRK0zyY5oO/T1xtvvKHEhg8f7mq/fubMGSOp0ffpjfxzJ1zHHIltzpw5SqxNmzauCkq94dPXnTx50tAZxaMAAAAAAMkCtyIDAAAAALTGwBYAAAAAoLVUhuaio6NdxdyYMWOGEnvyySf93uN9+vRpJTZp0iQlduDAgYDWC+Epbdq0fmNbt25NxDUC/MuTJw+bCREpW7ZsSmzo0KG25Q4dOiTiGiE5If+Q0GNI37o+jRs3dvVas2fPVmInNZ9TGyiu2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGhN++JRbpQsWVKJzZs3T4mVKFHC70Oi58+fr7SZNWuWEqNQVOS78847lVj+/Plty1mzZk3ENUJyt3HjRtvyuXPnlDYPPfSQEsuUKZMSu3jxYpDXDghM165dXeVn7969ldi9997r9/WdiqzcunUrXuuIyEX+IaEyZMigxFq1aqXEpk6dGtDrz5kzR4mZPsVuly5dqrR57bXXlNiRI0cMnXHFFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGtRpu/s4tga+hRR0qlY1Lp161xN5HbaFMOGDYtzWVy+fNlIjlymToLplHtffPGFbblp06ZKm8KFC7sqXhKoESNGKLElS5bEuaybxMo93fLP1/Lly5VY3bp1lVidOnWU2OrVq0O2Xo8++qgSe+GFF1wVbTl06JCR1Oj7Etcff/yhxKpXrx6013/rrbeUmNO+PhzQ9yU+8i/x80/n/a6Tr7/+Wom1aNEipNvLdPFZ/fXXX0qscePGSuz48eOGLrnHFVsAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtpTIikO/EZ6dCUU4TraOjo5XYgAEDgrx2iGS7du2yLadLl05p41RQ6rPPPgvaOlSpUsVvvutePArutGnTxlURiOeff95vwZTbt28HvNmLFStmWx49erTSJkWKFCEtqobk6cSJE3738xMmTEjENUJyQv5FPt8xRtu2bZU2rVq1CqgY0tq1a5XYggULlNjw4cOVWPny5W3Ln3/+udKmYsWKSqxz585K7L333jN0wRVbAAAAAIDWGNgCAAAAALTGwBYAAAAAoLWInGO7ffv2gB7qW6JECSW2aNEi2/L8+fOVNhMnToz3OiIybdy40W8bp3m3gWrdurUSK1mypBKbPHly0N4T+rhy5Yqrdk8++aTfXB41apSr16pVq5YSmzFjhm25UKFCruaGX7t2zdV7IrLce++9tuWsWbMqbf755x8lliNHDiW2ePFi23KnTp2Cso6IXOQfYpMlSxYl1qdPH9tyv379lDbXr19XYitXrvQ7V/aXX35R2ty6dcvVB/Tvv//alvfu3au0qVChghIrUKCAoTOu2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGgtynRZWSkqKsrQVfPmzZVYrly5XLXzLWjiVJzCqYhFhw4dlNiGDRuMSOK2KFdC6ZR7KVLYzxX99ttvSpsiRYq4Kvh06dIlv+/nVLjs2WefVWI1a9a0La9Zs8bQWWLlnm755ytlypRKrF27dkps2rRpfgtUuCmMJipXrux3PUaOHKm0cSq44bZIRmKj7wse38IromPHjrblMmXKuHqtFStWKLHXXnvNtrx+/XpDZ/R9wUX+hWf+JfZ+1+nYvnv37q5iefPm9fv6X3zxhRJr3769EUo5c+a0LR8/ftzV79WuXVuJ/f7774YuuccVWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK2lMpKB+fPnu2rnVIinUKFCcU7GFjNmzFBia9euVWJjxoyxLQ8dOlRpc+rUKVfrivB0+/Zt2/LXX3+ttBk9erQSe/3115XY22+/bVtOly6d0qZq1aoRVewIweVUfGnOnDlKzKk4T5cuXWzLxYoVU9pkz549oL5Vp0JRCC3ffazbYlHjxo3zWyhKXLlyJQFrh0hH/kEMGjRI2RDPP/98QBtn586dSuyZZ55J9A3do0cPv22ciumtW7fO0BlXbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWGNgCAAAAALSWLIpHJcSBAwfiXI6t0EXz5s2V2IQJE2zLly5d8ltgSpw8edL1+iK8OH2eLVu2VGIDBgxQYtWqVbMtp06dWmmzb98+JVahQoUA1hTJhVORpr59+/qN5ciRQ2nzxx9/uCrGsnnzZr/rgMjXv39/JXb69Gm/v3f06FEl9v777ysxCkWB/EMg0qZN66rdp59+qsRq165tWy5XrpzSpnv37kps1KhRRrA4HVf26dPH7+916tRJid24ccPQGVdsAQAAAABaY2ALAAAAANAaA1sAAAAAgNaiTNM0XTWMigr92kS4KlWq2JYXLlyotDl48KAS6927txJbtWqVkdRcpk6CRVruZc+e3dXDuwsXLmxbHjx4sKs5az179lRiNWvWtC2vWbPG0Fli5V4k5l+gcuXKpcSc8mj//v1K7MEHHzQiCX2ff061J5zyJWPGjEps9erVtuW2bdsqbY4cOWIkR/R97pB/eudfYu93J0+erMSefvppJdasWTMltnPnTtvy9u3blTYnTpxQYhUrVlRix44dsy1nyJBBaTNt2jQl1qpVKyV24cIFv/Nwly5dakRa7nHFFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGupknoFkpMNGzbYlh944AGlzYoVK5TYhAkTlJjv7548eTIo64jQO3v2rBIbOXJkQK/1119/BWGNAP+KFi2qxIoUKaLEpk6dyuZMhh566CG/he2cCkX5FktxKpiYXAtFwT3yDwlRqFAhV8WKKleurMTWrl1rW160aJGr36tWrZoS8y0WNWDAAKVNyZIlXRWebeZT6Grjxo1GcsAVWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK1RPCoJRUdHK7FXXnnFVWGh6dOn25YbN24c5LUDAPqY5K5gwYJK7M8//1RiN2/etC3//vvvfgsoioYNGyqxM2fOBLCmiETkHxLDSy+9pMS+/vprJeZUzMlXihTqNcM8efIosfnz5xuBWLp0qRIbPHiwEtuYTIpF+eKKLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNa0Kh5VuHBhV+32799v6GrmzJlKrH379kqsUaNGtuVevXopbUaPHh3ktUO4efjhh5N6FZBMFC1a1FW72bNnh3xdEBw1atSwLa9Zs0ZpkzJlSleFUG7dumVbXr58udLmvffeU2IUikq+yD+EczHXjz76SIn973//C6iglFuHDh2yLU+aNMnVPnbPnj1BWwfdccUWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaE2rObbNmzdXYm+++aYSO3jwoJHU5s2b5+oefl+5cuVSYjlz5lRit2/fti2XKFEi3usI/e3bt0+JnThxQolt3rw5kdYIker8+fOu2rVt21aJDRs2LARrhIRq06aN3/ljdevWdfVaU6dOtS1/9tlnSpubN2/Gex0Rucg/hLPJkye7Orbv0KGDbbls2bKuXv/q1atKbODAgbZlahDEH1dsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtBZlmqbpqmFUlKGLbt26+W1Tu3ZtJVayZEklVqVKFduy0+Zy2jZu2iXktVKksJ+TGD9+vNKme/fuRii5TJ0E0yn3EtszzzyjxJxy4b777rMtr1+/3tBZYuWeIP9i7x//+OMPJbZ3714l1qRJE9vysWPHDJ1FSt+XMmVK23LWrFmVNt98840Se+2115TYpk2b4ixwiOCIpL6P/NNPpPR9iNzc44otAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1lIZEWjixIlBaSPat29vW96xY4fSpmvXrkawlCpVSok5veepU6dsy5MmTQraOkBvN27cUGIXL15MknVB5IiOjlZiQ4YMUWKPPvqoEsufP39EFY+KFLdu3bItnzlzRmlTr169RFwjJCfkH4Bg44otAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1qJM0zRdNYyKCv3aQCsuUyfByL3Y/e9//1Nizz33nBK78847jUiSWLknyD8kVf6Re0iq3CP/4IS+D+Gee1yxBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjji0CxlwLJBXmmSEp0fch0nNPMMcbSZV/5B58MccWAAAAAJAscCsyAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaC3KNE0zqVcCAAAAAIBAccUWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWGNgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGsRNbC9ffu2UbZsWePdd99N6lWJGE888YTRpk2bpF6NsHfx4kUjd+7cxqxZs5J6VSIGuece+Uf+JRVyL/hq1KhhvP766yF45chD/gUf+UfuaZ17ZgCmTp1qen71l19+sf7/3r17Y34eHR1t9urVy7zvvvvMtGnTKj/3Jj+T1xOFCxc2Bw4caAZq5syZZpYsWcxz586ZkWrWrFnmqFGjgvJax44dM7t162bmz5/f+pxk+z/99NO2Nhs3bjRTpEhhbt682QwH/nLP44svvjBr1KhhZsiQwcyaNauVi8uWLYv5ufyO/K68hm8eBmLIkCFm0aJFzZs3b5qRauzYsQnaRmLHjh1mnz59zAoVKpiZMmUy8+bNazZp0sRct26d0jbccs9N/sl3SGJO/4oVKxbTjvxLmvzzfFcfe+wxM3fu3NbnEts+R8e+79ChQ2br1q2tPi9z5sxm06ZNzd27d9vakHtJl3vi33//Ndu1a2fmypXLTJcundUv9OvXz9Zm3rx51r7r6NGjZrhwk39LliwxH3jgATNHjhxWDlatWtWcPn26rQ35lzT559nuTv/mzJkT1vnnJvfkb6hUqZJ1LJszZ07rWPbkyZO2NuRe0vV9R44cMbt27WoWKVLE6vfkeLl3797mqVOngp57qYwQ+OOPP4wxY8YYpUuXNkqVKmVs3rzZSAzvv/++dZUna9asRqSaPXu2sXXrVqNXr14Jep2DBw8a999/v/X/n3/+eePOO+80jhw5Yvz555+2dpUqVTLuvfde44MPPjCmT59u6ODtt982Bg0aZLRq1cro3LmzcePGDWubHT58OCTvJ6//0UcfGb179zZSpkxpRKpPP/3UyJkzp7VNAzV58mTjs88+M1q2bGn06NHDOH/+vDFhwgTrLN1PP/1kPPzww1rn3ujRo60rCN72799v9O/f32jQoEFI3pP8ix/5LPLmzWvl1+LFi2Ntp1v+Sd7Vq1fP+k7169fPSJ06tTFq1Cijbt261j44R44cQX9Pci9+5HN44IEHrP3tq6++an0mBw4csPbH3h5//HEjS5YsVp8r+zIdfPfdd0azZs2M++67z9oHR0VFGV999ZXRsWNH49SpU9b+MdjIv/hr166d0aRJE1tMPjOd82/cuHHW8cRDDz1kfPjhh8ahQ4esY7L169cba9euNdKlSxf09yT34rdvkhy7dOmS9TkVLFjQ+Ouvv4xPPvnE+OWXX4wNGzYYKVKkCFruhWRg27RpU+PcuXNG5syZjZEjRybKwHbTpk3WhpKDkMR09epVI02aNDEfijf5EDNmzGiEo+eee85IlSqVsW7dOr8HPHIr8sCBA61Ey5QpkxHO1qxZY30ZJA9CsSN18sMPPxgnT55M9Fu248qvy5cvGxkyZDDCjexU5aDHO4+efvpp6wSYxL0HtrrlnpADO19Dhgyx/tu+ffuQvCf5Fz979+41ihQpYh1s58qVK862OuWfrOOuXbusk5NVq1a1Yo0bN7am50h/OHTo0KC/J7kXv6lSHTp0MEqWLGkdzKVPnz7WtnI8ISdm5YTKO++8Yw0Sw50cpObLl89Yvny5kTZt2pjjDPl7P//885Dsj8m/+KtcubLx1FNPxdlGp/y7fv26dSKvTp06xpIlS2LWtWbNmsZjjz1mTJo0yXjxxReD/r7kXvxOeskJftlmjzzySEz8jjvusI7XZewmJ5KDlXshmWMrKyuD2sT07bffWgNMSW5fcqXumWeeMfLnz291uHfddZfRvXt36wvhsWfPHqN169bWusuAQK4gLVy40PY6v/76q7WRv/jiC+usv5x1lbYXLlywrmLJgc/u3buts2Hy93sOZGWHJldyypQpY505ypMnj9Xhnz17VlnXRYsWWWfY5fflrIUcoMhVWiFnemWdJEFkPeSfHKB5yJnf6Ohov9tK2sj79OnTxxrUyuBczj7Fpn79+tYgSjqNcCfbWa7GvPzyy3LfinL1LFS5J5/D3Xff7bit5eBYDqDlQKZEiRLGW2+9pZyUkQNQ+bwlh+SsowzQvcmBgXzev/32m3XGS+bzFihQICYv5OBVznpJ/ktOSkcvrl27Zh2YFytWzMp9OVMm8xck7mvmzJlGtWrVrN/Pnj279Vo///yz9TP5+7Zt22a9vyf35H09JO/lnz9VqlRRBgiSg7Vr1zZ27Nihde7FRr6/0ufIjjYUyD/3+Se8+0x/dMq/uXPnWvsLz6BWyKBC+hO5chYK5J773JO+VO4ckv5Y9gVy8vHWrVtx5p7s6xPrjreEkuMg2W94BrVCTp7LXT5xDeITgvyLX9/nIX2a9/Gvzvkn3ym5kNa2bVvbIOjRRx+1jjXkeD0UyD3Dde5J3yBk7ONNToQJ3/4hobkXkiu2SeH333+3Du7l9itvcnutHKxL4nfr1s3a0ctAVw4CZMcig+Hjx49bB52y/NJLL1kH2tOmTbOuPEu75s2b215z8ODB1u+99tpr1gBB/r+4efOm0bBhQ6NWrVrWlWrPFTMZxMrApEuXLtbryxUDObspA5rVq1fHrLO0katXMgDu27evkS1bNquN3KL55JNPWgMiuc1MbrOQW8yE9yBBbvmRgcf/TxmN3dKlS2OSTA565Ayr3EIrySS3dPge+Mkt5ZJ4sq6+2yLcLFu2zPos5VZ4uVJ2+vRpa6Ar265nz54hyz05C+rr77//tgZs8vlK7sl2lU7g+++/jylwJoNFaSODWhlwSlu5NVcGjfJZVq9e3faaMqiVQfL//vc/a+fkIX+nDI7lVnw5GyufrZxQkRxetWqV9f5yVXTLli1W7uzcudPqmD3kzJhcMZVtJ2fQJKflFh7JDbmFVk4YyFlPyTfPwNy7k5I8Evv27QtoGx47dsw6APKlU+45ke+vDNh9T2YEE/mX8PyLjS75J9916W9k/+FL9n8yqPrvv/+CfsKZ3HOfe579rgz85BZ3OREp/azklVxtl5PqvicBheSe52pGOJN91nvvvWcMGDDA6NSpkzXIkJN6cjtoqE6skH/x7/tkXy8XNeTzkRyTYxGnaTK65J/nJL3TyROJyT5Y+kenuyoTgtwzXOeeXCSR7S8XnOTuIbkoI/sryT25y03GZUHNPTPE3n///TiLRwVLgQIFzJYtWyrxjh07WgVAnIrT3L592/qvFLqSdVy5cmXMz/777z/zrrvusiY637p1yzZpXSY9X7582fZanTp1sn725ptv2uLymhKXok/efvrpJ1tcCl5JsY/q1aubV65ccVxP8cgjj1gFapzUrVs3ZoJ9XF566SWrnRR4aNSokfnll19an5MU87n77rvNS5cuKb9TvHhxs3HjxmY4O3PmTMzfJX+L/E3yt8nfKPHx48cH/T1v3LhhRkVFma+++qryszp16lif6f79+2P9PJs1a2amSZPGVuBFJtnL78nv+xZPqFWrllKgyvO5+/59M2bMsHLfO6+FtJP2q1evtpZ37dpltWvevHlMrjuta5kyZaz3ciI5GVte+rNixQprGw4YMMDx5zrkXmwkL2Rbb9++PSSvT/4Fnn9SWCSu4lE65Z/nbxk0aJBj8Q/5mRR1DCZyL365J4W8PPun9u3bm3PnzrX6vFSpUpk1a9a09bUesm/o3r27qYOLFy+abdq0sfpyT1EiKQLz7bffhuT9yL/45Z8chzRo0MAcN26c+d1335mjR482CxUqZO37f/jhB8ff0SH/pO+TnHvmmWdscenvPHnoW6Aooci9+O93J0+ebGbLls1WtEzGTbItg517ETOwTZ8+vfnss8/aYnKQLlWSH3/8cb8HLtWqVVPiw4YNs9Z9y5YttoHtO++8o7T1DGx9BzEyiJTqgCdOnLC+gN7/ZPDlWeevv/7a+v358+fHua5xDWzdkmpx8l4yUPEeyEhVOYlPmjRJ+R0ZcEuFw3B24MCBmC+MVEX2kL+xdOnS1smPYDt+/Lj1flJp1Zt83hJ/+eWXY/1dGaDKjl8OBnw999xz1g7n/PnztoHttGnTlLYy2JRKgNeuXVMOpOQz9s27nTt32tbZ8x3dtGlTnH9rXAPbhGw/+VzkZJGcTHKiQ+45kby78847rUqNoUL+Bc7twFanvu+9995TfvbZZ5+5+n7HF7kXPw8++KD1OciJVqfjDKko7CtPnjxWlWsdyAFq//79rfWVYwl5SoWcnJXjnD/++CPo70f+Jdzp06etHCtRooTjz3XJv7Zt21oniEaOHGldJJCT5fLkhdSpU1vfrYMHDwb1/ci9+Fu0aJF1YkVOqMg455VXXrE+M6eLQgnNvYi5FVn43oIrBX3k3m65RTkuci+37y2fQm7d9Pzc+zVkvpwTmU/imffoIcU85PZhmRPp5MSJE9Z/Pfep+1vXYPDcsiFzP71vz5A5xlLcQm6xePbZZ5VtG84FBLz/LrmdVyafe8jfKPMvZG6TzEMuVKhQyHNP5mz7+zwlP+X2d5l365R7cvuMVMuUW9P95Z7M9/bcEu+de3IbbGwFcrxzT7aR3HaZmORWapkHI7dIyu3SsRXn0SH3nMit5DLtITGKmJF/od22uvR9TnPnpYaCd5tgI/fc8Wx/KaDnTaYZydQj2e/6Fs/TIfc8ZKqP1IbYuHFjzHGFHGPI/ktuQZSpLaFA/gVObn+XKXLDhw+3prj5Hr/qkn8yfevKlSvW9ED5J2RKltQ9mTdvXsgK/5F77sgtxXKsJ/2DTMMQcguyTMGTW+NlCo3v8WdCci9iBrYyL9apGFMoxHaAIHNnfO/jl8GJDGpnzZrl+Dv+qnKGghTRcprILfNsY9uOErvnnnuMcCadtBTnkrnJvo/d8ZxYkL8jmANbeU/58iV17jnFJffKlStnlb93IoWkkooUrmjRooU1z0IeuRLXCQAdcs+JfOelP/A9kA0m8i/0dMg/yQPZ/xw9elT5mSfm6feD+Z70fQnf73rvm3xJbRCn2gPhRvpzeYyb1InwPgaSk8xS+0Fqikgb35OvCUH+BYfnOODMmTPKwFaX/JNHfC5YsMC6cCHzPQsXLmz9k5ohcowtx4TBRO7F/8SD9HueQa2H1ICR2i5yUs93YJuQ3IuYga1MPpaiTN4koeWMgFRNi4t8Af755x8l7qkwLD8PlJwxkqIR8szYuM6YeyrqyrpKBdvYBOPsmWditu9zXWXH4/QIDCmKJVcOJQnDmexQK1asaD3CyHcnKkXEQnEiQa7Sy2fnm3tFixa1/htX7sm6SIGx2HJP/p6EDD5lvaSMukzwjytvpJ0Mgrdv325tv9gE68ytvJcUOpNCX1JURKqAx0aX3PMlV86++eYbq6BKsAcU3si/0NIl/6SvkJNYUqjHl1wpk/4o2IWjyL3473fl0SO++93Y9k3STvZjnjvHwpkUL5TvilOVZ3nigvT5cVWADgT5Fxyeu8t0zj8PuWjhuXAhAyMp0NayZcugvw+5Fz9SoDe2vkFI3xHM3AvJ436Sgjz8VwYR3rdiyc5eLndLFVqnHb7nNgJ5PI88+++PP/6w3SY5ceJEq5JtQm7RlFtx5AOVSsq+5MOUL5+QqnRy4DFs2LCYW8d811PIc0vl1mYnbh/3IwfbnqvI3u8lVZllXaU6sjcZ8Ei7UD2uJJjklmP5G6SqtYesu/yt8jmGYpAhueebX7KTkEpwU6ZMsT4Xp89TrirL5y5nGr2rykknINUkpbq2nJhJSO5JByEHU77kth1PVWX5jsh3RaohywGI07p6cs+Tr77i88gBqa785ZdfWpVA5aptXHTKPW8//vijta1C9exab+RfYI+8cEOn/JPpF3JSz7svkpNmUtlcppmEArnnPvcef/xx66r61KlTbf3s5MmTrf/67nfloFzokHtyPCFXxebPn297jIw8bk+Ov+TCQyhuhSf/3OefTH3yJccHcoxSvnz5mEev6Jh/TuT2fjnGDtVUIHLPcJ17xYsXt45r5ZGp3ubMmWP917fycYJzzwwBqfA7ePBg65+nIq1MEJbljz/+OBRvaa5fv956n8WLF9vihw4dMvPmzWsV6ZHqxxMmTDDffvttqxDO2bNnrTbHjh2zJipLkSepUjhq1CizYsWKVqW1efPmxbyWp3iUFHpyKh6VMWNGx3WTQkDye1JZU177k08+sYoK5c+f3/ZaUjVM2pUtW9YcOnSoVb3u+eeftyo7e4wYMcJq07t3b3P27NlWdbv4VkUWUoRI2kpRlDFjxpivvfaaNdG+du3aStVdmZAv2+/ChQtmuJNq1fLZyt8if5P8bfI3pkyZ0vzxxx9D8p5S3VK25T///GOLb9682SqcIVUw+/bta06cONHs16+fVdTAY+vWrVbeSJGhd9991yr+IoWUpBjUmjVrYtp5ikc5VfeWz13+ZqfiRU2aNLHy+IknnrC+ezJxX3LqjjvusL2W5L28vlTnlM9b2kreeVf57tGjh/Va8j2W4iDLli2Ld3U8yX95n/vuu8+q2uz7Typr6pp73qRCu3yG0heGGvkXv+qM06dPt3JYvpOSi/Xq1YvZX+3bt0/b/JN1lKr2uXPntvYT8l0rWLCgtZ+RYnahQO7FL/ekarXkXP369a1q1d26dbP61Hbt2ilte/bsaVWtdaqWHI6kGKH8bVIsT3JPvjulSpWyYlJIKhTIP/f517lzZ+v4To5/Pccicmwi1Wfl2Fbn/JMCbFJpXI73Pv30U6tIkVNRz2Ai90zXuScVquU4V46HZb8rT+aQPs/TFwY790IysJUKyN4lnb3/JbSib1zKly+vlPwWUqlYDtJz5cplHWzKwOGFF16wVZGVSmqtWrWyylGnS5fOqpLsWwI90IGtkI6kSpUqVvVmeZRLuXLlzNdff916tIs3GajK4ELaSUVnWQ8ZRHjIgf+TTz4ZUzbbe3vGZ2Ar5HVlkCXbRAb2kkxOB3BSFfSpp54ydSEV6+TzkMGb/G2y/vJ4pVCRPMqZM6d1YOxLBq7yGB1PXkn1Qd/H2mzcuNFs2LCh9aWXg2g50P79999tbQIZ2Irr169bg2X5uWyL7NmzW3kolb09FZc9pkyZYh2UeNrJ63pX6pQTQFKVW/JX1sW7QrLbDs5TPTy2f77V03XLPSHbVT7rFi1aJMr7kX/xG1x4+kmnf74HeLrln1T/lP2Y7DukP3n00Uetx3mFCrkXv9yTAzU5aShPYpCTr3LiQSoJSz/te1IyX7581s90Io8vlGMW2d/JMYx8f2QAECrkn/v8kwshUqVajoOlGq0cs8ixyYYNG5S2uuWfHKtL3smxiRxD1ahRw/zqq69C+p7knhmvvk8Gt7Jvkj5P+j75Pbn45Pt40WDkXsgf95OY5Ey8JLbnSiwSTh4RIWeUg/2oiEgjZ+Lluce+V7sROHKP/EtK5J879H3BJ4/DkIGh74lvkH+Jgfxzh74vPHMvSv7HiBAyb0XmCkgV0rfeeiupVyciPPHEE9Z2lSI/iJ3MJZICLaNGjUqUeZXJAbnnHvlH/iUVci808/dq165tjBgxIgSvHlnIv+Aj/8g9nXMvoga2AAAAAIDkJ2KqIgMAAAAAkicGtgAAAAAArTGwBQAAAABojYEtAAAAAEBrDGwBAAAAAFpL5bZhVFRUaNcE2kmsgtrkHpIq98g/OKHvQ1Kh70NSou9DuOceV2wBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtMbAFgAAAACgtVRJvQIAnE2ZMsW23KVLF6XNt99+q8SaN2/OJgUQMr///rsSGzp0qG35hx9+4BMAACQqrtgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABoTaviUVmyZFFiS5YsUWKZMmWyLVeuXFlpc+3atSCvHRBct2/fjnNZmKbJZgcQMuXLl1diFStWVGIFCxbkU0DQ1alTR4mNHz9eiZUoUcK2vGrVKqXNjh07XL1nqVKlbMu1a9d2te+NiopSYh06dLAtz5o1y9U6AAgMV2wBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0plXxqBo1aiixu+66S4lNmzZN20JRzZs3V2Ldu3f3+3tORbQ+++wzJXbmzJkErB2A5Grw4MFKrH///krstddeU2KjRo3yWwgN4alFixZKLF26dEpszpw5hg5Kly6txNKkSaPENm/enEhrhLiULFnSb6Eop2JOtWrVUtrcf//9rgo++b6WU6GoefPmKbHo6GglNn/+fCUGIHS4YgsAAAAA0BoDWwAAAACA1hjYAgAAAAC0ptUc2xdffFGJLV++XIn16dPH0MELL7zgah5btmzZlNiBAwdsy3Xr1lXaFC1a1NV7Mt8NgD+1a9d21XeMGDFCif3000+25W3btrHBkSh896lvvPGG0iZVKvVQaPjw4UqsX79+QV47+LNixQpX82IDqU2SkHVwmk8LxMZpznfNmjWVWJMmTWzLP/74o9ImX758SqxXr15K7ObNm7blt956y9UHNHHiRCV27tw5QxdcsQUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANBa2BaPypUrlxJ74IEHlJjbydDhoHr16n4LRW3YsEGJTZkyRYl9++23tuXx48crbZ577jklNmTIECV2+PDhONYaAAzjypUrrjbDBx98oMSOHTvGJtRUpUqVDF0UKlRIib355pt+CwFt2bLF7+855fGYMWMCXFO45VSkyTRNv7GTJ08qbebPn8+GR4I0btxYiZUpU8a2/Prrrytt0qdPr8QyZMgQUNFGJ06FHFOksF+7HDZsmKvXclr/Bg0a2JY3btxohCuu2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGgtbItH5cmTR4llzJhRiS1fvtzQRbdu3fz+PV27dlVi+/bt8/vaToWiGjVqpMQefPBBJTZjxgy/rw8gefPtv8SoUaNcxU6fPh2y9ULiF2Tatm2bErt06VKSfxQtWrRQYilTprQtf/LJJ66KCtWoUUOJvfrqq7blsWPHKm1u3brlen0RmFWrVimxWrVq2Zb79euntKF4FOLDqWDtV199FVARqFC7evWq3/FRkyZNXL1W9uzZldiIESNsy+3bt1faHD9+3AgHXLEFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWmNgCwAAAADQWtgWjxo2bJgS27NnjxKLjo42dJE3b17b8sSJEwMqFOV24rhTEQunolwA4M/BgweVWO/evZXY9evX2ZgR7s4771RiadOmtS3fuHHDCAf//vuvbXnRokWufm/u3Ll+C6iUK1dOabN58+Z4ryPiZ+jQoUps4cKFtuWSJUsqbZxiOh1DInFlypQpLAtFORkyZIgSe++992zLNWvWVNo4jUNKlCihxOrVq2dbvvfee/1+B5MKV2wBAAAAAFpjYAsAAAAA0BoDWwAAAACA1sJmjm26dOlsy1WqVFHa/Pfff0rs5s2bRjhq166dEmvUqJFt+YknnjASW6FChRL9PeFf4cKFlRifFXScd4vI5ztvVVy5csVIag888IAS27Vrl996FE5WrVoVtPVCcC1evFiJbdq0ye8xpNN+ljm2CCf79++3LX/wwQdKmyNHjiix9evXK7Hbt2/77dPGjBmjxMaOHWvojCu2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWgub4lGpU6e2LefNm1dps2TJEkMXNWrU8Fvo6o8//gja+5UtW1aJZc+eXYmtW7cuaO+J4Hn44YeV2EMPPWRbHjFihNJm7dq1fAxIFLlz51ZiJ06cYOsnQ077m4wZM9qWL1y4ENJ1yJAhgxKrVq1a0I4btm7dqsTOnTtnW65fv77SZvPmzQG9HxJmx44dtuXKlSsrbfr27euqEBUgtmzZ4qpwXrFixUJ2LLhnz56gvXbatGmVWIsWLQLqD522Tbjgii0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWwqZ4VNGiRf222bt3r6ELp7/np59+si0fOnQoaO/XrVs3VxPFV65cGbT3ROKaOnWqEtu5cycfAxLF2LFjXRXKeffdd/lEIsgPP/ygxCpUqKDEUqRIkaQFJ0XOnDmVWKDFV/LkyaPErl696reAFZLG0KFDbctPPfWU0qZ27dpKrE6dOkpsxYoVAa1Do0aNlFizZs38HquZpqnETp8+rcQaN25sW96wYUNA6wl39u/f72r/1qdPH9ty6dKlA97E3377rW35ww8/VNp8/vnnrl4rVSr7EG/RokVKm7p167p6ra5du9qWDxw4YIQrrtgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABoLWyKRzkVo/BXfClc5M+fX4nVqlVLiU2aNClo75kuXTrbctOmTV0V2zpy5EjQ1gGBadOmjavceOaZZ2zLFIpCUkqTJo0Sy5YtW5KsCxLPrVu3wnJzR0VF+S2WEls7NwoVKqTE8uXLZ1tesGBBQK+N4IuOjvZbkMkpVrJkSSV24sQJ23LLli39FoUSlStX9vuebtcrR44cSmzhwoW25QceeMDvdkBwTZ8+XYktXrzYtvzzzz8rbcqWLevq9cuUKWNbnjx5stJmwIABSqxJkyZKrECBAgEVivr333+V2PHjxw1dcMUWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaC1s5tg+9NBDtuX//vtPaXPw4EEjHDVo0ECJZc2aVYnt2LEjaO/ZuXNnv/OBnnvuOSV2/fr1oK0DAjN27FhXc2yAcDJq1CglduXKFSVWrFgxV3N2gMQQaN+aMWNGJXbjxg3bMvvT8LVq1SpXtU/Gjx/vN2ec5mk75ZVTu/nz59uWZ82a5erYcN68eUqsRIkStuUVK1Yobe69914lduDAASWG4PGdf+o0l9WpPxk6dKgSa9GihW05Q4YMSpsiRYoosdWrV7uqi+Fm3/zII48osf379xu64IotAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1sKmeFTmzJltyxcvXlTaHD582AhH99xzT0hf3+kB4u+8845ted26dUqb2bNnh3S9AESuihUrxtlHi7///luJUVAnsvzyyy9K7H//+58Sq1evXpxFc3TTqlUrJbZt2zbb8vbt2xNxjRAfToV5Fi5c6Op3fQtDnTp1ylVxp0mTJimx6Oho2/Lly5ddrYNTEajp06fblps3b660+fDDD13lMkLn3LlzrmKdOnVSYgUKFLAtP/DAA67eM3v27EYgli1bFnHFHrliCwAAAADQGgNbAAAAAIDWGNgCAAAAALTGwBYAAAAAoLWwKR6VN29eI9L98MMPftukTZtWic2ZM8dvu5dffllpc+nSpXivI4LLtxCASJkyZUCvlSqV+nUtVqxYQK81btw4JbZy5Uol1r17d7+vdf78eSV29OhRJVa4cGEldubMGdtymjRplDZ9+vTxuw4IvqxZs9qWc+bM6ffzQ+RxKpDk9LnXrFkzUYtHXbhwQYmtWbMmoNdyyu1atWopsalTpwb0+kh8pUuXVmJRUVGuYr6FpwYMGGAkNqciU75FoG7fvq20cSoolStXLiV28uTJBK8jgq99+/Z+j8uKFi0aUBG0v/76S2kzYcIEI9JwxRYAAAAAoDUGtgAAAAAArTGwBQAAAABojYEtAAAAAEBrYVM86r777rMtHzlyxIg0N27c8Fu0YOzYsUqsfPnySqxp06ZBKZqB0Prkk0/8FuWJzf33329bzpIli9Lmww8/DGi9nApH7Nq1S4m5KQCzevVqJTZt2jQl1rdvXyW2aNEi23KePHmUNpUrV/a7Dgi+2rVr25b/+ecfNnMy5NRX/Pfff66KLYWSU+Gca9euBfRavsWCYivQsmrVqoBeH4mvRIkSfovpiFOnTimxSZMmGTpw+nucYk4FpSZOnBiy9ULwiuJdvXo14Ne6fv26bblKlSpGcsAVWwAAAACA1hjYAgAAAAC0xsAWAAAAAKC1sJlj6zQvQGeHDx/2e7/7O++8o7R5+umnldi7776rxBYuXJjgdUTwNW7c2LZcoUKFgF+rS5cuftts2rTJ1TxtXydOnEj0nBo2bFhAv7d48WJXc+IQOKc8LVasmG35119/ZRPDsnv3bmVLPPjgg7bl1KlT+60zEWxOr+9by8K3jxadO3dWYpMnT1Ziv/32W4LXEYnDab/RsGFDV5/zgQMHjHBUp04dv3VanDCfNjylTJlSiU2ZMsW2XLp06URco8jAFVsAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACthU3xqKlTp9qW27dvr7R5/PHHldiCBQuMpLZy5Uol1r17dyXWsmVL23K/fv1cPfh+/fr1CV5HJI5y5crZlgsVKhTwa/nmgm/+iB07diixI0eOBPyeSJ6qV6+uxDp06OC3yAqSp+eff16Jbdmyxbb85ptvKm0GDx4c0vX68ccfldhLL73kt0Dj/v37lZjT/hnhoWTJkkpsxowZtuWqVau6OlY7deqUoYu+ffv6Lbq6ffv2RFwjJETPnj2VWOvWrdmoCcQVWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK2FTfGoAQMG+J1A7VSwYvHixUrs6tWrRmJyKj6QLVs2JTZt2jTb8pkzZ5Q2jRo1UmIUjwpPzZo1U2KlSpUKqMDJjRs3lNjYsWNty8uWLYv3OgJunD59Won179/ftpw/f342Jiz//vuvsiV+++032/LAgQOVNk77PN9+zq2UKVMqMaccLVKkiG354sWLSpvmzZu76pMRHho3bqzEKleubFuuU6eO0mbFihWGLpwKrTVo0MC2nCKFem1q+PDhIV0vBGeM41QMDMHBFVsAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtRZmmabpqGBVlJKYvvvhCibVp00aJjR49Won169cvaMWk0qZNa1uuW7eu0mbUqFGuigidPXvWtlyuXDmlzZEjRwxduEydBEvs3HPr9u3bfrfJ559/rrTp2bOnErty5UqQ1y6yJVbuhXP+BbPwSoYMGZTYN998Y1vOlCmT0sapEE9ykNz7Pie+OfTDDz8obWrVqqXEnIriRUdH25bz5cuntLnvvvuUWMGCBf2up9PxwObNm5XY66+/rsRWrVplJDX6PsMYP368sl2effZZ2/KmTZuUNpMmTQpom/vmoyhZsmTA7dwUL/MtFOX02X/77bdKm44dOyqxy5cvG8FC36fy3TdOmDDBVWFYpyKzwbR+/XrbcvXq1Q2duc09rtgCAAAAALTGwBYAAAAAoDUGtgAAAAAArYXtHNs77rhDia1bt06J3XXXXUrs5MmTcd5nLvbt26fEUqdOrcQef/xx23KuXLmUNufPn3c1R7hz58625UGDBilthg0bZugiuc+1eO6555RYzZo1bcu9e/dW2pw5cyak65UcMM8scPPmzVNiefLkUWL3339/At4lsiX3vs+NjBkzupob2b59e7+vdfToUSW2evVqJTZmzBgltnXrVtty2bJllTbnzp3z+3vhgr7PMAYPHuy3torTd8dp27lplxSv5TQvdujQoUl+vEjfp6pfv75t+aeffjLCQa9evWzLH3/8saEz5tgCAAAAAJIFbkUGAAAAAGiNgS0AAAAAQGsMbAEAAAAAWgvb4lFOnAo3tW3bVonVq1fPtlypUiWlTZEiRZTYjRs3lNhXX31lW/7mm2+UNosXL1Zi169f91voqkaNGkqbJUuWKLHjx48b4YgiAoj03AuXvi9QuXPnVmI9evRQYocPH1ZikyZNCtl66Y6+D5Gee+Hc92XIkEGJ9e3b17acM2dOV69VqlQpv2127Njh6vecjlG3b99uWz516pSr9froo4+UWHR0tJHU6PtUc+fOtS03b97cSGy+eSYee+wxw1/RXJ1QPAoAAAAAkCxwKzIAAAAAQGsMbAEAAAAAWmNgCwAAAADQmlbFoxBeKCKASM+9SOz75s2bp8SyZ8+uxAYNGmRb/uWXX0K6Xjqh70Ok514k9n1IOPq+xC0e5VTU9v3331diU6dOVWJ79uwxIgnFowAAAAAAyQK3IgMAAAAAtMbAFgAAAACgNQa2AAAAAACtUTwKAaOIAJIKBVSQlOj7EOm5JygehaTKP51yr0uXLrblyZMnB/xa7777rm15y5YtSpuvv/7aSI5Ml7nHFVsAAAAAgNYY2AIAAAAAtMbAFgAAAACgNebYImDMtUBSYZ4ZkhJ9HyI993Sb54jEQd+HpMIcWwAAAABAssCtyAAAAAAArTGwBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKC1KNM0zaReCQAAAAAAAsUVWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWGNgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtRdTA9vbt20bZsmWNd999N6lXJWK8+eabRvXq1ZN6NcLexYsXjdy5cxuzZs1K6lWJGDVq1DBef/31pF4NLZB/wffEE08Ybdq0CcErRxZyL/jo+8i/pET+uUPfF6b7XTMAU6dONT2/+ssvv1j/f+/evTE//+abb8w2bdqYd911l5k+fXqzePHi5iuvvGKePXtWeS35XXk9UbhwYXPgwIFmoGbOnGlmyZLFPHfunBmpZs2aZY4aNSpBr3H58mXz6aefNsuUKWNtr4wZM5rly5c3R48ebV6/ft3W9ujRo2batGnNBQsWmOHAX+5J/kjM95/8Dd7kdyQur+Gbh4EYMmSIWbRoUfPmzZtmpBo7dmyCtpHHrl27zJYtW5rZsmWz+of777/fXL58udJu3rx5ZoYMGawcDBf+8k8cOnTIbN26tZk1a1Yzc+bMZtOmTc3du3fb2pB/SZN/O3bsMPv06WNWqFDBzJQpk5k3b16zSZMm5rp165S2GzduNFOkSGFu3rzZ1CX35syZY1aqVMnq73LmzGn18ydPnrS1IfeSru/zPV6Rz1D2v77o+9xj3+uO53vv9E/6jXDOPzd935IlS8wHHnjAzJEjh7XvrVq1qjl9+nRbG/q+pOn7Dh8+bLZv394aC8p+1/P5fP755+bt27eDvt9NZYRAt27djPz58xtPPfWUUahQIWPLli3GJ598Yvz444/Gxo0bjfTp04fibY3333/fGu1nzZrViFSzZ882tm7davTq1Svg17hy5Yqxbds2o0mTJkaRIkWMFClSGL///rvRu3dvY+3atdZ7eOTNm9d4/PHHjZEjRxpNmzYN0l8ReuPGjTMyZcoUs5wyZcqQvdeNGzeMjz76yNp+oXyfpPbpp58aOXPmNDp37hzwaxw8eNC47777rO3Up08fI2PGjMbUqVONBg0aGMuWLTPq1KkT01byLkuWLNb7Dho0yNDlDG69evWM8+fPG/369TNSp05tjBo1yqhbt66xefNmI0eOHEF/T/LPvcmTJxufffaZ0bJlS6NHjx7W5zRhwgTrCsVPP/1kPPzwwzFtK1WqZNx7773GBx98YEyfPt3Qoc+Tv+mhhx4yPvzwQ+PQoUNWv7R+/XqrX0+XLl3Q35PcC7yfkLtRpP9zQt9H/gV73+vRrl0769jPm+yTdc6/7777zmjWrJn1d7z99ttGVFSU8dVXXxkdO3Y0Tp06ZR2bBRt9n3vyGcj+qFWrVtaYULbdkiVLrHz+559/jKFDhwZ3vxuKsyeeq2Depk2bZrWbNGlSSK7YyihfXmvp0qVmYrpy5Yp569Ytx59dvHgx6O/3yCOPWNspFHr27GltQ9+zdHPnzjWjoqKUq07hfMXW9yqFr2CeuZOzm/L7//77r5mY4sqvS5cuBf395Ap/3bp1E/QaPXr0MFOlSmVGR0fb1rVgwYJm5cqVHXNS8t33rF645t97771nxf7880/bVcKUKVOaffv2jYmRf0mTf+vXrzf/++8/W+zUqVNmrly5rDsHfI0cOdK6oub7O+GWe9euXbPugKhTp47tu/L9999b7caMGRMTI/eSJve8vfHGG2aJEiWsqxhOV2wFfZ9/7Hvd83zv33//fVftwyn//O1369evb+bPn9+8evVqTOzGjRvm3Xffbd2N6EHfl/R9n7dHH33U6v9873RM6H43JHNsH3jgASXWvHlz6787duwIxVsa3377rZEmTRrbFR+Pw4cPG88884x1FTlt2rTGXXfdZXTv3t24fv16TJs9e/YYrVu3Nu644w4jQ4YM1hn8hQsX2l7n119/tc4EffHFF0b//v2NO++802p74cIF68yDXCHcvXu3dTYsc+bMRvv27WPm/o4ePdooU6aMddY8T548xnPPPWecPXtWWddFixZZV3fk9+WMWdWqVWOuoMp2lXXav3+/tR7yT664ehw4cMCIjo4OeBt6XuvcuXO2uOcqxoIFCwxdyFhVPpf/H7OGluSebLu7775b+Zl8HjJfIFeuXNadCiVKlDDeeustW5tNmzYZjRs3tj5vySG54rJmzRpbm88//9z6vH/77TfrqozM5y1QoEBMXsjc8g0bNlj5LzkpVwvFtWvXjIEDBxrFihWzcr9gwYLWlQKJ+5o5c6ZRrVo16/ezZ89uvdbPP/9s/Uz+PrnKL+/vyT3v77nkvfzzZ+XKldYZOdkOHvJ+cjeA3M2xa9cuW/v69etb+S5XO3Uwd+5c6zsr/zxKlixpfaZyBjkUyD/3+VelShXbnRxCrqLXrl3bcd8k+Xfp0iXr7HI4k7t4pN9u27at9d30ePTRR62/V/ZZoUDuuc89D+nj5C4OuaqeKlXsN83R95F/wdz3epM+zfv4V/f8k2M9OWaRYxwP+W7JVe5Q3SFK32cElHveJLcvX76s5GJC97shuRXZybFjx6z/SqKFgtxKKwf3cuuftyNHjlgH67LTl1uk5SBTBrpyACobVAbDx48fN2rWrGktv/TSS9aBzrRp06yDbWnnGZR7DB482Pq91157zRogyP8XN2/eNBo2bGjUqlXLunVXDtiFDGJlYNKlSxfr9ffu3Wvdmi0DmtWrV8ess7R5+umnrQFw3759jWzZsllt5Ba5J5980hoQya1zcklfdozC+yBNbruQzs/tYE6SSToEuTVZbleTdS5cuLA1CPImt3bLoE3WNRS3dIRC0aJFrdu95FYvuUVFbmuQEwqhyr3KlSsr8b///ts6YJbPV3JPvsTSCXz//fcxBc5khyVtZFArA05pK7dGyo5LPkvfwl0yqJVB8v/+9z/ri+9x+vRpa3Ast+LLFAD5W+WEiuTwqlWrrPcvVaqUNS1Acmfnzp1Wx+zxzjvvWLfwyPdAbj2SnJbbF5cvX27dJiwnZl588UUr3zwDc+/tKQM3sW/fvji3lXxfZAfky/NdkcH5PffcYxuICMk9GRCHM9ne8pnLd9iX9EFykuC///6zTloFE/nnPv/i2j857ZtKly5tHRhJ/vnuB8KJ50SV00GcxGQ/Ivkp006CidyLf+7JNCKZriAnwOM62UXfR/4Fc9/rva+XaUAyQJYck2MReR+d80+Ol9577z1jwIABRqdOnay/TS4IyXFtqE4o0/cZ8c49GWvIcascm8vxrUxDk9vHffdbCd7vmonkmWeesW7H27lzZ0hev0CBAlZBGl8dO3a0JiI7FQfx3GLRq1cv69aGlStXxvxMLoFL8asiRYrE3GrsuQVCigRJASZvnTp1sn725ptv2uLymhKXok/efvrpJ1tcCl5JoZnq1atbtzc7rae/W5HldoH4fKRSMMC7gMC9995r/v33345tGzRoYJYqVcoMd1IAS26hke0qt1C//PLL1q2v99xzj3n+/Pmgv5/c7iK3ab/66qvKz+S2QPlM9+/fH+vn2axZMzNNmjS227yPHDli/Z78vu+tOLVq1VJu2/B87uPHj7fFZ8yYYeW+d14LaSftV69eHVPMSdo1b95cua3ee13juiVFctLNLfKPPfaYdcvkhQsXbPH77rvPWie5BcWXbJ/u3bub4U5uf5e/YdCgQY4FGORn3rdgBwP5F7/8c7JixQrrOzxgwADHn0vBi8aNG5vhnnvyN8h+1pvkm6d/l1uug4nci3/u/fDDD9b+aNu2bTHHDbHdiizo+8i/YO175ThEjuPGjRtnfvfdd9axUqFChax9v+Slzvkn07KkYK30gZ7+TopfffvttyF5P/q+wPa7w4YNs405HnroIfPAgQNB3+8myuN+5MyJFOx49dVXbVdjgkmuWPleCZIz1HJV6rHHHrMmI/vy3LIlRa3kiopcafWQs2NylUvORGzfvt32e3JGKLbbG+QWZ29ff/21dcVTLq3LBGrPP88tcb/88ovVTi65y9UcebyOb5EP71vL4iK3Ssfn1ls5ayzvK+v4/PPPW1cLva8CepNtK+sd7l5++WXj448/tq5wS4EYOdspV9/l9i8phBBsZ86csba5b+6dPHnSWLFihXX1TibLO32et27dsq7iyRVlucLskS9fPmv95UqrXFH31rVrV8cCVXILjtwR4E0+V7lKK3cpeOfegw8+aP3ck3vyHZHvilwF9r2i4zb35Hvi5qydfD88t0zKVSS5cixXMOTMqueMnq6551l379uhPDzfaae/LyHIv/jln68TJ05Y3zWZnhLbo6V0yD+52ixTHqSvk7tTZGqN3PYv3zPPHUHkXtL2fXKHlNzxJPtauSLhhg65J+j7wn/fK8chixcvtvJPjonlWEn2wXIHmByb65x/ss8tXry4VZxozpw51rQqOeaXu9d8p3UFA/vdwPa7UrhMxhwyJpT9blz7pYTkXshvRZadq8xvlVt0Q/18Wd9BnQwuZGAgtyjHReYROD2rVTomz8+9X0MOgpzIPf2eeY8eMqCS24dlTmRsB1bCc5+6v3UNJrmdxXNLi3QIUplMBuCyzlIN2Xfbuu1ow418gaTjXrp0qXXiIDFyTw4s/X2ekp9y+7v3fFPv3JMdnlQRllvT/eWezPf23BLvIZ+jzBuUHZe/3JOdqtuDrYSQ26XlxIN8Dp7bt+XWd+kbZGDhO/9Rp9zznOxymr989epVW5tgI//iT07iyRxUOaEoJ5Gcck+n/JMpDHKQIFNk5J+QAzuZRjJv3rxY/76EIvfckSkgcqAmt4LGZ9vqkHv0feG/73UiNWXkhPjw4cOtKW6+x6+65F/Pnj2tAazU6fCcIJATfXLsJAN4mVYVCvR98SNTHeWfZ5ArFw+lho9URvY9NkpI7oV0YPvXX39Zc/zk4F7mqsZVKCGhZF6sUzGmUIjt4FTOGvmedZPBiQxqZ82a5fg7sXV8SUEGtzKHQ4pEybxgb7JtQzU/OjFI0SQ5yxaKHYN8+ZI695ziknvlypWzipTEtk2SaickO1OZjyqD8YoVK1p3dAg56+pLrvDqkHuSC9IHHD16VPmZJyYF7IL9nuRf/MnVsxYtWlg5KFcx4joBJd/tUN1pFExyZ5D03VJEUM6iew4iZN687GekZkMwkXvuycntIUOGWDUS5GS7504cmWsmB3DyeUmdAd8T4PR95F+oeY4D5PjId2CrQ/5JXy7HD3Ji3Pv4W+5UkRPpUs9G2vie+E8I+r7gjTkmTZpk3d0oFz+Dtd8N2UhTzkQ1atTI6qjlVt9QnS32kFs+pCiTN9mZS1EeqRgZF9n5yxkDX54Kw54zDIGQs+VypfD++++P82qNp6KurKtv8SZvoTx75rklQHbCvmTbVqhQwdCR58AhFAUQ5GSNfHa+uee5tTiu3JP8lIOZ2HJPOumEDD5lveTkkkzwjytvpJ0MguWWexlkJkbuSVEv72fnyXdEvh/yPfEmhd5kp+S5eyKcyeclJxI8t1V7kzPGkhPBLhxF/sWf5LoU2pPnJkthEalCHxspCCh3Tej0DG+55dAz/UEOTKUgm0zLCDZyzz05SJNB7IgRI6x/vuROHHl2qHdBP/o+8i8xjvs8d5f5XmTRJf9kGqL00zK1y5c8L1X6e6efJQR9X2jHHAnd74Zkjq1UmJQqa3KgJ2fDE+OqpBwkyyDC+zZAeX+ZvyhVaJ0ONj23EUh1wj///NP4448/bLepTZw40apkm5DbROR2CPlSSSVlX/LheR6tI9tLDnqHDRsWc9ui73p6BgROA8/4PO5Hbodymos7efJk67++85Hl/eREhZz5D3dye6+vcePGWXE50RKq3PPNL8l5eVzOlClTrM/Fm2fby1xZ+dzlKov3PAWp0i1zEGTOt5yYSUjuyc5JzojFVp1OyHdEvitSDVl2Ak7r6sk930dBBaPsu1QXlFslZcqCXHXyJgflQofc85yBXLdunS0f5MSFVJeWx4mFAvkXv/yTCqNffvmlNedertrGRU72SH+sS/75kur6sp8JVTV7cs9d7skJ/vnz5yv/pM6FzL+X/y+flTf6PvIvmPtep2MjOT6QY5Ty5ctbtT10zD/5bsndKPId8n5sjJxIkmN/uegViilA9H1GgnJPyJV2OWnj+1SRBO93zRCoUKGCVfHq9ddftyqzev/7+eefQ/GW5vr16633XLx4sS1+6NAhM2/evFaFNKl+PGHCBPPtt9+2qsydPXvWanPs2DEzT548ZtasWa3KmKNGjTIrVqxoVViTB4B7eKoif/3118r7x1Xd8LnnnrN+Typ8yWt/8sknVrVeeaC092tNnjzZale2bFlz6NChVvW6559/3qrs7DFixAirTe/evc3Zs2db1e3iWxVZ1kEeDi8PiZftIZVo5QHX8rtStdaXVBeWn/37779muEufPr3ZuXNn84MPPrAq0bZr1876HOXzvHTpUkje07N9/vnnH1t88+bNZqZMmcwcOXKYffv2NSdOnGj269fP+n54bN261cqbO++803z33XfN9957z6q6nTZtWnPNmjVKVWSn6t7yuUs++5IKx02aNLH+/ieeeML8+OOPrUqIklN33HGH7bUk7+X1a9asaeWDtJW8867y3aNHD+u1Bg8ebFXUXrZsWbyr4+3bt8+sVq2aOWTIECvfJY/lM6tUqZJSKVlIhWup3BgOD4l3Q/4GeSh87ty5re+qfNcKFixofddPnDgRkvck/9znn3wekudShdt33yT/pLqmN/kuyL7DKTfDjVScbN++vTlmzBjz008/tSqgyt8q37VQIfcSVpE7ruMG+j7yL5j7Xjkuql27tnX86zkWkWMTqXwsx7Y655/0cdLXyXGE9PHSb8tTPCQ2c+bMkLwnfZ/pOvdkvCNPXenfv7+Ve8OHDzerVq1qfT4vvvii0j6h+92QDGy9yzn7/outZHkwlC9fXnncgafMuRyk58qVyxowyMDhhRdeMK9duxbTRh630qpVK+tRJOnSpbMOvn1LoAc6sBXyYVapUsU6iJdHuZQrV84a+MujXbzJQFUGF9IuS5Ys1npIR+YhB15PPvmktZ6yLt5J5XZgKwOa1q1bW52WbA9Z78qVK5sffvihVcbcV9u2ba3HzOjg2WefNUuXLm1t49SpU5vFihWzBvChPDCVPMqZM6e10/ElA1d5jI4nr+SEgu9jRTZu3Gg2bNjQGgTLl7levXrm77//bmsTyMBWXL9+3Rosy8/ls86ePbuVh++8847y+KMpU6ZYOwZPO3ndJUuWxPxcTgDJ46Zk2/p+l912cGfOnDEff/xx62ST7FDlkVqxfT4yMM+XL5/VGerk4MGDVl8i31/5TB999FHrkUqhQv65zz/PY9li+7d3715be3n82lNPPWXqQPZXsr+Q76f0IzVq1DC/+uqrkL4nuReagS19H/kX7H2vXAiRRwjKcbA8ckqOWeTYZMOGDRGRf/KIR+n/5FhLjp+l75bBZ6jQ95muc08uaMpxkJzgl+NyyeP777/fOq51OnGS0P1uoj3HNjFMnz7d2mCeK7FIuKNHj1oDslA9DyxSyLNLZZDm+4xZBG7+/PnWDsr35A/Iv8SwadMm6yqJ/Bexo+8LPvo+98g/8i+pkHvhud9NlOfYJpb27dtbRTPGjh2b1KsSMeQ5sFIQRwpbIHYyh03mdHzxxRdspiB57733rArKvnN/QP4lBnkEhsyZjqugGuj7QoG+zz32veRfUiH3wnO/GyWj26CuFQAAAAAAiSiirtgCAAAAAJIfBrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAa6ncNoyKigrtmkA7iVV3jNxDUuUe+Qcn9H1IKvR9SEr0fQj33OOKLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtMbAFgAAAACgNQa2AAAAAACtMbAFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWGNgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABojYEtAAAAAEBrDGwBAAAAAFpLldQrAESyXLlyKbHBgwcrsebNmyux6Ohov6+fIUMGJVa1atV4rSOSl9SpUyuxJ598Uok99NBDtuUOHToYie2pp55SYl9++aVt+ebNm4m4RgBgGClSqNeFqlWrFtJN0759e9ty+vTplTZnzpxRYt98840SW7t2bZDXDggPXLEFAAAAAGiNgS0AAAAAQGsMbAEAAAAAWmNgCwAAAADQWpRpmqarhlFRoV8baMVl6iSYTrlXsmRJ2/KiRYuUNoUKFXK1LX3/bqc2V65cUWJDhw5VYsOGDTMiSWLlnm7551vQxKng01tvvaXEihUrZuhi7NixtuV+/fopbS5evBjSnKHvQ1Kh7wsPTzzxhBKbPXu2EY7++usvJVapUqWAXou+L3gqVKigxOrUqWNbzpkzp9Kmf//+SmzlypVK7PHHH7ctnz9/3tCZ29zjii0AAAAAQGsMbAEAAAAAWmNgCwAAAADQGnNs/ahZs6ZtefXq1Uqbb7/9Vom1bdtWiV2/ft2IJMl9rkWuXLmU2Pjx423LzZo1U9ocPHhQia1YsUKJ7d+/3+9cwo0bNyqxqlWrGpGOeWbOihcvbluOjo42kqPs2bMrsWDOL0rufR+SDn1feBg8eLCr+gW+zp49q8QyZMigxP755x8ldvr0advyjz/+qLQ5cuSIEvv555/9vpZb9H3+NW7cWImNHj1aiWXLlk2J5ciRI2j7iunTp9uWu3TpYuiMObYAAAAAgGSBW5EBAAAAAFpjYAsAAAAA0BoDWwAAAACA1lIl9QqEk/Tp0yux5cuX+5283LRpUyU2Z84cJda+fXvb8tWrVwNcU4SDvn37KjHfYlFO+XLvvfcqsVOnTvktTuH0WpMmTVJiixYtUmIdO3a0LTdv3txVAavkWnxIB6lSqd13ixYtkmRdws3AgQOV2CuvvJIk64K4vfzyy0ps1KhRfgujDBkyRIkNGDAgoM1dsWJFJbZs2TK/RclGjhyptHn99dcDWgeEr2eeeUaJ9enTR4kdOnTI7zHB0aNHlTZp0qRRYidPnlRily9fdrW+SNpiUVOnTlXa5MyZU4k59WvBLM5VuHBh23LGjBmVNpcuXTIiDVdsAQAAAABaY2ALAAAAANAaA1sAAAAAgNYY2AIAAAAAtBZlupyp7DTJOdKkTZtWiR08eNDvBHDfolBi5syZfosIfP/994bOgjnJPdxzz3cSvvjzzz+VWK5cuWzL7777bsAFTrZv3x7nsihdurQSK1GihN9t6PTZde/eXYlNnDjRSM65Fy7556Rnz55KbMyYMSF7v+PHjyuxcePG+S2kISpXrmxbTp06tZHYUqQI3nnc5NT3uZUlSxbb8vjx45U2TsWWnIo0+b6WkwMHDiixhx56SInt2bPHb6GoJUuWKLE77rjD7zrcunXLb6E+8cUXXxjBQt8Xep06dbItT5kyxdV388yZM34LkH7wwQdKm7Vr1xq6SE59n9M6+Ba2Ex06dLAtZ82aNeDXD3T7unmtRx55RGmzePFiQxdutw1XbAEAAAAAWmNgCwAAAADQGgNbAAAAAIDWGNgCAAAAALRG8Sg/fAtIpEuXzlXBgCNHjiixv/76y7Zcr149Q2fJqYhAlSpVXBV88F3XoUOHBlw8yrcwidP2dto28+bNU2ItWrTw+1qzZ89WYq+88ooSO3XqlJHUklsBlWzZsimxdevWKbG77747oNe/evWqEhs2bJjfQmJOBaWcPPbYY7bl4cOHK21KlSplhBLFo0Lr008/tS137dpVaXP27FklliNHjqCtw5AhQ5TY/Pnz/RZLcSoK6cbFixf99rWxFcgKVHLr+0KtVatWSuzDDz+0LRcoUCBo73f9+nUl1qVLFyU2Z84cIxwlp+O+0aNHuyraGCg3BZ++++47pc3+/fuV2EsvveT3tXbt2pXo+91gongUAAAAACBZ4FZkAAAAAIDWGNgCAAAAALSWKqlXINw5zZ91w2ku0b///huENUJScJov5jQ/wjfmNN8qV65cSqx58+Z+X8tpbmvdunWVWHR0tBIbPHiwbblfv35Km6eeekqJXbp0SYl1795diSG0ChcuHLT5tE42b97sN2cS4vvvv7ctZ8+eXWnz+eefB+39EDxOn9WoUaOUWJMmTfzOaXY7n/b06dO25RdffFFpc/jwYSV248YNJda5c+egzKd18tZbb4V0Pi2Cq06dOkps2rRpSix9+vR+X2vp0qWujvtKlChhWy5fvrzSZurUqUps9+7dSuzPP//0u15QFSlSxO/c0ueee05pU7t27ZBuTqe5snPnzrUtv/32267+Hqc5tm5+r0OHDkpsxowZhs64YgsAAAAA0BoDWwAAAACA1hjYAgAAAAC0xsAWAAAAAKA1ikeFSK9evZRYpkyZQvV2CLH58+crsWeffdbv7/kWjhAlS5Z09eDpFStW2JZfeeUVV4WinAwbNsy23KxZM6VN6dKltX54dyRzKmwBJIamTZu6KjgSqD179iix559/PqCCTE7FqXyLWiXEypUr4yz0gvDhtH8eN26cEkuZMqUSu3Dhgm15/PjxrgqH3bp1S4llzpw5zhyKraCUU3FAikcFr2jYhAkTbMupU6c2guXIkSNKzKlg56BBg5RYmTJlbMtff/21q2NIN06ePOn3ODMScMUWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAaxSPCoJ06dIpsYULFyqx27dvB+PtkAQ2btyoxDZt2qTEqlSpEtBr7dixQ4kNHTo0oEJRTi5fvmxbvnr1qtImKipKidWuXTvg90TwHDx4MKSb8/DhwyF9feghTZo0SuzNN98M6LWcCuI5FVtq3769qyI8vgoWLKjE3n33XVdFeNy4efOmEnv11Vdty8eOHQvotRF8NWvWtC2PHTs2oEJR4tFHH7Utr1q1KuD1+u+//2zLO3fudFU8qnXr1kpszpw5Aa9HcjZ9+nS/xb8C7SecZMiQQYk5Ff8sXLiwEitUqFDQ1sPNejmtw/79+w2dccUWAAAAAKA1BrYAAAAAAK0xsAUAAAAAaI2BLQAAAABAaxSPCkJxjWHDhimx1atXuyqcAT2cPHlSiTVu3Digif9OxaMS27x585RY5cqVXRWAQeL7/PPPXRXKCWYBvLp16/r9PacCak7FWBLbrFmzknoVtORUyKl48eJBKZojnnjiiYBe64UXXlBiFSpUcLX+gVqwYIES27BhQ9BeH8H12muv2ZZTp06ttDl37pwSa9SokRL7888/g7x2CDcrV660LRcrVszV7/31119KbMWKFbblpk2buuqvUqRIEbQis25eK3v27Eqbnj17uiok+ttvvxm64IotAAAAAEBrDGwBAAAAAFpjYAsAAAAA0FqU6XISndM915EmVSp1yrHvPI3hw4crbXLmzOnqofYHDx40Iklizb9MDrmX2EqWLKnEtm/f7uoznj17tm25Q4cORmJLzLm/iZ1/FStWVGLjxo1TYtWrVzeS2ubNm13NrRw/frzfv7FPnz5BW6977rlHie3evTtorx+pfV/KlCmV2JYtW5RYiRIlAnr9GzduBG3fHOpt4zTX7ddff7Ut169f30hskdz3JYRvzYFq1aopbbp27arE9u3bF9L1ypw5s215//79Spts2bIpsc6dOyux6dOnG0ktUvq+DBky2Jbz5cvn6vfOnz+vxE6dOmVbXrhwodKmYcOGrv7GQLdvMF/rgkOdDN999oEDB4zE5vbv4YotAAAAAEBrDGwBAAAAAFpjYAsAAAAA0BoDWwAAAACA1tSKDMmkUM4rr7yixCpXrqzEKlWqZFs+fPiw0sbpwc7Xr18PYE2BxBEdHe1qYr5TzOn7hMD5Prj9xx9/VNrkzZs3LDexUxEoJ7Vr1w7ZOsycOVOJ7d27N2TvF8lu3bqlxDZs2BC04lG+xRjD2bBhw5TYmDFjkmRdIpVvYSXfvlDUq1dPia1fv16JLVq0yLb8zjvvhMVxWf78+f0WinJy8uTJEK0RxOXLl0NWXDCYn92RI0eU2KVLl5RY7969/R6/Pf/88672zdmzZ9e67+aKLQAAAABAawxsAQAAAABaY2ALAAAAANAaA1sAAAAAgNaSRfGo6dOnK7F7773X1e+uXbvW70Tuxo0bKzGnIlO+k9PPnj3rah2AxBAVFeWqXa5cuWzLOXPmVNqcOnUqaOsV6Z588kktCkWFq0yZMiX1KkS0sWPHKrEmTZooMbdFcZLanj17lFj79u2V2F9//aXErl27FrL1So7mzp1rW65fv37QXtvp8xs0aJAScyrWF+jnXKBAASX2/fff+/09p2NBCuDpq1evXq4KnrVs2VKJffPNN37zZ//+/QGt1+LFi5XY6NGjldiLL76oxFq1amVbfu+994xwxRVbAAAAAIDWGNgCAAAAALTGwBYAAAAAoDUGtgAAAAAArUWZpmkGs7BMOMiXL59tefLkyUqbHTt2KLH+/fsrsRs3bvh9v5QpUyqxtm3bKrGbN2/alhcsWKC0uXz5sqELl6mTYDrlns5u3brl6jPetGmTbblq1apGpOZesPPPqUiNb/GGHDlyBO39kqvKlSsrsc2bNwft9ZN739e8eXMl9tlnn9mWs2bNaoQD30IrTgWKfAs7hjNd+z43f4vT37ZlyxYlVqxYMSWWPn36gNZhxYoVfvPbqbhTihTqdaFx48Ypsa5du/pdh3r16imx3377zQhHyb3vc8OpkF6HDh2U2Mcff2wktVY+RaHE119/rcRu375tW37kkUeUNj/99JMRDrnHFVsAAAAAgNYY2AIAAAAAtMbAFgAAAACgtVRGBDp69Kjfe8FDPTfRaV7F/fffb1suXry41nN9EFkmTZrkKo8zZsxoW86QIYPWc8UT08yZM5N0zlxy0aJFi5DOsU3u5s+fr8ROnjxpWy5YsGDArz9kyBDbcpEiRQJ+rVGjRtmW2cfqZfjw4a7mnzZs2NBvnZMGDRoosTp16iixMmXK2JbXrl2rtHn++ecDmk+7detWJUbfFFly5szptx+KbS7u4MGDjcQ0d+5cVzlaqlSpOJcTY46tW1yxBQAAAABojYEtAAAAAEBrDGwBAAAAAFpjYAsAAAAA0FpEFo8KB07FBk6fPh3nMvRSsmRJJdayZUvbcrNmzZQ2VatWNcKRUzEAp8JGFDtCuNu0aVNSr0Kys2rVqoB+L2XKlEosTZo0tuUpU6a4eq2bN2+6iiE8/Pzzz7bl+vXr+y0KJebMmaPEpk6daltetmyZ0mbfvn2u1qt27dq25Y4dOyptnn32WVevtW3bNr9/z/nz5129FiJLly5dlNiSJUtsy1u2bFHaXLp0KWjr0L1794B+r1u3bq4KZCUFrtgCAAAAALTGwBYAAAAAoDUGtgAAAAAArTGwBQAAAABoLcp0WQkmKioq9GujqcyZM7sqxHPixImAChmEq8QqIhSuubdo0SIl1qhRI9vy7du3lTadOnVSYjt27FBiGzZs8LsOhQsXVmI5c+b0O9HfqahVrly5XH3GvgU/GjdubCS2xCxgFcz8oxhX/Hz++edK7MqVK7bl33//XWkze/ZsV9s+UMm97wuml19+WYl9+OGHfn/v+PHjfvtf8ffffxuRRNe+z0n16tVty7/99pvSJkUK9frLxx9/rMQ+/fRT23K+fPmUNitWrDCC5cKFC0ps7NixSmzo0KEhK/yTFOj7/CtWrJgSi46OdvX98t2+ToXznAo3uZHL4Rhv8eLFSqxChQp+12vXrl2uxj1JkXtcsQUAAAAAaI2BLQAAAABAawxsAQAAAABaY2ALAAAAANAaxaPiqXPnzkrs0KFDSmz16tV+i57oLrkXEXCawD9u3Di/28hNwQCxadMmv+tQqFAhJZYjRw6/7+l2vbZv367E6tWrZ1s+deqUkdh0LaDSunVrJfbll18autq7d68SW7BggRKbMGGCbfnkyZOuXv/cuXNKzKkgW2JL7n1fOBSP2rx5sxKrUqWKEel07fvcGDlypBJ75ZVXjKS2bt06V3m7Zs0aI9LR9wVWpMm36KaoWLFi0PZvTkXWbofwtf7991+lTYkSJYxQongUAAAAACBZ4FZkAAAAAIDWGNgCAAAAALTGHFs/ChcubFv+5JNPlDYtWrRQYjdu3DAiHXMtVG+99ZZt+c0331TaZMyY0dW2DHRerJt2TnMcn3/+eSU2f/58IxzpOs/M6bWaNm1qW+7fv3+izx08fPiwEps1a5bf35s0aZIS2717txHp6PuC58iRI0osT548tuUTJ04obdq1a6fEfv31VyPS6dr3Bfp+jRo1UmL9+vVTYuXLl7ctp0uXTmkTHR2txM6fP6/Ehg4dalteunSp0ubmzZtGckTfF5icOXMqsV9++UWJlSpVKqDXd3ssGKzX+u6771yNhYKJObYAAAAAgGSBW5EBAAAAAFpjYAsAAAAA0BoDWwAAAACA1ige5cdDDz1kW86SJYs2BXZCjSIC/pUsWdJVMQynIlO+D/l22t6nT59WYvPmzVNip06d8lv058CBA4YuIrmASsqUKZVY1qxZlVivXr2U2JkzZ2zLY8eODXh7JtfiKG7Q9wXP9u3blViJEiVsyxMmTFDa9OjRw0iOIrnvS4jixYvbljNkyKC02bx5cyKuUWSi7wtdcVoxc+bMOIuixVaANNTFo/bu3Wtb7ty5s9Jm1apVRihRPAoAAAAAkCxwKzIAAAAAQGsMbAEAAAAAWmNgCwAAAADQWrIoHpU5c2YlVq5cOSXWsGFDJTZu3Djb8rFjx4K8dvqiiAAiPfd07/sQGvR9wbN48WIlVq1aNdvyfffdp7SJjo42kiP6PiSH/GO/+/8WLlzoaqzyf+3dC5xNVf/48T3lNsPoQhd3laSUEsko5HGtXAspPUqU8ngKpae7UG5hPCGSSClCusldTyXGLZE7xcgt9xiU6/69vuv/OvM/+6w9c/acOWfmrDOf9+vlVfs76+yzZ886a+/v3nt9TziLR3V3KUz50UcfOZaPHj1q5TSKRwEAAAAA8gQeRQYAAAAAGI3EFgAAAABgNBJbAAAAAIDR8lkxKLBY1LRp07Q2+/bt02Ju7SgWBQBAZMyaNUuL7d+/37GcVwtFAcjb7r333tzeBONwxxYAAAAAYDQSWwAAAACA0UhsAQAAAABGi8k5tmlpaY7lJk2a5Nq2AACA7M27BQAgGO7YAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo8XZtm17ahgXF/mtgVE8dp1so+8ht/oe/Q9uGPuQWxj7kJsY+xDtfY87tgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAIG8UjwIAAAAAIBpxxxYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABgtphLb48ePW5dffrn18ccf5/amxIyaNWtazz//fG5vRtSj74Vfu3btrLZt20ZgzbHn/Pnz1o033mi9+eabub0pMYP+5w1jX/hx3PWO/kf/yy30vSgd++wQTJgwwfa99H//+5/6/+3bt6f/vHfv3ioW+K9gwYKO9chrJC7rEPL/su5QvfHGG/bVV19tnz171o5Vo0aNytY+8tm6dat9//332xdffLEdHx9v33HHHfa3336rtZsxY4adkJBg7927144GwfpeuXLlXPue/KtQoUJ6O/pe7vS9jRs32r169bJvvvlmu0iRIvaVV15p33PPPfaKFSu0tqtWrbIvuOACe/Xq1Xa0CNb/5PPSqFEju0SJEnaBAgXsUqVKqc/Z2rVrtXX5j3fSb2XcDNWkSZPsokWL2n/++acdqz7++GM7OTk52+vJaHwYMGBAVPe/YH1PTJ482a5atao61hYvXtx+7LHH7AMHDjjaMPbl3nE38DMrf8PChQtH/XFX0P9yT17vf1763q5du+w2bdrYF110kZ2YmGg3b97c/u233xxtGPtyp+/59rvbPzlmhbvv5bMiaPTo0VaRIkXSly+88MKIvdeZM2es//73v1aPHj0i+j657Z133rGKFy9uPfrooyGvY+fOnVZSUpLaT7169bIKFy5sTZgwwWrUqJG1cOFCq06dOultW7RoYRUtWlS9b9++fa1oN3z4cHUVzd+OHTusV155Rf1+kUDf827cuHHW+++/b91///1W165draNHj1rvvvuuuko3Z84cq0GDBultq1atalWvXt0aOnSo9eGHH1omWLt2rXXJJZdYzzzzjPqc/vHHH9b48eOtGjVqWCkpKdbNN98ckfd966231B3Giy66yIpVn3zyibVu3Tqre/fu2V5Xw4YNrQ4dOjhi0t8Cl03qf3K8lc9U/fr1rWHDhlm7du1Sx8SVK1day5YtswoVKhT292TsC40co+SuhBx73Zh23BX0v+g+74vl/ie/T7169dT5xEsvvWTlz5/fSk5OturWrWutXr3aKlasWNjfk7Ev6x588EHrnnvuccQkFwl334toYtu6dWv1YcwJM2fOtA4cOJDjjy6eOHEiw8Hh5MmTVkJCghVtBg4caP3555/qJPG6665Tsccff9yqVKmSujDw008/pbe94IIL1N9RTuz69OljxcXFWdGsZcuWWuyNN95Q/23fvn1E3pO+l7WB7fXXX3dc8Hrssces66+/XsX9E1shn+fevXurQc7/NdHqtdde02KdO3e2SpcurU78xowZE/b3/Pnnn601a9aoBCwn/f3331aBAgXUGJGVcTEaVKxY0Xr44YeDtjOl/50+fVqd0MlFyfnz56eP07Vq1bKaNWtmvffee9a///3vsL8vY19o5JiUmJioTsa/+OIL7eemHXfpf9F/3hfL/U/G561bt1rLly+3brvtNhW7++671fQcOS72798/7O/J2Jd1t956a9Djbjj6XkTn2MpTX8eOHVP/jTT5cJYvX9665pprtJ9t2rRJnaBcdtllVnx8vErmXn75Ze3kUD4IcqVATmDkqvfSpUsdbT744AO1k7///nt1ZVzm88oJq7jrrrvUh0iSQjm5kIFNTjTEqVOn1MlRhQoVrIIFC1plypRRV8skHmjSpEnq7o68Xu78yLrmzZunfia/3/r169X7y3bIP3lfn99++039C2bRokXqboQvqRXyfs2bN7dWrVqlBojAuxty11OufJl6p+eqq65SJ3mRQN/z3veqVaumJQhyNbV27drWxo0btfbS9yRJkpN1U8k4IZ8vuZgUqf4nCab/kxY+u3fvtjp16mSVLFlSjT3yOXjqqafUiajPtm3brDZt2liXXnqp2k65e/7NN9841vPdd9+p8WbKlCnq6YdSpUqptjK+y10E+ZvK31+uxsoJk+8iksz9lacoKleurO4YXnHFFVaXLl2sI0eOaNs6e/ZsdYVdXi/jsJygyGdXyDgn2yTjkG/sk/HQ5/fff1fjfFb89ddfKjnPjCn9Ty5SSv964IEHHCcCTZs2VX8b+btFAmOf97HPR46vcjdJ7qrny5fxvQWTjrv0v+g/74vl/jd9+nR1vPAltUJu1Mh5/NSpUyPynox9Vpb7npDjqf/5RyT6XkTv2F599dXqEQG5ci930uTKiZzYRMKSJUvU1YBAv/zyizpplkcTnnjiCTVIyB/i66+/Ti+0IoOGtJGTKRl4pK08HimDhwwmt99+u2OdktRKkix3Z+SP5HPo0CGVHMsjgXJVQn5XObGThPHHH39U7y93puRxRRlYtmzZ4rhaJlcn5K6VJGByC15OVuURsm+//VY9RisniHLVXU5UfIm5//6UD7FITU3NdF/JwCqDZyDfVUZJzq+99lpHMiIWL16sPa4X7eSChSRMgRcywom+573vZUQe2XV7uuOGG25QF6Ok77Vq1coyhSQZ8qiS/F7yuZUE0LePItH/5KKajFv+9uzZo06WZFtk7JEDvSS6chIgdxVkfNm3b58ab2T56aefVhcZJk6cqMYsaRe4z/v166de99xzz6lxRP5fnD171mrcuLF15513WkOGDEkfSySJlQuCHTt2VOvfvn27NXLkSPW5lL+pb5uljdy5lwT4xRdftC6++GLVRh5Pf+ihh9TnVx4zk8drZewU/hdI5LFiGau9XkSV95Or/NJexmRJ1uV9TO1/vpNl2dZAEpN9Kccit7vr2cHYl/WxTx6llztlchEos5Nuk4679L/oP++L1f4n+1rO8+X4EUiOf3KBIC0tTV0wDSfGPivLfU/6ukx/lIsz0r8kB3ObIpjtvmdHwPDhw+1u3bqpYh/Tp0+3n3nmGTtfvnz2tddeax89ejTs73fmzBk7Li7OfvbZZ7Wf1alTR00k37FjhyN+/vz59P9v2bKlKvTiP9F8z5496nXy+sAJ7HfeeadWoKpu3brqZ2PGjHHEP/roI1WAZNGiRY64tJP2ixcvTi/mJO1atWplnzt3LsNtrVy5snovN1KARv4F06xZM1U06tixY454UlKS2qYhQ4Zor5H989RTT9mmkT4hv9OGDRsisn76Xtb6npsffvhBfX5fffVV159XrFjRvvvuu22TXHfddenFEaRI1iuvvKJ9rsOldOnSqkBVoA4dOqgxxa0wl29M6d69u9pG//EpLS3Nvuqqq+zy5cunb7OvYIcU5zt58qRjXY888oj62QsvvOCIyzolLscBf3PmzHHEpeCVjLW33367/ddff7lup7j33nsz7GO+8deLWrVqqWPUl19+aY8ePdq+8cYb1WvfeecdY/ufFIiSz1CnTp0c8U2bNqX3w4MHD4b1PRn7sj72zZw5U50LrV+/Pv2z41a8x7TjLv0v+s/7YrX/Sd+Tfdq3b1/XwkfyMxkHw4mxL2t9T/IvKaopx9uvvvpKHX/Lli2r+r70yXD3vYgktm7kJMat8mQ47Nu3T61bqiL7279/v4pLYp0RSVClAlfbtm21n3Xp0kXteF8y7ktsJ06cqLWVQUcqUZ46dcoRl8psMijJh8//35YtWxzb/NZbb6nln3/+OdPfNbMBzqtZs2ap95KTNan8uXnzZrWP8ufPr+L9+vXTXnPFFVeoinMmkQOFVKWVKqGRQt/L/v6TxEwSJkmo3EjCc9ttt9kmWbJkiUrgJFmSbZcLLKdPn47Ie0lV886dO2t9X6okt2jRItPXStJWo0YNLS7jtIwFvmrOvsS2T58+WltfYht48fDpp59WFSplHA4c/yTZ923ztGnT1Os///zzTLc1s8Q2O2TMluRWLvYFJu0m9b8HHnhAnbTKhUm5SCsXjKT6uG9c37lzZ1jfj7Ev6/1MLu7LRX+fYImFScdd+l90n/fFav/7/fff1T4cNGiQ9rP333/f0/7NKsa+7Dt06JDqX3ITINx9L8e+x1Ye87ryyiutBQsWROw9Ah9Dk7ljQh7Ty4gUnJLH8Pznm/rI4yPymINUEfYn89TcyLwz36N5/vMZ5FFneXTZ/58ULxH79+9X/5XHo+UxMXn0LdLkcekRI0ZYP/zwg3p8W353mb/mezTbrUiK7NtoLyAQSB5NlEcvI1U0yh99L+vkMX6ZAyiPCX355ZcZFucxse9JpT95NFfms86dO1fNoZJHbHOq/8m4Jo8/Zzb2CZnHktHY5/u5l7FP5mr56g34j33y+LDMMQ4c/2SKiv/YJ4Jta6TImN2tWzf1yLZ/4TzT+p9Mn5HHC+Uxcak1IfP0brrpJlU8SkSq+BVjnzfyGOrBgwfV43hZ2bcm9D1B/4vu875Y7X++6Rduc5d9NRTcpmiEA2Nf6KSmh0xR2rx5s5piFM6+F9E5toFk8vzhw4cjsoNkB7gVJImEjD4kbnFJjOXkQibqZ7RPcoOcyEmnkrkJcmJ3yy23qK9hEb7B15+c9OVUhetw+fjjj9VBQyrxRgp9LzRSPOC+++5T/U8Sv8ySGvlc+8/5No3MZ//HP/6h+qPMPw03mReb22OfFEcJnL8pY58ktfJ7u5ETvWjhG4fdjk+m9D/5qie5QCSFtGTOU7ly5dQ/mbsn+1rmLYcTY593coFHKtFKfQ654CT/hFzgkRM4+XvJvHT5vJh63KX/Re95Xyz3PxmH5Pizd+9e7We+mBRPDPd7knOE97gbeGE8O30vxxJb34cnEpPQ5W6BXKGWwiSBxat8FfsyIgd8+UDLVYNAUmVTTtayMwjJdslXccgk68yuPkg7GQw3bNigksyMhPPqmRT18v8OKbmbLieud9xxh6Od3PWURMR3F8cEcvXus88+UwXAwj2o+aPvZZ30cyn2I9+ZLMUrpBJuRqQokTwxIYU4TCYVeOXkIhKkKFTg2CfjmhTDy2zsE5L4ZDT2+X4eKhnTZEyR8SSzK+a+SvayrVJBNCORvHPge7onMNk2sf+VLVtW/RO+u9DyvdHhxtjnnVwckSRi8ODB6l8geRJCvr/Rv6iQicddQf+LvvO+WO5/co4uFxHk+7oDSREuyQPCXTiKsS+yx93s9r2IPIosj8EFku9wlHiTJk0i8ZYqQQvs2LKz5HGs8ePHq6vYbo8QXHjhhaoql1zp9q/sJdVC5asmpMqnnCCGSr5mSP5I8j2Cbie7vqrKUjVaPqBSFU8GOrdt9SWjGX1tSCilt/0rvM2YMUN9NYhcefXnezwvUl+XEwmzZs1S+yknHkOm72Wt70mFx08//VRVpZW7tpmRA748TmRK3/M9YuZPxhVJ4qtXrx6x/idJof+jWDKWyJgi1d/dDvi+MUUeXZXv/ktJSUn/mYxJY8eOVRXks/OInIx9586dU5WUA0nC6BvHZPyVE48BAwZoX78TOPZldHHA69f9uB2b5FF4qTwqV4d91RhN7X+B5PF32dfy/eSRwNjnbeyTO2Gff/659k+q08rXYMn/B05VMPG4G4j+Fx3nfbHe/+R7T1esWOE41skFW6ksLV9lFwmMfZbn8z63467kRZKbValSxSpRokR4+54doWImjz76qD106FBVlezBBx9UFRtvueUW+8SJE5F4S1V9WX4dKYTkb/Xq1apQSbFixewXX3zRHjt2rP3SSy+poho+69atUxPopdDQm2++qSahSzEbKQa1dOnS9Ha+4lFuVUZlYr9M8A8kRVzuuece9fu3a9fOHjFihKoI9uSTT9qXXnqpY11SFVbWL1U7pQCItJXKpv7VRrt27arWJQWeJk+ebC9cuDDLFcpSU1NVwRgpYDBu3Di7R48e6m8mRZYCKyULKTYgFcz8q/RFO6kSK38/qbgaafQ9730vOTlZ9XGpwC2VIwP/HT9+3NFePgdS3M2tX0ajyy+/XI13MobIWNOrVy/1OS9UqFB6JcxwW7lypdqnc+fOdcR37dplX3nllWr/SfXjd99913799dfVOHXkyBHV5o8//lBFGqTIk4w/8veRcVrGmBkzZqSvy1c8Sgo9BcqsAIkU4PMVqpN1jxw5UhWqK1mypGNdMg5JOyni1L9/f1U9UcZIGf98Bg8erNrIePXJJ5+o6opZrYrcu3dvNfZLlWr5+0gxLOm38vtOmjRJa29S/5OCX+3bt7fffvttVbRMqlC6FVUMJ8a+7FWEz+yzY9pxl/4X3ed9sdz/ZHy+5ppr1PFXjhNyrClTpow6zkjxwkhg7LM99z3JB2vXrq3OP3w5mORkUvlYzi3C3fcikthKtcsbbrhBfYWDVGSsUKGC/Z///CeiJwdS8a148eKuFX0lcZVy6lL1Uk4wpQpX4FeLSHXgxo0bqyRYTmTq1aunKpv6CyWxFVINVU505eeSbF1yySV2tWrV1ElV4NcfjR8/XiWYvnay3vnz56f/XE5EpTqo7FvZFv9KeV472eHDh1W1VDnplY4lX+2R0d9HBugSJUqoE0FTyD6Vv/N9992XI+9H3/Pe93wVdDP6t337dq0i7cMPP2ybQhKn6tWrq8+uVKiVA6uc2Pzyyy8Rfd8qVapoX/UipFKxnCRddtllakyRC3b/+te/HNXbpYJu69at08dHuegVWII/1MRWyIFMxju5eCbj1k033WQ///zz6ivV/EmiKid30k4qOst2yEmcj1z0eOihh9R2yrb49zevie28efPshg0bqrFPjk2yLkkA/U8UTe1/8jeTfSb7WI5hNWvWtKdOnRrR92Tsi0xiYeJxl/4X3ed9sd7/pOq7HMfk2CHn8U2bNlVfpxQpjH22574nF6Llq1PlPETOiyRXk5zsp59+ikjfy7Gv+8kJ8j1WkqQFfscsQidfwSEnmoEnoaDvRZqU6Jer1OEu1R+LPvzwQ3XS47sTi+yj/3nDcTf8OO56R/+j/+UW+l50jn059nU/OUHmEckE+SlTpuT2psSMQYMGqQrKgc/Aw4m+F34DBw5Uc2cyK6qB/0fmkkvRllGjRrFL6H85irEv/Dju0v9yE/3PG8a+6Ox7cZLdhnWrAAAAAADIQTF1xxYAAAAAkPeQ2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBo+bw2jIuLi+yWwDg5VXeMvofc6nv0P7hh7ENuYexDbmLsQ7T3Pe7YAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADBavtzeAABAZDRo0ECL5c+fX4s1a9ZMiz355JOO5bi4OK2NbduetiM1NdWxfPToUa3N22+/rcUmTJjgaf0AAOQFF1yg35OsUaNGRN+zffv2juX4+HitzeHDh7XYZ599psWWLVtmRRJ3bAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNHibI/VP9wKhyBv81o4Jrvoe8itvmda/2vXrp1jedKkSWH7fdxel5aWpsUGDhyoxRYvXhy0eMSpU6csUzD2Idb7nmljH3IGY1/0HefFJ598YkWjNWvWaLGqVatGtO9xxxYAAAAAYDQSWwAAAACA0UhsAQAAAABGy2flAcOGDdNi3bt312KrVq3SYqmpqY7lUqVKBZ0/Jn7++Wct9t133zmW9+7dq7U5f/68FgMAL6pVq+ZY3rVrl9amTJkyYduZbvNpBwwYELb1A1nRr18/LdapUyfHcsmSJdmpiIh69eoFnQ+ZnJzsaRxt0aKFFmvSpIljee7cuSFuKUxWuXLlkF535MgRLZaQkKDFNm/erMUOHTrkWJ41a5bWZs+ePVps3rx5Vk7jji0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADBanO3xG29N+qLuBg0aOJbHjh2rtXn22We12N9//63FWrZs6VguX7681qZ69eqetisxMdGxvHTpUq1N+/bttdjOnTutaMQXdee+mjVrarGyZctqsdatWwddV5s2bTwVXnP77MRq3zNt7AtUpEgRLda4cWMtNnXq1KDrOnPmjBZr2rSpFluwYIEV6xj7wmf+/PlBj4P79+/X2lxyySVabPv27VrsxIkTQQtAmoSxLzrUrVtXi3399ddBx+Dly5drbWrUqKHFVq5cGbQ4VWDfzgmMfTkrsPidGDVqlBZzGyNbBuQvbgVrCxQooMUOHDigxU6ePGnlNq99jzu2AAAAAACjkdgCAAAAAIxGYgsAAAAAMBqJLQAAAADAaPmsGPTYY485lnfv3q21+fzzzz2ta/bs2WHbrsCiO9u2bdPaHD58OGzvh+hUpkwZT8WdkpKSghZ3irSePXsGbRMNxaTg7vjx41rs1ltvDWl3LVy4ME8WikJoChcurMUefvhhLVaoUCEtdvbs2ZCKqhQtWlSL5UaBHcSW4sWLa7GhQ4d6KtaXmpoatFCUm59++kmL0Zdj3yOPPBK0+K1bQcuEhAQt9sILLwTts8uWLbNiDXdsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0eJs27Y9NXSZrByt1q9f71hevny51qZjx445uEWxyWPXybZo7Xtt27YNWgSqdOnSQYtCZWTnzp2O5V27dmltUlJSPK1r2rRpjuWlS5dqbZYsWRLStpYtWzbotpva96K5/3lRoUIFT0VJEhMTg66rQ4cOWmzSpElWXpTXxz43BQsWdCxPmTJFa9OkSRMtVqpUqZCKKD733HNabPDgwVps7969Qd/PJIx9kXfRRRc5lidOnKi1ad68uRZbsWJF0IJPd911V9A+KipVqqTF0tLSrNzG2Bc+bkVDhw0bFvQcMlSnT5/2lAtNnjzZikZe+x53bAEAAAAARiOxBQAAAAAYjcQWAAAAAGC0fJbhAuf1uMXWrVuXg1uEvKxmzZqO5TJlyniaf+o2X2zq1KlWTs4PDnXub6Tn0yJ0jRs31mJFihTxNHdlzpw5mc7TBvx169bNsdygQQNtB913330hzad147Z+N4xPyMyFF16oxR599NGg82nnz5+vxWbMmBF0zuSOHTu0Nu3bt4/K+bQInzp16mgxt7nb8fHxQde1YMECLXbkyBEtdt111zmWq1SporWZMGGCFvvtt9+0mFutomjFHVsAAAAAgNFIbAEAAAAARiOxBQAAAAAYjcQWAAAAAGA044tHuX3ZesmSJTP9sm0gHNyKO0Wy4FN2BBaxGjJkSMiFV5KTk8O2XYgst6JkXo0YMcKxfOrUqTBsEWJB6dKltVjXrl0dyxs3btTazJ49O6T3q1+/vqeYG7eCPkBm/SjwGOdW4MztONioUaOgxYCmT5+utVmyZAl/kBjSuXNnLTZ69GhPhcuOHTvmWB4zZozW5uWXX9Zi586d02KJiYmO5UWLFmlt3ApKXXPNNVqM4lEAAAAAAOQQHkUGAAAAABiNxBYAAAAAYDQSWwAAAACA0YwvHrVt2zYt9tVXXwUtoBJYGEUcOHAgbNs1ePBgLTZ//vxMl4FI6d69e6bFpDKydOlSLUbxKHPYth3yawOLSqxYsUJrc+TIES12/vz5kN8TZnjiiSe02FVXXRW2wmWB/vnPf3oqvLJv3z4tNn78+LBtB8zWsWNHLfbaa69psTNnzgQt/LNnzx4t1qVLFy32559/OpYnTpzoeXthhlq1ajmWR40aFVKhKNG0aVPH8o8//hjydqWlpTmWt2zZ4ql4VJs2bbTY5MmTLVNwxxYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABjN+OJRbrZu3epYLlSokNamefPmWuz9998P2zZUq1ZNi8XFxTmWKR6F7KpZs6YW69mzp6diAF5Mnz49pNchOtx6661abO3atVqsZMmSWmzAgAGO5YEDB2ptpkyZosW++eaboNuVmpqqxRYvXhz0dYgOF198sRZbs2aNY3nWrFkhr79YsWJBj6fLli3TYkePHtViBw8eDHk7YK5KlSp5KhRVrlw5LTZmzBjH8quvvupp7IuPj9diY8eOdSyvW7cuk62GiQIL5eXPnz9oETHRpEkTLbZ8+fIwb13ewx1bAAAAAIDRSGwBAAAAAEYjsQUAAAAAGC0m59iuWrUqaBu3ebehcpu/6Da/Y9y4cWF7T8SWMmXKaLHu3bs7lpOSkrQ2brFwSklJiej6EVlu83ratWsXdE6ZqFy5cqY1AjJal1vMixdeeEGLDR48OKR1IbIKFiyoxbZs2eJYPnXqVMjrf/nllx3Ll112mdZm27ZtnrYLeXPed58+fTzNp125cqUWe+ONNxzLd999t6c6LYH1XcSIESMy2WrEgo0bNzqWExMTtTaPP/64p1oT4ZQYsB0NGjTw9LovvvjCMhl3bAEAAAAARiOxBQAAAAAYjcQWAAAAAGA0ElsAAAAAgNFisnjUl19+6VhevHixp0IlH3zwgRY7ceJE0Pdr2LChFitRooQW2759e9B1IfbVrFlTi02dOtVTQamc9vvvv2uxnTt3Opbbtm2rtVm6dGlEtwuhcxsPb7vtNi2WkJDgWG7fvr3WxrZtT0VVAgu7VKtWTWvTr18/T/1vypQpWgw5a9++fVqsfv36QcevkydPajG3fhUYcyu8MmjQIC22Y8eOTLYasaxOnTpBi3q6nc+NHj1ai7Vu3TromOZWgPTs2bNarGLFikGLniHnBRZWuvnmm7U29erV81RsbPbs2UELl50+fdrKaSVLlsz0OJyRAwcOWCbjji0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADBanO1W/cOtYVycZaqnn35aiw0fPtxT8ZLXX389aMGAJUuWaLFbbrlFiyUlJcVUgR2PXSfbTO57bgVUPv3006B9IzsCizuJ5ORkx/Lu3bu1NqVKldJiPXr0CKmoVc+ePYNugwl9z/T+Fy2aNWvmWP7iiy887We3MbJWrVpWbsvrY9+DDz6oxSZNmuRY3rNnj9bm3LlzWqxs2bJarG/fvpkuiw0bNmix1NRULdakSRMrljD2WVaxYsW0/TJ58mTHcoMGDSL6d/jjjz+02P79+7XY5s2bHcsdO3b0VFQtWsXK2Dd37tygRWBDtWbNGi3mNobNmjVLi506dSqk9yxdurQW+/bbbx3LFSpU0NocOXJEi91xxx1abNOmTZYpfY87tgAAAAAAo5HYAgAAAACMRmILAAAAADAaiS0AAAAAwGh5oniUm++//16L1a5dO+gE8/z582tt0tLStFiLFi2CFj2heJQ30dr33IooDR06NOiE/kgXinKb+O/WLlSBBaW8/j5t27YN2zZQQCU6FC5c2FORjM6dOzuWixQp4ulz/t5772mxLl26WLktVgqohFNg8ZWSJUuGvK6ZM2c6lq+//nqtzQ8//KDF5s2bp8UoHhV7/a9mzZqeinhG0tixY7XYk08+qcUSExODni+aJFbGvsDfw+33Wrt2rRZzK8AUHx8f0ja4jWGtWrUKWtzpggv0e5KjR4/WYo8//njQbahXr56n/CgaUDwKAAAAAJAn8CgyAAAAAMBoJLYAAAAAAKPls/Koli1barFOnTppsXLlyjmW+/Xrp7V55ZVXwrx1MGE+bevWrYPOZXWbCxSqlJSUoPO2c0JycnKmy4hNpUqV0mLjxo3TYo0aNQpp/dOmTdNi3bp1C2ldyHnz58+P2LorVarkqd306dMjtg2IHgcPHtRiY8aMCTovu1evXlrM7Rj99ttvO5Z3796ttRkxYoSnbTV9Tm1eNnDgQE/zTxs3buxYfuCBBzwdF+vUqaPFKleu7FhetmyZp7ncXubTrlu3ToutXr3aijXcsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEbLs8Wj3L70eMiQISGta82aNWHYIpjGrZhToJ49e4a8/mHDhjmWn3322ZDXBWRV9erVHctTp07V2pQvXz5shaLcCm4A4vjx4552xNGjR9lhecCvv/6qxbp27Rr0dQkJCVqsT58+QV83d+5cLbZ+/fqgr0P0mjdvnmO5YcOGQYtCicmTJ2uxCRMmOJYXLlyotUlNTfW0XbVr13Ysd+jQQWvTuXNnT+taH9BH3X6fWBwzuWMLAAAAADAaiS0AAAAAwGgktgAAAAAAo5HYAgAAAACMlmeLRwEZKVOmjBbbuXOnFktKSgqpAJlb0aldu3ZpseHDh/NHQoaFw7Zt26bFFi1a5Fg+ePCg1qZ06dJarEuXLlqsV69ejuUCBQpobWzb1mJ///23Fps0aZJjuVu3blobICNNmjTRYnFxcewwZEnbtm21WIMGDbTYsmXLHMs9evRgT8eY1157zbFct25drc1DDz2kxQ4dOqTF3nnnHcdyiRIlQt6uN998M2ibY8eOabFRo0Zpsf79+zuWT5w4YeUF3LEFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGo3hUGLgVH4A5haHcikAF6t69uxbr2bNn0Ne5tUlOTg76OsBf69attR1So0YNLbZ///6gRSYuvfRSTzEvNmzYoMVefPFFLTZz5syQ1g9kxK1wGeATHx+v7Yz27dtrsfz582uxBQsWOJbT0tLYsTEmsEDYyJEjPZ2/uRUSi2RxsRUrVmixZ555RostXbo0YttgGu7YAgAAAACMRmILAAAAADAaiS0AAAAAwGjMsQ2D1NTUoHPdxOrVq8PxdsiGoUOHarGUlBTHcps2bbQ2pUuX1mLMn0VO+eqrr7RY1apVtdgVV1zhWL788stDfs9Tp045lvv06aO1+eCDD7TYvn37Qn5PwKsTJ05osYULF7IDobRr107bE3Xr1g0611L079+fvZjH9OrVy9N48tJLL2mxKlWqOJYLFSqktdm0aZMWO3r0aNC+FzjfW5w9e1aL4f/jji0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaxaPC4Ndff9VixYoV02I33nijY3nlypXheHtkgVthKLdYoKSkJC3GF2IjpwwYMECLbd26VYsNGjTIsVy+fPmQi1P17t3bsfzLL794WheQEwoXLqzF6tevr8WmTZvGHyQPKFq0qGO5a9euWpt8+fRT3mHDhmmxv/76K8xbh2hn27YWmz17tqdYxYoVHcsJCQlaG4rH5hzu2AIAAAAAjEZiCwAAAAAwGoktAAAAAMBoJLYAAAAAAKNRPCpCzpw5o8WOHz8eqbeDR26FRFJSUhzL06dP19rs3LmTfYyo4tZP3WJALIqLi8vtTUAU6dmzp2O5WrVqWpuRI0dqsc8++yyi24XYt2XLltzeBPjhji0AAAAAwGgktgAAAAAAo5HYAgAAAACMRmILAAAAADAaxaPCoFSpUlrsyJEjWmzTpk3heDtkQ9u2bdl/AGCQzZs3a7GVK1dqsYULF+bQFiE3XX311VqsVatWjuU5c+Zobfr27avFzp8/H+atA5CbuGMLAAAAADAaiS0AAAAAwGgktgAAAAAAo8XZtm17asiXoSOAx66TbfQ95Fbfo//BDWMfcgtjH3ITYx+ive9xxxYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYDQSWwAAAACA0UhsAQAAAABGI7EFAAAAABiNxBYAAAAAYLQ427bt3N4IAAAAAABCxR1bAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIDRSGwBAAAAAEYjsQUAAAAAGI3EFgAAAABgNBJbAAAAAIBlsv8D5umEZtsK050AAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1200x1200 with 30 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_wrong_images(convnet)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### The Structure of ConvNets in More Detail"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "A very simple case:\n",
    "* input is a 28 &times; 28 grayscale image\n",
    "* receptive field size is 3 &times; 3\n",
    "* convolution layer has only 1 filter (it learns only 1 feature)\n",
    "\n",
    "The 3 &times; 3 receptive field implies that:\n",
    "* the feature map is 2 pixels smaller than the input image in each dimension\n",
    "* the filter consists of 9 weights + 1 bias"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from keras.layers import Conv2D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_3\"\n",
      "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
      "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
      "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
      "│ conv2d (Conv2D)                 │ (None, 26, 26, 1)      │            10 │\n",
      "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
      " Total params: 10 (40.00 B)\n",
      " Trainable params: 10 (40.00 B)\n",
      " Non-trainable params: 0 (0.00 B)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cnn = Sequential()\n",
    "cnn.add(Input(shape=(28,28,1)))\n",
    "cnn.add(Conv2D(1, (3,3), activation='relu'))\n",
    "cnn.summary(print_fn=print)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "The size of the input image doesn't matter.  All units in the convolution layer share the same weights and bias (the \"filter\") for learning a feature, no matter where the feature occurs in the input image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_4\"\n",
      "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
      "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
      "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
      "│ conv2d_1 (Conv2D)               │ (None, 998, 998, 1)    │            10 │\n",
      "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
      " Total params: 10 (40.00 B)\n",
      " Trainable params: 10 (40.00 B)\n",
      " Non-trainable params: 0 (0.00 B)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cnn = Sequential()\n",
    "cnn.add(Input(shape=(1000,1000,1)))  # changed 28x28 to 1000x1000\n",
    "cnn.add(Conv2D(1, (3,3), activation='relu'))\n",
    "cnn.summary(print_fn=print)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "If the receptive field size is 5 &times; 5, the \"kernel\" consists of 25 weights + 1 bias, and the feature map is 4 pixels smaller than the input image in each dimension."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_5\"\n",
      "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
      "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
      "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
      "│ conv2d_2 (Conv2D)               │ (None, 24, 24, 1)      │            26 │\n",
      "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
      " Total params: 26 (104.00 B)\n",
      " Trainable params: 26 (104.00 B)\n",
      " Non-trainable params: 0 (0.00 B)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cnn = Sequential()\n",
    "cnn.add(Input(shape=(28,28,1)))\n",
    "cnn.add(Conv2D(1, (5,5), activation='relu'))  # changed 3x3 kernel to 5x5\n",
    "cnn.summary(print_fn=print)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "With RGB images, the input depth is 3, so a 3 x 3 kernel consists of 27 (9 &times; 3) weights + 1 bias"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_7\"\n",
      "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
      "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
      "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
      "│ conv2d_4 (Conv2D)               │ (None, 26, 26, 1)      │            28 │\n",
      "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
      " Total params: 28 (112.00 B)\n",
      " Trainable params: 28 (112.00 B)\n",
      " Non-trainable params: 0 (0.00 B)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cnn = Sequential()\n",
    "cnn.add(Input(shape=(28,28,3)))\n",
    "cnn.add(Conv2D(1, (3,3), activation='relu'))\n",
    "cnn.summary(print_fn=print)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "If we add another filter, the Conv2D layer will learn two independent features, doubling the number of parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_8\"\n",
      "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
      "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
      "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
      "│ conv2d_5 (Conv2D)               │ (None, 26, 26, 2)      │            56 │\n",
      "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
      " Total params: 56 (224.00 B)\n",
      " Trainable params: 56 (224.00 B)\n",
      " Non-trainable params: 0 (0.00 B)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cnn = Sequential()\n",
    "cnn.add(Input(shape=(28,28,3)))\n",
    "cnn.add(Conv2D(2, (3,3), activation='relu'))  # changed 1 filter to 2\n",
    "cnn.summary(print_fn=print)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "We can include as many filters as we like in a layer.  Each *N* x *N* filter has its own kernel (set of weights + bias), which is shared by all the units in the layer.  Each unit receives inputs from its specific *N* &times; *N* patch of the input image, including at all depths of the input image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_9\"\n",
      "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
      "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
      "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
      "│ conv2d_6 (Conv2D)               │ (None, 26, 26, 10)     │           280 │\n",
      "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
      " Total params: 280 (1.09 KB)\n",
      " Trainable params: 280 (1.09 KB)\n",
      " Non-trainable params: 0 (0.00 B)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cnn = Sequential()\n",
    "cnn.add(Input(shape=(28,28,3)))\n",
    "cnn.add(Conv2D(10, (3,3), activation='relu'))  # changed 2 filters to 10\n",
    "cnn.summary(print_fn=print)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}