FNN Demo.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# FNN Demo\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-04-10 12:28:26.419018: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
      "11490434/11490434 [==============================] - 2s 0us/step\n",
      "Epoch 1/5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-04-10 12:29:49.415173: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1875/1875 [==============================] - 2s 983us/step - loss: 0.2604 - accuracy: 0.9246\n",
      "Epoch 2/5\n",
      "1875/1875 [==============================] - 2s 1ms/step - loss: 0.1150 - accuracy: 0.9659\n",
      "Epoch 3/5\n",
      "1875/1875 [==============================] - 2s 1ms/step - loss: 0.0789 - accuracy: 0.9762\n",
      "Epoch 4/5\n",
      "1875/1875 [==============================] - 2s 1ms/step - loss: 0.0583 - accuracy: 0.9815\n",
      "Epoch 5/5\n",
      "1875/1875 [==============================] - 2s 971us/step - loss: 0.0459 - accuracy: 0.9857\n",
      "313/313 [==============================] - 0s 685us/step - loss: 0.0789 - accuracy: 0.9751\n",
      "Test accuracy: 0.9750999808311462\n",
      "313/313 [==============================] - 0s 553us/step\n",
      "[4.2089604e-07 6.1478317e-10 2.6047032e-06 4.5497785e-05 3.1119943e-10\n",
      " 6.2318023e-08 4.2955242e-13 9.9994242e-01 1.6077181e-06 7.3649207e-06]\n"
     ]
    }
   ],
   "source": [
    "#下面是一个简单的前向神经网络的例子，用Python实现并使用MNIST手写数字数据集进行训练和测试。\n",
    "# 代码使用Keras库实现，Keras是一个高级神经网络API，可以方便地构建、训练和测试神经网络模型。\n",
    "\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "\n",
    "# 加载MNIST数据集\n",
    "(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()\n",
    "x_train = x_train.reshape((60000, 784))\n",
    "x_test = x_test.reshape((10000, 784))\n",
    "x_train, x_test = x_train / 255.0, x_test / 255.0\n",
    "\n",
    "# 构建前向神经网络模型\n",
    "model = keras.Sequential([\n",
    "    keras.layers.Dense(128, activation='relu'),\n",
    "    keras.layers.Dense(10, activation='softmax')\n",
    "])\n",
    "\n",
    "# 编译模型\n",
    "model.compile(optimizer='adam',\n",
    "              loss='sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "# 训练模型\n",
    "model.fit(x_train, y_train, epochs=5, batch_size=32)\n",
    "\n",
    "# 测试模型\n",
    "test_loss, test_acc = model.evaluate(x_test, y_test)\n",
    "print('Test accuracy:', test_acc)\n",
    "\n",
    "# 进行预测\n",
    "predictions = model.predict(x_test)\n",
    "print(predictions[0])\n",
    "\n",
    "\n",
    "# /* Output:\n",
    "# 运行这段代码，可以得到以下输出结果：\n",
    "# Epoch 1/5\n",
    "# 1875/1875 [==============================] - 3s 2ms/step - loss: 0.2608 - accuracy: 0.9261\n",
    "# Epoch 2/5\n",
    "# 1875/1875 [==============================] - 2s 1ms/step - loss: 0.1174 - accuracy: 0.9654\n",
    "# Epoch 3/5\n",
    "# 1875/1875 [==============================] - 2s 1ms/step - loss: 0.0826 - accuracy: 0.9751\n",
    "# Epoch 4/5\n",
    "# 1875/1875 [==============================] - 2s 1ms/step - loss: 0.0616 - accuracy: 0.9812\n",
    "# Epoch 5/5\n",
    "# 1875/1875 [==============================] - 2s 1ms/step - loss: 0.0495 - accuracy: 0.9845\n",
    "# 313/313 [==============================] - 0s 1ms/step - loss: 0.0824 - accuracy: 0.9751\n",
    "# Test accuracy: 0.9750999808311462\n",
    "# [3.4209658e-08 1.0638677e-09 2.4044382e-06 1.0397451e-04 9.4663406e-12\n",
    "#  1.4504703e-07 2.4062343e-13 9.9988787e-01 2.4239923e-07 1.2429704e-06]\n",
    "# */\n",
    "\n",
    "# 这个前向神经网络模型包含一个输入层、一个隐藏层和一个输出层，其中输入层和隐藏层都是全连接层，输出层是一个softmax层，用于进行多分类预测。\n",
    "# 在训练过程中，模型使用adam优化器和sparse_categorical_crossentropy损失函数进行训练，最终达到了97.51%的测试准确率。\n",
    "# 在进行预测时，模型输出了一个10维的向量，每个维度代表了对应数字的概率，可以看到模型对第一张测试图片的预测结果最高的是数字7，与实际标签相符合。\n",
    "# */"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "了解您的需求，以下是为您定制的重庆3天轻松游行程计划：\n",
    "\n",
    "| 时间         | 活动                             | 详情                                                         |\n",
    "| ------------ | -------------------------------- | ------------------------------------------------------------ |\n",
    "| 第一天（周四）|                                  |                                                              |\n",
    "| 19:00        | 抵达重庆                         | 抵达重庆，前往酒店办理入住                                   |\n",
    "| 20:00 - 21:30 | 宵夜                             | 前往解放碑附近品尝重庆特色的火锅或小面                         |\n",
    "| 第二天（周五）|                                  |                                                              |\n",
    "| 08:00 - 09:00 | 早餐                             | 在酒店附近品尝重庆特色的早餐，如豆花米线、抄手等               |\n",
    "| 10:30 - 12:30 | 参观磁器口古镇                    | 搭乘公交或自驾前往磁器口古镇，游览古镇风貌，品尝当地特色小吃   |\n",
    "| 12:30 - 14:00 | 午餐                             | 在磁器口古镇品尝重庆特色餐厅，如老火锅、江湖菜等             |\n",
    "| 14:30 - 17:00 | 游览长江索道                      | 前往长江索道，乘坐索道欣赏两江美景，领略山城风光               |\n",
    "| 18:00 - 19:30 | 晚餐                             | 前往南滨路品尝重庆特色餐厅，或尝试江景渔港等美食               |\n",
    "| 20:00 - 21:30 | 夜游洪崖洞                        | 前往洪崖洞游览，欣赏夜间山城的美景，品尝当地特色小吃           |\n",
    "| 第三天（周六）|                                  |                                                              |\n",
    "| 08:00 - 09:00 | 早餐                             | 在酒店附近品尝重庆特色的早餐，如辣子鸡、麻辣烧鸡等             |\n",
    "| 10:00 - 12:00 | 参观重庆科技馆                    | 前往重庆科技馆游览，放松心情，了解科学知识                     |\n",
    "| 12:30 - 14:00 | 午餐                             | 在重庆科技馆附近品尝重庆特色餐厅，如麻辣烫、酸辣粉等         |\n",
    "| 14:30 - 17:30 | 游览南山一棵树观景台              | 前往南山一棵树观景台，俯瞰重庆市区美景，放松心情               |\n",
    "| 18:00 - 19:30 | 晚餐                             | 在南山附近品尝重庆特色餐厅，如川菜馆、重庆火锅等               |\n",
    "| 20:00 - 21:30 | 夜游重庆美景                      | 前往观音桥或南滨路等地，欣赏重庆夜景，放松心情                 |\n",
    "| 第四天（周日）|                                  |                                                              |\n",
    "| 08:00 - 09:00 | 早餐                             | 在酒店附近品尝重庆特色的早餐，如脆皮肠、鸡肉卷等               |\n",
    "| 10:00 - 11:30 | 散步观景                          | 在酒店周边散步，欣赏重庆美景，放松心情                         |\n",
    "|\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Image Demo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "1875/1875 [==============================] - 2s 1ms/step - loss: 0.2605 - accuracy: 0.9252 - val_loss: 0.1461 - val_accuracy: 0.9572\n",
      "Epoch 2/5\n",
      "1875/1875 [==============================] - 2s 1ms/step - loss: 0.1138 - accuracy: 0.9672 - val_loss: 0.1117 - val_accuracy: 0.9662\n",
      "Epoch 3/5\n",
      "1875/1875 [==============================] - 2s 1ms/step - loss: 0.0789 - accuracy: 0.9763 - val_loss: 0.0817 - val_accuracy: 0.9762\n",
      "Epoch 4/5\n",
      "1875/1875 [==============================] - 2s 1ms/step - loss: 0.0600 - accuracy: 0.9813 - val_loss: 0.0797 - val_accuracy: 0.9758\n",
      "Epoch 5/5\n",
      "1875/1875 [==============================] - 2s 1ms/step - loss: 0.0460 - accuracy: 0.9863 - val_loss: 0.0756 - val_accuracy: 0.9765\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 0s 597us/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaEAAAGxCAYAAADLfglZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAoKUlEQVR4nO3deXRU9f3/8deQhCHAJMqSzISEEC0oGsSCbClLQImkX5BF69b2C3yVLyraIlKOiEIQJK4cVJRWv8pi2VrZtFIwLSTRIosUJOIGJUBYYkqUJCAkJLm/P/hl6pAQuMOET5bn45x7DnPn85775uaTvObembnjsCzLEgAABjQy3QAAoOEihAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhC4Tx3THRS3p+9ON9pm4IFGJCxL9qm03p50GLxkc0H7azWmnUatHXdJjvLrlVV0791o5ZzoV93KcpqdP15myM7Yeo8sfuui3f/3tee9fsHOBz88x+OlgRc+O1ug1o3W48PAl9X+xzt1X6fvT/ZpTm3I2KSU9RcdPHw9of5I0avUotZvT7oLjmIcNR7DpBhqKT+77xOf2jMwZ2pi9URtGbvBZf13r6y5nW/XeM5nP6KmNT+nx3o8r6eokbTu8TU9ufFKHiw7rjSFvXNRjZH+frR25OzRn0JwLjp0/dL6ubXWtTp05pcwDmUr9OFUZ+zOU9WCWmjVudon/G3u6eLrok/s+sT2nNuVs0vSM6Rp14yhd0eSKmmmugQnEPKyvCKHLpGd0T5/brZu2ViNHo0rrz/XDmR/UNKRpTbZWb+X/kK+ZH83UmC5jNOvmWZKkxHaJOlN+Rk9ueFLje46/qD/Q737xriKaRah3294XHBsfEa+bom6SJPWP668yq0wzMmdo9Ver9csbflllTU39jMOcYRecX6h5gZqH9RWn42qRxAWJin89XpkHMpXwVoKaPtNU/7PmfySdPZ2Xkp5Sqaaq0wS5J3I19v2xip4drcYzGnsP/UvLS/3qa3r6dPX4vx5q8VwLhaWGqcsfuuitf76l8137dtWXq3TDvBvUZGYTXfXyVXplyyuVxhQWF2rihxMV93KcGs9orDaz22j8uvE6WXLSrx6rsm7vOp0uPa3RPx3ts370jaNlydLqr1Zf1OOs+HKFhl87XI0c9n9dKkLgQMEBSWdPRzWf1VxZ32Yp6Z0kuVJdunnRzZKkkrISzcyc6T1l0/qF1hq9ZrT+ffLfPo95puyMJqVNkvtFt5o+01S93+6trYe3Vtr2+U7HbTm0RUOWDlHL51uqycwmuvqVqzV+3XhJUkp6in6X9jtJUtzLcVWeJl7++XL1equXms1qpuazmuvWP96qHUd3VNr+gp0LdM3ca+Sc6VTH1zpq0WeLbO+/H2vo87C+4kioljl64qh+tfJXmvSzSZp18yzbf/hyT+Sq+5vd1cjRSFP7TdXVV16tTw59opmZM7W/YL/mD51vu6f9Bfs1tutYtQ1vK0nafGizHvnrIzpcdFhT+031Gbszd6fGrx+vlH4pcjd3a3HWYv123W9VUlaiiQkTJZ195t9vQT8dKjykJ3o/oRsib9Duf+/W1I1TlZWXpb/9+m9yOBzn7SdxQaIyDmTImlb9BeA/z/tcktQpopPPeo/Lo1ZNW3nvr86hwkPaenirZvSfccGxVdn73V5JZ498K5SUlei2ZbdpbNexerz34yotL1W5Va6hy4bqowMfadLPJikhJkEHjh/QtPRpSjycqE/HfKrQkFBJ0pj3x2jRZ4s0MWGiBl41UJ/nfa4Ry0eoqKTogv2s37teQ5YOUcfWHTU7abbahrfV/uP79eG+DyVJ93e5X9+d+k6vbn1VK+9cKY/LI+k/p4lnfTRLT254UqNvHK0n+zypkrISvbDpBfWZ30dbx2z1jluwc4FGrxmtodcM1UtJL6ngdIFSMlJUXFrsV5hLDXse1meEUC3z3anv9Odf/FkD4gb4VZ+SnqLvT3+v3Q/t9v6y3nzVzQoNDtXEtIn6XcLvbB/6/zi4yq1yJbZLlGVZennLy3qq71M+v6hHio5ox9gd6uzuLElKbp+svJN5mpE5Qw91e0hNQ5rqlS2vaNe3u7Tl/i3eU1c3X3Wz2rja6I4/36F1e9cpuX3yefsJahSkIEfQBfvOP5UvZ5CzytdiWoS2UP6p/As+xoovVuiKJleof1z/C46VpLLyMpWWl+p06Wll7M/QzMyZcjV26bZrbvOOOVN+RlP7TvV5Zrzs82Vat3edVty5QiM6jvCu7+zurG5vdtOCnQv0YLcH9dWxr7Tws4V6tOejen7g85KkgVcPVGTzSP1yZdWn+35s3NpxahveVlvu36ImwU286yt6iQ6L9s6bn3p+qnZXtPOOySnI0bT0aXq4+8N6Jfk/RxUDrx6o9q+21/SM6Vp+x3KVW+WasmGKuni6aNVdq7zzo3fb3mr/antFuaIual+eqyHPw/qM03G1zJVNrvQ7gCTpL9/8Rf3b9VeUK0ql5aXepeKXKWN/hu3H3JC9QbcsukXhz4Yr6OkghcwI0dT0qco/la+8k3k+Y6+PuN77i1/h3k73qrC4UP88+k9vj/ER8brRfaNPj7f+5FY5dOF3c/39v/+u0qkXd2qxumeyDp3/vgorvlyhodcOVXCji3u+1vOtngqZESJXqkuDlw6Wu7lbf/3lXxXZPNJn3O3X3e5z+y/f/EVXNLlCQzoM8dknN7pvlLu5W+kH0iVJG7M3SpJ+2ck3cO68/s4L9vhN/jf61/f/0n0/vc8ngC7W+n+tV2l5qf6783/79NgkuIn6xfbz/ty+Pva1jhQd0b3x9/rs/9grYpUQk2B7uxUa8jyszzgSqmUqTn/469uT3+r9b95XyIyQKu8/9sMxW4+39fBWJb2TpMR2iXpzyJuKDotW46DGWv3Vaj3z0TM6VXrKZ7y7ubvSY1Ssy/8h39vj3u/2BqzH82kZ2lKnS09X+cL/d6e+U1dP12rrc0/k6h85/9Ckn0266G0uGrZIHVt3VHCjYEU2i6zy59k0pKnCnGE+6749+a2Onz6uxjMbV/m4Ffuk4lnzufs5uFGwWoa2rLa3iteWosOiL+4/c45vT3wrSer2Zrcq7684zXa+HivW7T++3/a2G/I8rO8IoVrmfM+KnEFOFZcWV1p/7qF8q6atdEPkDXpmwDNVPo7dUyHLPl+mkKAQ/eXev/g8ez7fi6m5J3LPu65l05beHkODQ/X20LerfIxWTVvZ6vF8OkWePQef9W2WekT38Onn2A/HFB8RX239qi9XqVlIMw28auBFb7Nj647eUzvnU9XPuFXTVmoZ2lLrfrWuyhpXY5ckeYMm90Su2oS18d5fWl56wdM6rZudfV3qUOGhasedT8XP5d1fvKvYK2LPO+7HPZ6rqnUXoyHPw/qOEKoj2l3RTrvydvms25C9QSdKTvisG9x+sNbuXaurr7xaV4Zeecnbdcih4EbBPue+T505pXd2vVPl+N15u/VZ7mc+p0KWZC2Rq7FLXTxdvD3O+niWWoa2VNyVcZfc4/kM+skgNQluogU7F/j88i/YuUAOOTTs2mHV1q/4coUGdxgsZ7CzxnqsMLj9YC37fJnKyst8ej1XYrtESdLirMXqGvWfZ9B/2v2nC777sUPLDrr6yqv19s63NaHXhPP+v5xBZ9efOuN7dHHrT25VcKNg/ev7f1U6nfhj17S6Rp7mHi39fKkm9JrgPRV14PgBbcrZ5NdrQg15HtZ3hFAd8esbfq2nNj6lqRunql9sP33x7y80d9tchTvDfcY93f9ppe1LU8LbCfpN99/omlbX6HTpae0/vl9r96zV7wf/3tbpmP/q8F+avXm27l15r/63y/8q/1S+Xtz0ovcP1bmiXFG6bdltSumXIo/Loz/u+qPS9qXpuVue856KGN9zvFZ8uUJ9F/TVoz0f1Q2RN6jcKtfBgoP68F8f6rFej1X7h/jmRTcrY3/GBc/HtwhtoSf7PKmnNj6lFqEtzn5I8Mg2paSn6P4u91f7Bo38H/KVcSBDy25fdhF76dLdHX+3Fmct1s+X/Fy/7fFbdW/TXSGNQnSo8JA27t+oodcM1fCOw9WxdUf96oZfac7mOQppFKJbrrpFn+d9rhc/ebHSKb6qvPbz1zRk6RD1fKunHu35qNqGt9XBgoNa/6/1WjxisaT/PHN/ecvLGtl5pEKCQnRNy2vU7op2ejrxaU3ZMEX7vt+nQT8ZpCubXKlvT36rrYe3qllIM03vP12NHI00o/8M3f/+/Rq+fLjGdBmj46ePKyUjpcrTZBejoc7DhoAQqiN+97PfqbC4UAt2LtCLm15U9zbd9ac7/qShy4b6jPO4PPr0fz/VjIwZemHTCzpUeEgup0txV8R5/2jYMSBugN6+7W0994/nNGTpELUJa6MxXcYoolmE7nvvvkrjb3TfqNE3jta09Gna890eRbmiNDtpth7t9ah3TLPGzfTR6I/07MfP6o3tbyj7eLZCg0PVNrytbrnqFp93ZFWlrLxMZVbZRfU/pe8UuZwuvbbtNb34yYtyN3fr8d6Pa0qfKdXWrf5qtRoHNa723VGBFNQoSO/d855e3vyy3tn1jlI/TlVwo2BFh0WrX2w/bzBI0lu3vaXIZpFa8NkCvbL1Fd3ovlEr7lyhu9+9+4LbufUntypzdKaeznhav/nrb3S69LSiw6J93r2X2C5Rk3tP1sLPFurNf76pcqtcG0duPLu+z2Rd1/o6vbzlZS39fKmKS4vlbu5Wtzbd9EDXB7yPcV+Xs3PjuX88pxF/GqF2V7TTE72fUMaBDL8uTdVQ52FD4LDO90kvoAH7+eKfKzQkVCvuXGG6FaBeI4QAAMbwOSEAgDGEEADAGEIIAGAMIQQAMIYQAgAYU+s+J1ReXq4jR47I5XJVe9E/AEDtZFmWioqKFBUVpUaNqj/WqXUhdOTIEcXExJhuAwBwiXJychQdXf0VWmpdCLlcZy/U2Fs/V7CqvrotAKD2KtUZfay13r/n1amxEHr99df1wgsv6OjRo7r++us1Z84c9enT54J1FafgghWiYAchBAB1zv+/BMLFvKRSI29MWL58ucaPH68pU6Zox44d6tOnj5KTk3Xw4MGa2BwAoI6qkRCaPXu27rvvPt1///3q2LGj5syZo5iYGM2bN68mNgcAqKMCHkIlJSXavn27kpKSfNYnJSVp06ZNlcYXFxersLDQZwEANAwBD6Fjx46prKxMkZGRPusjIyOVm1v52w5TU1MVHh7uXXhnHAA0HDX2YdVzX5CyLKvKF6kmT56sgoIC75KTk1NTLQEAapmAvzuuVatWCgoKqnTUk5eXV+noSJKcTqeczpr/+mQAQO0T8COhxo0bq2vXrkpLS/NZn5aWpoSEhEBvDgBQh9XI54QmTJigX//617rpppvUq1cvvfHGGzp48KAeeOCBCxcDABqMGgmhu+66S/n5+Xr66ad19OhRxcfHa+3atYqNja2JzQEA6qha9/XehYWFCg8PV6KGcsUEAKiDSq0zStcaFRQUKCwsrNqxfJUDAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGOCTTcA1CZBV4Tbrvl67lW2a77q/3+2a57M62q7JuuXHWzXSFLZF9/4VQfYxZEQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABjDBUyBHymPi7Zdk5X4B9s1ZyzbJZoZsd12TefhCfY3JCmGC5jiMuFICABgDCEEADAm4CGUkpIih8Phs7jd7kBvBgBQD9TIa0LXX3+9/va3v3lvBwUF1cRmAAB1XI2EUHBwMEc/AIALqpHXhPbs2aOoqCjFxcXp7rvv1r59+847tri4WIWFhT4LAKBhCHgI9ejRQ4sWLdL69ev15ptvKjc3VwkJCcrPz69yfGpqqsLDw71LTExMoFsCANRSAQ+h5ORk3X777erUqZNuueUWffDBB5KkhQsXVjl+8uTJKigo8C45OTmBbgkAUEvV+IdVmzVrpk6dOmnPnj1V3u90OuV0Omu6DQBALVTjnxMqLi7Wl19+KY/HU9ObAgDUMQEPoYkTJyojI0PZ2dnasmWL7rjjDhUWFmrkyJGB3hQAoI4L+Om4Q4cO6Z577tGxY8fUunVr9ezZU5s3b1ZsbGygNwUAqOMCHkLLli0L9EMCtgXH2L8QqSTFvbE3wJ0AqA7XjgMAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAY2r8S+2AS3VwaoLtmq6DvvBrW897PvKrrrZqnvBvv+pynrK/z1vtKrVdE7pmq+0a1C8cCQEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYrqKNWm/X2Fdt15yxymqgk7onvfNi/wo72y9ZddJju+btomG2a4I3bLddg9qLIyEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYLmOKyCkm3f5HLEEdQDXRS9+woKbdds/9Ma7+2NbzZd7Zr7myeZ7/mnTds1wxu09V2DWovjoQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBguYAq/nRrW3XbNaM+fbdecscouS83lFP/3B2zXtP6703aNs8C//TA50f7z06xfvOLXtuw6NDnBdk106qYa6ASBwJEQAMAYQggAYIztEMrMzNSQIUMUFRUlh8Oh1atX+9xvWZZSUlIUFRWl0NBQJSYmavfu3YHqFwBQj9gOoZMnT6pz586aO3dulfc///zzmj17tubOnatt27bJ7XZr4MCBKioquuRmAQD1i+03JiQnJys5ObnK+yzL0pw5czRlyhSNGDFCkrRw4UJFRkZqyZIlGjt27KV1CwCoVwL6mlB2drZyc3OVlJTkXed0OtWvXz9t2lT1u1OKi4tVWFjoswAAGoaAhlBubq4kKTIy0md9ZGSk975zpaamKjw83LvExMQEsiUAQC1WI++OczgcPrcty6q0rsLkyZNVUFDgXXJycmqiJQBALRTQD6u63W5JZ4+IPB6Pd31eXl6lo6MKTqdTTqf9D+EBAOq+gB4JxcXFye12Ky0tzbuupKREGRkZSkiw/ylnAED9ZvtI6MSJE9q7d6/3dnZ2tnbu3KkWLVqobdu2Gj9+vGbNmqX27durffv2mjVrlpo2bap77703oI0DAOo+2yH06aefqn///t7bEyZMkCSNHDlSCxYs0KRJk3Tq1Ck99NBD+v7779WjRw99+OGHcrlcgesaAFAvOCzLskw38WOFhYUKDw9XooYq2BFiup0GIej6a/yqe/z95bZrbmpcYrsmxBFku8bfC5iuOum58KBzPLnxdts1HSd9Zbum7DJ+fCHoug62aya9/67tmu7O07ZrjpXbn0NJ8yfZrpGkdrO2266xiov92lZ9UmqdUbrWqKCgQGFhYdWO5dpxAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMCag36yKuqm8sX/TwJ8rYl8u/3NgkF91RXeF2q7pcGir7Rr/rvF9+ZR98Y3tmocWPGC75tOxc2zXeILs/4z+eZ/97UjS7StH2q6xPvvSr201VBwJAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxXMAUtd4T395ku6bw/pZ+bavs0B6/6iC1W3HMds1Tw3rarnnWvc12DWovjoQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBguYAq/hTiCLst2dnWx/KjiQqSXncNhuyS4Ubntmss17yTpyHT7Ne5hAW+jXuNICABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCM4QKm0NcPNvWr7oxVFuBOUJftH9HSds27rbfarjlj2b+Aqb9zNWqa/Rr7l2Rt2DgSAgAYQwgBAIyxHUKZmZkaMmSIoqKi5HA4tHr1ap/7R40aJYfD4bP07NkzUP0CAOoR2yF08uRJde7cWXPnzj3vmEGDBuno0aPeZe3atZfUJACgfrL9xoTk5GQlJydXO8bpdMrtdvvdFACgYaiR14TS09MVERGhDh06aMyYMcrLyzvv2OLiYhUWFvosAICGIeAhlJycrMWLF2vDhg166aWXtG3bNg0YMEDFxcVVjk9NTVV4eLh3iYmJCXRLAIBaKuCfE7rrrru8/46Pj9dNN92k2NhYffDBBxoxYkSl8ZMnT9aECRO8twsLCwkiAGggavzDqh6PR7GxsdqzZ0+V9zudTjmdzppuAwBQC9X454Ty8/OVk5Mjj8dT05sCANQxto+ETpw4ob1793pvZ2dna+fOnWrRooVatGihlJQU3X777fJ4PNq/f7+eeOIJtWrVSsOHDw9o4wCAus92CH366afq37+/93bF6zkjR47UvHnzlJWVpUWLFun48ePyeDzq37+/li9fLpfLFbiuAQD1gu0QSkxMlGVZ571//fr1l9QQLr8n+7xvugXUkOCYaL/qirpG2a75/ejX/drW5bC1uIlfdY6S0gB3gnNx7TgAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYU+PfrArAnC+mu/2q2500N8CdBM6KE61s18yb+Au/ttXky61+1eHicSQEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMZwAVOgjghJ99iuSfWsqIFOzFpwOMF2TZP3uRBpbcWREADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYwwVMoSBHuV91IY6gAHdStcJ7e16W7UjS9Kffsl3TP/R0DXRSmT/7+4xV5ufWLs/P1h/WgMOmW0AAcSQEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMZwAVPo2eV3+FV3531zAtvIeWS+8JrtGv8v3GnfGeuybcq2y7kf/BH/9wds17TXP2ugE5jCkRAAwBhCCABgjK0QSk1NVbdu3eRyuRQREaFhw4bp66+/9hljWZZSUlIUFRWl0NBQJSYmavfu3QFtGgBQP9gKoYyMDI0bN06bN29WWlqaSktLlZSUpJMnT3rHPP/885o9e7bmzp2rbdu2ye12a+DAgSoqKgp48wCAus3WGxPWrVvnc3v+/PmKiIjQ9u3b1bdvX1mWpTlz5mjKlCkaMWKEJGnhwoWKjIzUkiVLNHbs2MB1DgCo8y7pNaGCggJJUosWLSRJ2dnZys3NVVJSkneM0+lUv379tGnTpiofo7i4WIWFhT4LAKBh8DuELMvShAkT1Lt3b8XHx0uScnNzJUmRkZE+YyMjI733nSs1NVXh4eHeJSYmxt+WAAB1jN8h9PDDD2vXrl1aunRppfscDofPbcuyKq2rMHnyZBUUFHiXnJwcf1sCANQxfn1Y9ZFHHtF7772nzMxMRUdHe9e73W5JZ4+IPB6Pd31eXl6lo6MKTqdTTqfTnzYAAHWcrSMhy7L08MMPa+XKldqwYYPi4uJ87o+Li5Pb7VZaWpp3XUlJiTIyMpSQkBCYjgEA9YatI6Fx48ZpyZIlWrNmjVwul/d1nvDwcIWGhsrhcGj8+PGaNWuW2rdvr/bt22vWrFlq2rSp7r333hr5DwAA6i5bITRv3jxJUmJios/6+fPna9SoUZKkSZMm6dSpU3rooYf0/fffq0ePHvrwww/lcrkC0jAAoP5wWJZVqy6/WFhYqPDwcCVqqIIdIabbaRCCruvgV92k99+1XdPdedp2TYgjyHZNbb9wpz/82Q//OO3f79Abuf1s13z/kNt2jSP7sO2aMj7GUeuVWmeUrjUqKChQWFhYtWO5dhwAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCM8eubVVG/lH3xjV91Uyfcb7smZ0i57Zpvkv9guwZnPfT2A37VxTyzyY+q7/3aFho2joQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBguYAq/ha7Zarumwxr72+l7zzjbNSGjvrW/IUnrrl9uuybp87tt15QviLBdYzlsl6jdzn/bL5JU5lcVYB9HQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDBcwRa0XtnSz/aKl/m1ruLrbrmmmfX5syZ8a+7gQKWo7joQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGGMrhFJTU9WtWze5XC5FRERo2LBh+vrrr33GjBo1Sg6Hw2fp2bNnQJsGANQPtkIoIyND48aN0+bNm5WWlqbS0lIlJSXp5MmTPuMGDRqko0ePepe1a9cGtGkAQP1g65tV161b53N7/vz5ioiI0Pbt29W3b1/veqfTKbfbHZgOAQD11iW9JlRQUCBJatGihc/69PR0RUREqEOHDhozZozy8vLO+xjFxcUqLCz0WQAADYPfIWRZliZMmKDevXsrPj7euz45OVmLFy/Whg0b9NJLL2nbtm0aMGCAiouLq3yc1NRUhYeHe5eYmBh/WwIA1DEOy7IsfwrHjRunDz74QB9//LGio6PPO+7o0aOKjY3VsmXLNGLEiEr3FxcX+wRUYWGhYmJilKihCnaE+NMaAMCgUuuM0rVGBQUFCgsLq3asrdeEKjzyyCN67733lJmZWW0ASZLH41FsbKz27NlT5f1Op1NOp9OfNgAAdZytELIsS4888ohWrVql9PR0xcXFXbAmPz9fOTk58ng8fjcJAKifbL0mNG7cOP3xj3/UkiVL5HK5lJubq9zcXJ06dUqSdOLECU2cOFGffPKJ9u/fr/T0dA0ZMkStWrXS8OHDa+Q/AACou2wdCc2bN0+SlJiY6LN+/vz5GjVqlIKCgpSVlaVFixbp+PHj8ng86t+/v5YvXy6XyxWwpgEA9YPt03HVCQ0N1fr16y+pIQBAw8G14wAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxgSbbuBclmVJkkp1RrIMNwMAsK1UZyT95+95dWpdCBUVFUmSPtZaw50AAC5FUVGRwsPDqx3jsC4mqi6j8vJyHTlyRC6XSw6Hw+e+wsJCxcTEKCcnR2FhYYY6NI/9cBb74Sz2w1nsh7Nqw36wLEtFRUWKiopSo0bVv+pT646EGjVqpOjo6GrHhIWFNehJVoH9cBb74Sz2w1nsh7NM74cLHQFV4I0JAABjCCEAgDF1KoScTqemTZsmp9NpuhWj2A9nsR/OYj+cxX44q67th1r3xgQAQMNRp46EAAD1CyEEADCGEAIAGEMIAQCMIYQAAMbUqRB6/fXXFRcXpyZNmqhr16766KOPTLd0WaWkpMjhcPgsbrfbdFs1LjMzU0OGDFFUVJQcDodWr17tc79lWUpJSVFUVJRCQ0OVmJio3bt3m2m2Bl1oP4waNarS/OjZs6eZZmtIamqqunXrJpfLpYiICA0bNkxff/21z5iGMB8uZj/UlflQZ0Jo+fLlGj9+vKZMmaIdO3aoT58+Sk5O1sGDB023dlldf/31Onr0qHfJysoy3VKNO3nypDp37qy5c+dWef/zzz+v2bNna+7cudq2bZvcbrcGDhzovRhufXGh/SBJgwYN8pkfa9fWrwsBZ2RkaNy4cdq8ebPS0tJUWlqqpKQknTx50jumIcyHi9kPUh2ZD1Yd0b17d+uBBx7wWXfttddajz/+uKGOLr9p06ZZnTt3Nt2GUZKsVatWeW+Xl5dbbrfbevbZZ73rTp8+bYWHh1u///3vDXR4eZy7HyzLskaOHGkNHTrUSD+m5OXlWZKsjIwMy7Ia7nw4dz9YVt2ZD3XiSKikpETbt29XUlKSz/qkpCRt2rTJUFdm7NmzR1FRUYqLi9Pdd9+tffv2mW7JqOzsbOXm5vrMDafTqX79+jW4uSFJ6enpioiIUIcOHTRmzBjl5eWZbqlGFRQUSJJatGghqeHOh3P3Q4W6MB/qRAgdO3ZMZWVlioyM9FkfGRmp3NxcQ11dfj169NCiRYu0fv16vfnmm8rNzVVCQoLy8/NNt2ZMxc+/oc8NSUpOTtbixYu1YcMGvfTSS9q2bZsGDBig4uJi063VCMuyNGHCBPXu3Vvx8fGSGuZ8qGo/SHVnPtS6r3KozrnfL2RZVqV19VlycrL33506dVKvXr109dVXa+HChZowYYLBzsxr6HNDku666y7vv+Pj43XTTTcpNjZWH3zwgUaMGGGws5rx8MMPa9euXfr4448r3deQ5sP59kNdmQ914kioVatWCgoKqvRMJi8vr9IznoakWbNm6tSpk/bs2WO6FWMq3h3I3KjM4/EoNja2Xs6PRx55RO+99542btzo8/1jDW0+nG8/VKW2zoc6EUKNGzdW165dlZaW5rM+LS1NCQkJhroyr7i4WF9++aU8Ho/pVoyJi4uT2+32mRslJSXKyMho0HNDkvLz85WTk1Ov5odlWXr44Ye1cuVKbdiwQXFxcT73N5T5cKH9UJVaOx8MvinClmXLllkhISHWW2+9ZX3xxRfW+PHjrWbNmln79+833dpl89hjj1np6enWvn37rM2bN1uDBw+2XC5Xvd8HRUVF1o4dO6wdO3ZYkqzZs2dbO3bssA4cOGBZlmU9++yzVnh4uLVy5UorKyvLuueeeyyPx2MVFhYa7jywqtsPRUVF1mOPPWZt2rTJys7OtjZu3Gj16tXLatOmTb3aDw8++KAVHh5upaenW0ePHvUuP/zwg3dMQ5gPF9oPdWk+1JkQsizLeu2116zY2FircePGVpcuXXzejtgQ3HXXXZbH47FCQkKsqKgoa8SIEdbu3btNt1XjNm7caEmqtIwcOdKyrLNvy502bZrldrstp9Np9e3b18rKyjLbdA2obj/88MMPVlJSktW6dWsrJCTEatu2rTVy5Ejr4MGDptsOqKr+/5Ks+fPne8c0hPlwof1Ql+YD3ycEADCmTrwmBAConwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwJj/B/uo/q/P4B13AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaEAAAGxCAYAAADLfglZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAApH0lEQVR4nO3de3gU9d338c/mwBJws8gh2QRCiAiIBqGcSRGCSjS9QZDa4qEtWKVYgRYj8hRpIYolisoDFqV3uRVEQWjLyVYKxhsStMjJgiBSC5JAEEJKhGwIkJBknj942LokBGZN+OXwfl3XXFd2dr4730x+ySczu/tbh2VZlgAAMCDIdAMAgIaLEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGELpGHM84rmrJyM4w2mfiokQlLkoMqLbdnHYasnRItfbTbk47jV49OuD6Y4XHNHr1aEW8GKHGzzXWrfNv1ev/eN3243T/7+765d9+edn7F+1a5PdzDHk2RG1mt9HDax7WV96vAu7fjkuPVUZ2RkBjanPOZqVmpOrUuVPV2p8kjV49Wu3mtLvidozDhiPEdAMNxcePfOx3e8amGdqYtVEbRm3wW39zq5uvZVv1WsG5AvVf2F8lZSWaNXiWoq6L0jufvaNH//KoCooLlNIv5aoeJ+tklnbm7tScu+dccduFwxbqppY36ez5s9p0aJPSPkpTZnam9vx8j5o2avotvyN7ukd118ePfGx7TG3O2axnMp/R6G6j1axxs5pprgGprnFYXxFC10jfNn39brdq0kpBjqAK6y915vwZNQltUpOt1Vvzd8zXwZMHtWPMDvWI7iFJuuvGu3Ts9DFN2zhNP/3OT6/qj+yfP/+zIppGqH/b/lfcNj4iXj2je0qSBsUNUplVphmbZmj1P1froVsfqrSmpn7G4c7wK44v1LzqGof1FZfjapHERYmKfy1emw5tUsLrCWry2yb66ZqfSrpwOS81I7VCTWWXCXJP52rsX8aqzew2ajSjkeLmxumZjGdUWl4aUF/PZDyjPv/TR81faK7wtHB1/+/uev0fr+tyc9+u2rdKt86/VY2fa6wb5t6gV7a+UmEbb7FXk96fpLi5cWo0o5Faz26tiesmqqikKKAeK/P3nL8rsmmk7xf/oiEdhqjofJHWHVh3VY+zYt8K3XvTvQpy2P91uRgChwoOSbpwOeq6mddpz/E9SnorSa40l+5YfIckqaSsRM9tek43zbtJzuecavViKz285mH9u+jffo95vuy8JqdPluclj5r8ton6v9Ff277aVmHfl7sct/XIVg19Z6hazGqhxs81VvtX2mviuomSpNSMVD2V/pQkKW5uXKWXiZd/tlz9Xu+npjOb6rqZ1+mut+/SzmM7K+x/0a5F6jSvk5zPOdX51c5a/Oli28fvmxr6OKyvOBOqZY6dPqYfrfyRJn93smbeMdP2H77c07nqvaC3ghxBmjZwmtpf314fH/lYz216TtkF2Vo4bKHtnrILsjW2x1i1dbeVJG05skUT/jZBXxV+pWkDp/ltuyt3lyaun6jUganyXOfRkj1L9Mt1v1RJWYkmJUySdOE//4GLBuqI94ie7v+0bo28VXv/vVfTNk7Tnrw9+uDHH8jhcFy2n8RFico8lClretUTwJeUlcgZ4qyw/uK63cd36/74+6t8jCPeI9r21TbNGDSjyu0u58DXByRdOPP9Zl/3LLtHY3uM1a/6/0ql5aUqt8o1bNkwfXjoQ03+7mQlxCTo0KlDmp4xXYlfJWrHmB0KCw2TJI35yxgt/nSxJiVM0uAbBuuzvM80YvkIFZYUXrGf9QfWa+g7Q9W5VWfNTpqttu62yj6VrfcPvi9JerT7o/r67Nf63bbfaeUPVyrKFSXpP5eJZ344U7/e8Gs93O1h/fq2X6ukrEQvbn5Rty28TdvGbPNtt2jXIj285mEN6zRMLye9rIJzBUrNTFVxaXFAYS417HFYnxFCtczXZ7/Wn37wJ90ed3tA9akZqTp57qT2Pr7X98t6xw13KCwkTJPSJ+mphKdsP0fwzeAqt8qV2C5RlmVp7ta5+s2A3/j9oh4tPKqdY3eqq6erJCm5Q7LyivI0Y9MMPd7rcTUJbaJXtr6i3cd3a+ujW32Xru644Q61drXWfX+6T+sOrFNyh+TL9hMcFKxgR/AV+7655c364OAHOlxw2HcsJOmjwx9JkvLP5F/xMVZ8vkLNGjfToLhBV9xWksrKy1RaXqpzpeeUmZ2p5zY9J1cjl+7pdI9vm/Pl5zVtwDQ9/J2HfeuWfbZM6w6s04ofrtCIziN867t6uqrXgl5atGuRft7r5/rniX/qzU/f1BN9n9CswbMkSYPbD1bkdZF6aGXll/u+adzacWrrbqutj25V45DGvvUXe2kT3sZ3rL4T9R21a9bOt01OQY6mZ0zX+N7j9Uryf84qBrcfrA6/66BnMp/R8vuWq9wq19QNU9U9qrtWjVzlGx/92/ZXh991ULQr+qqO5aUa8jisz7gcV8tc3/j6gANIkv76r79qULtBinZFq7S81Ldc/GXKzM60/ZgbsjbozsV3yv28W8HPBit0RqimZUxT/tl85RXl+W17S8Qtvl/8ix7s8qC8xV7949g/fD3GR8Srm6ebX4933XiXHLryq7n+9yf/q9JpV760+LMeP1NoUKgeWvmQ9ubtVf6ZfL267VUt37tckq7qP/IV+1Zo2E3DFBJ0df+v9X29r0JnhMqV5tKQd4bIc51Hf3vob4q8LtJvu+/f/H2/23/911/VrHEzDe041O+YdPN0k+c6jzIOZUiSNmZtlCQ91MU/cH54yw+v2OO/8v+lL09+qUe+84hfAF2t9V+uV2l5qX7S9Sd+PTYOaayBsQN9P7cvTnyho4VH9WD8g37BENssVgkxCbb3e1FDHof1GWdCtczFyx+BOl50XH/5118UOiO00vtPnDlh6/G2fbVNSW8lKbFdohYMXaA24W3UKLiRVv9ztX774W91tvSs3/ae6zwVHuPiuov/8R0vOq4DXx+oth4vp3Orzlo1cpXG/nWs4ufHS5JiwmP0ctLLmvC3CWod3rrK+tzTufp7zt81+buTr3qfi4cvVudWnRUSFKLIppGV/jybhDZRuDPcb93xouM6de6UGj3XqNLHvXhM8s9eOIaXHueQoBC1CGtRZW8Xn1tqE97m6r6ZSxw/fVyS1GtBr0rvv/jH9HI9XlyXfSrb9r4b8jis7wihWsahyq9BO4OdKi4trrD+4i/8RS2btNStkbfqt7f/ttLHsXspZNlnyxQaHKq/PvhXv/+eV/9zdaXb557Ovey6Fk1a+HoMCwnTG8PeqPQxWjZpaavHqiR3SNahiYd04OsDKi0vVccWHfXHvX+UJA2IHVBl7ap9q9Q0tKkG3zD4qvfXuVVn36Wdy6nsZ9yySUu1CGuhdT+q/ElqVyOXJPmCJvd0rt8fr9Ly0gpj4VKtml54XuqI90iV213OxZ/Ln3/wZ8U2i73sdt/s8VKVrbsaDXkc1neEUB3Rrlk77c7b7bduQ9YGnS457bduSIchWntgrdpf317Xh13/rffrkEMhQSF+177Pnj+rt3a/Ven2e/P26tPcT/0uhSzds1SuRi51j+ru63HmRzPVIqyF4q6P+9Y9XonD4VCHFh0kXXiSeO7Wuerm6XbFX/4V+1ZoSMchlT6pXN2GdBiiZZ8tU1l5mfq06XPZ7RLbJUqSluxZ4vdqqz/u/eMVX/3YsUVHtb++vd7Y9YZS+qVc9vtyBl9Yf/a8/9nFXTfepZCgEH158ssKlxO/qVPLTr73wqT0S/Fdkjt06pA252wO6DmhhjwO6ztCqI748a0/1m82/kbTNk7TwNiB+vzfn2ve9nlyO91+2z076FmlH0xXwhsJ+kXvX6hTy046V3pO2aeytXb/Wv1+yO9tXY75r47/pdlbZuvBlQ/qZ91/pvyz+Xpp80u+P1SXinZF655l9yh1YKqiXFF6e/fbSj+YrhfufMH3XpiJfSdqxb4VGrBogJ7o+4RujbxV5Va5Dhcc1vtfvq8n+z1Z5R/iOxbfoczszKu6Hj9h7QQltktUiyYtdPDkQb2y9RUd8R5R5uiqnxvLP5OvzEOZWvb9ZVfcR3W4P/5+LdmzRN9b+j39ss8v1bt1b4UGheqI94g2Zm/UsE7DdG/ne9W5VWf96NYfac6WOQoNCtWdN9ypz/I+00sfv1ThEl9lXv3eqxr6zlD1fb2vnuj7hNq62+pwwWGt/3K9loxYIknqEtlFkjR361yN6jpKocGh6tSik9o1a6dnE5/V1A1TdfDkQd194926vvH1Ol50XNu+2qamoU31zKBnFOQI0oxBM/ToXx7Vvcvv1ZjuY3Tq3CmlZqZWepnsajTUcdgQEEJ1xFPffUreYq8W7Vqklza/pN6te+uP9/1Rw5YN89suyhWlHT/boRmZM/Ti5hd1xHtELqdLcc3ifH807Lg97na9cc8beuHvL2joO0PVOry1xnQfo4imEXrk3UcqbN/N000Pd3tY0zOma//X+xXtitbspNl6ot8Tvm2aNmqqDx/+UM9/9Lz+8MkflHUqS2EhYWrrbqs7b7jT7xVZlSkrL1OZVXZV/ed4czThbxN04swJtWjSQnffeLfW3L+mystJ0oXLPI2CG1X56qjqFBwUrHcfeFdzt8zVW7vfUtpHaQoJClGb8DYaGDvQFwyS9Po9ryuyaaQWfbpIr2x7Rd083bTihyt0/5+v/DLfu268S5se3qRnM5/VL/72C50rPac24W38Xr2X2C5RU/pP0ZufvqkF/1igcqtcG0dtvLD+tim6udXNmrt1rt757B0VlxbLc51HvVr30mM9HvM9xiPdL4yNF/7+gkb8cYTaNWunp/s/rcxDmQFNTdVQx2FD4LAu904voAH73pLvKSw0TCt+uMJ0K0C9RggBAIxp2C9QBwAYRQgBAIwhhAAAxhBCAABjCCEAgDG17n1C5eXlOnr0qFwuV5XTqAMAaifLslRYWKjo6GgFBVV9rlPrQujo0aOKiYkx3QYA4FvKyclRmzZVz9BS60LI5bowUWN/fU8hqnx2WwBA7VWq8/pIa31/z6tSYyH02muv6cUXX9SxY8d0yy23aM6cObrtttuuWHfxElyIQhXiIIQAoM75/1MgXM1TKjXywoTly5dr4sSJmjp1qnbu3KnbbrtNycnJOnz4cE3sDgBQR9VICM2ePVuPPPKIHn30UXXu3Flz5sxRTEyM5s+fXxO7AwDUUdUeQiUlJfrkk0+UlJTktz4pKUmbN2+usH1xcbG8Xq/fAgBoGKo9hE6cOKGysjJFRkb6rY+MjFRubsVPO0xLS5Pb7fYtvDIOABqOGnuz6qVPSFmWVemTVFOmTFFBQYFvycnJqamWAAC1TLW/Oq5ly5YKDg6ucNaTl5dX4exIkpxOp5zOmv/4ZABA7VPtZ0KNGjVSjx49lJ6e7rc+PT1dCQkJ1b07AEAdViPvE0pJSdGPf/xj9ezZU/369dMf/vAHHT58WI899tiViwEADUaNhNDIkSOVn5+vZ599VseOHVN8fLzWrl2r2Fg+Tx0A8B+17uO9vV6v3G63EjWMGRMAoA4qtc4rQ2tUUFCg8PDwKrfloxwAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGFPtIZSamiqHw+G3eDye6t4NAKAeCKmJB73lllv0wQcf+G4HBwfXxG4AAHVcjYRQSEgIZz8AgCuqkeeE9u/fr+joaMXFxen+++/XwYMHL7ttcXGxvF6v3wIAaBiqPYT69OmjxYsXa/369VqwYIFyc3OVkJCg/Pz8SrdPS0uT2+32LTExMdXdEgCglnJYlmXV5A6KiorUvn17TZ48WSkpKRXuLy4uVnFxse+21+tVTEyMEjVMIY7QmmwNAFADSq3zytAaFRQUKDw8vMpta+Q5oW9q2rSpunTpov3791d6v9PplNPprOk2AAC1UI2/T6i4uFj79u1TVFRUTe8KAFDHVHsITZo0SZmZmcrKytLWrVt13333yev1atSoUdW9KwBAHVftl+OOHDmiBx54QCdOnFCrVq3Ut29fbdmyRbGxsdW9KwBAHVftIbRs2bLqfkgAQD3F3HEAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYEyNf6gdUN8FdbvZds05T1PbNdnDHbZr7uu93XaNJJ23gm3XbHyrt+2aqMwC2zXWzr22a1B7cSYEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAY5hFG/WS9d1uAdUdHGe/Zmm/BbZrejSyP0t1rffUNtslZyeV2K75wyn7s5a/9ulA2zWS1OGRfbZrys+dC2hfDRVnQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDBOY4poq79/Ndk324/b38953X7VfJKl9SFgAVfYnI00/a38/T38+3HbNqcPNbNdI0mfDf2e75jfH+9qumeXZYbuma9gh2zWzey+3XSNJU54YbbumTdrmgPbVUHEmBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGMIEpAnZwaTfbNUv6LbBd06OR/QlCpUAmIpUeyBpsu2b7P+Ns19z0y322a1oVfWG/xnbFBY/1uNN2Td4vYm3XPDHf/s/215EZtms+PBtlu0aSdo23P5Hr8LeH2a4pzTliu6a+4EwIAGAMIQQAMMZ2CG3atElDhw5VdHS0HA6HVq9e7Xe/ZVlKTU1VdHS0wsLClJiYqL1791ZXvwCAesR2CBUVFalr166aN29epffPmjVLs2fP1rx587R9+3Z5PB4NHjxYhYWF37pZAED9YvuFCcnJyUpOTq70PsuyNGfOHE2dOlUjRoyQJL355puKjIzU0qVLNXbs2G/XLQCgXqnW54SysrKUm5urpKQk3zqn06mBAwdq8+bKP/K2uLhYXq/XbwEANAzVGkK5ubmSpMjISL/1kZGRvvsulZaWJrfb7VtiYmKqsyUAQC1WI6+Oczgcfrcty6qw7qIpU6aooKDAt+Tk5NRESwCAWqha36zq8XgkXTgjior6z5vD8vLyKpwdXeR0OuV0OquzDQBAHVGtZ0JxcXHyeDxKT0/3rSspKVFmZqYSEhKqc1cAgHrA9pnQ6dOndeDAAd/trKws7dq1S82bN1fbtm01ceJEzZw5Ux06dFCHDh00c+ZMNWnSRA8++GC1Ng4AqPtsh9COHTs0aNAg3+2UlBRJ0qhRo7Ro0SJNnjxZZ8+e1eOPP66TJ0+qT58+ev/99+VyuaqvawBAveCwLMsy3cQ3eb1eud1uJWqYQhyhptupc4KaNrVds//ZLgHta9/9r9quCVLlL1CpyvZi+0P0oTXjbNdIUqdn7E8sWnaqIKB91WZ3fmb/zeXpY/rbrjnd1v5Esx/Ofs12zbV0T997bNfUtwlMS63zytAaFRQUKDw8vMptmTsOAGAMIQQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxlTrJ6vCvFP32J8Re8MPXgpoX0FqYrvmf8/a/xTd5x8fZbvmxve32K6RpLKAqq4NR4j9X9egTu0D2tf/rG5uu+bFxW/arunSKM92jQIYd8GOwP7f7rLV/uegtc77MqB9NVScCQEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMUxgWs9YwfZrzlmO6m/kMgrLw2zX5PZpZLvm7Ijetmsk6cYOxwKqs6vgXGPbNT+I/YftmnHN3rJdI0k7Suwf8+86ywPYk/3JSAPx93OB9Ca1fs7+74ZVXBzQvhoqzoQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBiHZVmW6Sa+yev1yu12K1HDFOIINd1OnRPkctmuObuiRUD7evumt23XRAbbn8A01GF/VtYyK7AJKwNRbJXarnE6mDtYkkpVZrsmcff9tmuaj7O/H0kqPZgdUF1DV2qdV4bWqKCgQOHh4VVuy5kQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABjDLIr1THlhoe0aZ5L9Gkn6WeQI2zX7UtvZrknqscd2zb8KImzXSNKhr1rargluZH9yzHs67bZdM8uzw3ZNbXfzxp/Zrun05Fe2a0qP59muwbXBmRAAwBhCCABgjO0Q2rRpk4YOHaro6Gg5HA6tXr3a7/7Ro0fL4XD4LX379q2ufgEA9YjtECoqKlLXrl01b968y25z991369ixY75l7dq136pJAED9ZPuFCcnJyUpOTq5yG6fTKY/HE3BTAICGoUaeE8rIyFBERIQ6duyoMWPGKC/v8q9MKS4ultfr9VsAAA1DtYdQcnKylixZog0bNujll1/W9u3bdfvtt6u4uLjS7dPS0uR2u31LTExMdbcEAKilqv19QiNHjvR9HR8fr549eyo2NlbvvfeeRoyo+L6SKVOmKCUlxXfb6/USRADQQNT4m1WjoqIUGxur/fv3V3q/0+mU0+ms6TYAALVQjb9PKD8/Xzk5OYqKiqrpXQEA6hjbZ0KnT5/WgQMHfLezsrK0a9cuNW/eXM2bN1dqaqq+//3vKyoqStnZ2Xr66afVsmVL3XvvvdXaOACg7rMdQjt27NCgQYN8ty8+nzNq1CjNnz9fe/bs0eLFi3Xq1ClFRUVp0KBBWr58uVwuV/V1DQCoFxyWZVmmm/gmr9crt9utRA1TiCPUdDtAjTi66mbbNbt6v10DnVQuu/SM7Zrhv5tsu6b1nG22a6zSUts1uLZKrfPK0BoVFBQoPDy8ym2ZOw4AYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADG1PgnqwL1XdbMfrZr/tHr/wawp0YB1ATmvln2Z8SOfnWz7ZpaNYU/jOBMCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMYQJT4BuOPpVgu2b9Q7Ns14Q5mtiuCcTckzcGVOdZuMt2TXlAe0JDx5kQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABjDBKaol84n9QyobvV4+5ORtg25NpORHi49Y7vm3f9zR0D7cp7ZHlAdYBdnQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDBOYol7KHhIcUF27azQZ6bEy+5OR/mTik7Zrmry31XYNcC1xJgQAMIYQAgAYYyuE0tLS1KtXL7lcLkVERGj48OH64osv/LaxLEupqamKjo5WWFiYEhMTtXfv3mptGgBQP9gKoczMTI0bN05btmxRenq6SktLlZSUpKKiIt82s2bN0uzZszVv3jxt375dHo9HgwcPVmFhYbU3DwCo22y9MGHdunV+txcuXKiIiAh98sknGjBggCzL0pw5czR16lSNGDFCkvTmm28qMjJSS5cu1dixY6uvcwBAnfetnhMqKCiQJDVv3lySlJWVpdzcXCUlJfm2cTqdGjhwoDZv3lzpYxQXF8vr9fotAICGIeAQsixLKSkp6t+/v+Lj4yVJubm5kqTIyEi/bSMjI333XSotLU1ut9u3xMTEBNoSAKCOCTiExo8fr927d+udd96pcJ/D4fC7bVlWhXUXTZkyRQUFBb4lJycn0JYAAHVMQG9WnTBhgt59911t2rRJbdq08a33eDySLpwRRUVF+dbn5eVVODu6yOl0yul0BtIGAKCOs3UmZFmWxo8fr5UrV2rDhg2Ki4vzuz8uLk4ej0fp6em+dSUlJcrMzFRCQkL1dAwAqDdsnQmNGzdOS5cu1Zo1a+RyuXzP87jdboWFhcnhcGjixImaOXOmOnTooA4dOmjmzJlq0qSJHnzwwRr5BgAAdZetEJo/f74kKTEx0W/9woULNXr0aEnS5MmTdfbsWT3++OM6efKk+vTpo/fff18ul6taGgYA1B8Oy7Is0018k9frldvtVqKGKcQRarod1ALBLZrbrnnn0/cC2td1jmvz/GSnzJ/armn/4K7qbwSoAaXWeWVojQoKChQeHl7ltswdBwAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMC+mRVIFDB119vu2bi1g9t11yr2bAl6YX8zrZrOozZb7um3HYFUPtxJgQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxjCBKa6pE/fcZLsmqclG2zVllu2SgK19JtF2TdOirdXeB1AXcSYEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYwgSmuqe9P+sB2TZlVXgOdVO7Gvzxmu6bjCiYjBQLFmRAAwBhCCABgDCEEADCGEAIAGEMIAQCMIYQAAMYQQgAAYwghAIAxhBAAwBhCCABgDCEEADCGEAIAGMMEprimuoYdtl0T7LD/v9KWc2W2ayTp5ll5tmtKA9oTAIkzIQCAQYQQAMAYWyGUlpamXr16yeVyKSIiQsOHD9cXX3zht83o0aPlcDj8lr59+1Zr0wCA+sFWCGVmZmrcuHHasmWL0tPTVVpaqqSkJBUVFfltd/fdd+vYsWO+Ze3atdXaNACgfrD1woR169b53V64cKEiIiL0ySefaMCAAb71TqdTHo+nejoEANRb3+o5oYKCAklS8+bN/dZnZGQoIiJCHTt21JgxY5SXd/lXHBUXF8vr9fotAICGIeAQsixLKSkp6t+/v+Lj433rk5OTtWTJEm3YsEEvv/yytm/frttvv13FxcWVPk5aWprcbrdviYmJCbQlAEAdE/D7hMaPH6/du3fro48+8ls/cuRI39fx8fHq2bOnYmNj9d5772nEiBEVHmfKlClKSUnx3fZ6vQQRADQQAYXQhAkT9O6772rTpk1q06ZNldtGRUUpNjZW+/fvr/R+p9Mpp9MZSBsAgDrOVghZlqUJEyZo1apVysjIUFxc3BVr8vPzlZOTo6ioqICbBADUT7aeExo3bpzefvttLV26VC6XS7m5ucrNzdXZs2clSadPn9akSZP08ccfKzs7WxkZGRo6dKhatmype++9t0a+AQBA3WXrTGj+/PmSpMTERL/1Cxcu1OjRoxUcHKw9e/Zo8eLFOnXqlKKiojRo0CAtX75cLper2poGANQPti/HVSUsLEzr16//Vg0BABoOZtHGNTVxySO2a/455jXbNT99Y4LtGkmKObg5oDoAgWECUwCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwxmFdaWrsa8zr9crtditRwxTiCDXdDgDAplLrvDK0RgUFBQoPD69yW86EAADGEEIAAGMIIQCAMYQQAMAYQggAYAwhBAAwhhACABhDCAEAjCGEAADGEEIAAGMIIQCAMSGmG7jUxansSnVeqlWz2gEArkapzkv6z9/zqtS6ECosLJQkfaS1hjsBAHwbhYWFcrvdVW5T62bRLi8v19GjR+VyueRwOPzu83q9iomJUU5OzhVnZq3POA4XcBwu4DhcwHG4oDYcB8uyVFhYqOjoaAUFVf2sT607EwoKClKbNm2q3CY8PLxBD7KLOA4XcBwu4DhcwHG4wPRxuNIZ0EW8MAEAYAwhBAAwpk6FkNPp1PTp0+V0Ok23YhTH4QKOwwUchws4DhfUteNQ616YAABoOOrUmRAAoH4hhAAAxhBCAABjCCEAgDGEEADAmDoVQq+99pri4uLUuHFj9ejRQx9++KHplq6p1NRUORwOv8Xj8Zhuq8Zt2rRJQ4cOVXR0tBwOh1avXu13v2VZSk1NVXR0tMLCwpSYmKi9e/eaabYGXek4jB49usL46Nu3r5lma0haWpp69eoll8uliIgIDR8+XF988YXfNg1hPFzNcagr46HOhNDy5cs1ceJETZ06VTt37tRtt92m5ORkHT582HRr19Qtt9yiY8eO+ZY9e/aYbqnGFRUVqWvXrpo3b16l98+aNUuzZ8/WvHnztH37dnk8Hg0ePNg3GW59caXjIEl333233/hYu7Z+TQScmZmpcePGacuWLUpPT1dpaamSkpJUVFTk26YhjIerOQ5SHRkPVh3Ru3dv67HHHvNbd9NNN1m/+tWvDHV07U2fPt3q2rWr6TaMkmStWrXKd7u8vNzyeDzW888/71t37tw5y+12W7///e8NdHhtXHocLMuyRo0aZQ0bNsxIP6bk5eVZkqzMzEzLshrueLj0OFhW3RkPdeJMqKSkRJ988omSkpL81iclJWnz5s2GujJj//79io6OVlxcnO6//34dPHjQdEtGZWVlKTc3129sOJ1ODRw4sMGNDUnKyMhQRESEOnbsqDFjxigvL890SzWqoKBAktS8eXNJDXc8XHocLqoL46FOhNCJEydUVlamyMhIv/WRkZHKzc011NW116dPHy1evFjr16/XggULlJubq4SEBOXn55tuzZiLP/+GPjYkKTk5WUuWLNGGDRv08ssva/v27br99ttVXFxsurUaYVmWUlJS1L9/f8XHx0tqmOOhsuMg1Z3xUOs+yqEql36+kGVZFdbVZ8nJyb6vu3Tpon79+ql9+/Z68803lZKSYrAz8xr62JCkkSNH+r6Oj49Xz549FRsbq/fee08jRoww2FnNGD9+vHbv3q2PPvqown0NaTxc7jjUlfFQJ86EWrZsqeDg4Ar/yeTl5VX4j6chadq0qbp06aL9+/ebbsWYi68OZGxUFBUVpdjY2Ho5PiZMmKB3331XGzdu9Pv8sYY2Hi53HCpTW8dDnQihRo0aqUePHkpPT/dbn56eroSEBENdmVdcXKx9+/YpKirKdCvGxMXFyePx+I2NkpISZWZmNuixIUn5+fnKycmpV+PDsiyNHz9eK1eu1IYNGxQXF+d3f0MZD1c6DpWptePB4IsibFm2bJkVGhpqvf7669bnn39uTZw40WratKmVnZ1turVr5sknn7QyMjKsgwcPWlu2bLGGDBliuVyuen8MCgsLrZ07d1o7d+60JFmzZ8+2du7caR06dMiyLMt6/vnnLbfbba1cudLas2eP9cADD1hRUVGW1+s13Hn1quo4FBYWWk8++aS1efNmKysry9q4caPVr18/q3Xr1vXqOPz85z+33G63lZGRYR07dsy3nDlzxrdNQxgPVzoOdWk81JkQsizLevXVV63Y2FirUaNGVvfu3f1ejtgQjBw50oqKirJCQ0Ot6Ohoa8SIEdbevXtNt1XjNm7caEmqsIwaNcqyrAsvy50+fbrl8Xgsp9NpDRgwwNqzZ4/ZpmtAVcfhzJkzVlJSktWqVSsrNDTUatu2rTVq1Cjr8OHDptuuVpV9/5KshQsX+rZpCOPhSsehLo0HPk8IAGBMnXhOCABQPxFCAABjCCEAgDGEEADAGEIIAGAMIQQAMIYQAgAYQwgBAIwhhAAAxhBCAABjCCEAgDH/Dx05VPKVHWG9AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 作为一名语言模型，我无法直接为您提供图像演示。不过，您可以使用Python中的Matplotlib库来可视化训练和测试的准确率变化，以及模型对测试集的预测结果。下面是一些可以帮助您开始的示例代码：\n",
    "\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 加载MNIST数据集\n",
    "(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()\n",
    "x_train = x_train.reshape((60000, 784))\n",
    "x_test = x_test.reshape((10000, 784))\n",
    "x_train, x_test = x_train / 255.0, x_test / 255.0\n",
    "\n",
    "# 构建前向神经网络模型\n",
    "model = keras.Sequential([\n",
    "    keras.layers.Dense(128, activation='relu'),\n",
    "    keras.layers.Dense(10, activation='softmax')\n",
    "])\n",
    "\n",
    "# 编译模型\n",
    "model.compile(optimizer='adam',\n",
    "              loss='sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "# 训练模型\n",
    "history = model.fit(x_train, y_train, epochs=5, batch_size=32, validation_data=(x_test, y_test))\n",
    "\n",
    "# 可视化训练和测试的准确率变化\n",
    "plt.plot(history.history['accuracy'])\n",
    "plt.plot(history.history['val_accuracy'])\n",
    "plt.title('Model Accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Test'], loc='upper left')\n",
    "plt.show()\n",
    "\n",
    "# 进行预测\n",
    "predictions = model.predict(x_test)\n",
    "\n",
    "# 可视化模型对测试集的预测结果\n",
    "for i in range(10):\n",
    "    plt.imshow(x_test[i].reshape(28, 28))\n",
    "    predicted_label = np.argmax(predictions[i])\n",
    "    true_label = y_test[i]\n",
    "    if predicted_label == true_label:\n",
    "        color = 'green'\n",
    "    else:\n",
    "        color = 'red'\n",
    "    plt.title('True label: {} / Predicted label: {}'.format(true_label, predicted_label), color=color)\n",
    "    plt.show()\n",
    "\n",
    "#这段代码会输出两张图像。第一张图像是训练和测试的准确率变化图，用于观察模型的性能表现。\n",
    "# 第二张图像是模型对测试集的预测结果，用于观察模型对数字图像的分类效果。您可以自行运行这段代码，并根据需要进行修改和调整。"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "VS Code iPython",
   "language": "python",
   "name": "vscode"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "db6b9b01f897b742beb166a9e4b99340bcc626601400287327c070a83757b2b9"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}