Lecture02 - MNIST.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src='https://www.anadronestarting.com/wp-content/uploads/intel-main_opt.png' width=50%>\n",
    "\n",
    "# 모바일넷을 이용한 이미지분류 - MNIST\n",
    "<font size=5><b>(Image Classification using Mobilenet)<b></font>\n",
    "\n",
    "<div align='right'>성  민  석<br>(Minsuk Sung)</div>\n",
    "\n",
    "<img src='https://upload.wikimedia.org/wikipedia/commons/thumb/2/27/MnistExamples.png/440px-MnistExamples.png' width=60%>\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>강의목차<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#필요한-라이브러리-및-옵션\" data-toc-modified-id=\"필요한-라이브러리-및-옵션-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>필요한 라이브러리 및 옵션</a></span><ul class=\"toc-item\"><li><span><a href=\"#기본-라이브러리(Library)\" data-toc-modified-id=\"기본-라이브러리(Library)-1.1\"><span class=\"toc-item-num\">1.1&nbsp;&nbsp;</span>기본 라이브러리(Library)</a></span></li><li><span><a href=\"#Keras,-Tensorflow와-손을-잡다!\" data-toc-modified-id=\"Keras,-Tensorflow와-손을-잡다!-1.2\"><span class=\"toc-item-num\">1.2&nbsp;&nbsp;</span>Keras, Tensorflow와 손을 잡다!</a></span><ul class=\"toc-item\"><li><span><a href=\"#신경망-구현-순서¶\" data-toc-modified-id=\"신경망-구현-순서¶-1.2.1\"><span class=\"toc-item-num\">1.2.1&nbsp;&nbsp;</span>신경망 구현 순서¶</a></span></li></ul></li><li><span><a href=\"#옵션(Option)\" data-toc-modified-id=\"옵션(Option)-1.3\"><span class=\"toc-item-num\">1.3&nbsp;&nbsp;</span>옵션(Option)</a></span></li></ul></li><li><span><a href=\"#예제---MNIST\" data-toc-modified-id=\"예제---MNIST-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>예제 - MNIST</a></span><ul class=\"toc-item\"><li><span><a href=\"#MNIST-데이터-불러오기\" data-toc-modified-id=\"MNIST-데이터-불러오기-2.1\"><span class=\"toc-item-num\">2.1&nbsp;&nbsp;</span>MNIST 데이터 불러오기</a></span></li><li><span><a href=\"#MNIST-데이터-형태-확인하기\" data-toc-modified-id=\"MNIST-데이터-형태-확인하기-2.2\"><span class=\"toc-item-num\">2.2&nbsp;&nbsp;</span>MNIST 데이터 형태 확인하기</a></span><ul class=\"toc-item\"><li><span><a href=\"#Train-데이터셋\" data-toc-modified-id=\"Train-데이터셋-2.2.1\"><span class=\"toc-item-num\">2.2.1&nbsp;&nbsp;</span>Train 데이터셋</a></span></li><li><span><a href=\"#Validation-데이터셋\" data-toc-modified-id=\"Validation-데이터셋-2.2.2\"><span class=\"toc-item-num\">2.2.2&nbsp;&nbsp;</span>Validation 데이터셋</a></span></li><li><span><a href=\"#Test-데이터셋\" data-toc-modified-id=\"Test-데이터셋-2.2.3\"><span class=\"toc-item-num\">2.2.3&nbsp;&nbsp;</span>Test 데이터셋</a></span></li></ul></li><li><span><a href=\"#데이터-시각화하기\" data-toc-modified-id=\"데이터-시각화하기-2.3\"><span class=\"toc-item-num\">2.3&nbsp;&nbsp;</span>데이터 시각화하기</a></span></li><li><span><a href=\"#데이터-전처리\" data-toc-modified-id=\"데이터-전처리-2.4\"><span class=\"toc-item-num\">2.4&nbsp;&nbsp;</span>데이터 전처리</a></span></li><li><span><a href=\"#모델링\" data-toc-modified-id=\"모델링-2.5\"><span class=\"toc-item-num\">2.5&nbsp;&nbsp;</span>모델링</a></span><ul class=\"toc-item\"><li><span><a href=\"#모델-구성\" data-toc-modified-id=\"모델-구성-2.5.1\"><span class=\"toc-item-num\">2.5.1&nbsp;&nbsp;</span>모델 구성</a></span></li><li><span><a href=\"#모델-컴파일\" data-toc-modified-id=\"모델-컴파일-2.5.2\"><span class=\"toc-item-num\">2.5.2&nbsp;&nbsp;</span>모델 컴파일</a></span></li><li><span><a href=\"#모델-확인\" data-toc-modified-id=\"모델-확인-2.5.3\"><span class=\"toc-item-num\">2.5.3&nbsp;&nbsp;</span>모델 확인</a></span></li></ul></li><li><span><a href=\"#신경망-모델-학습\" data-toc-modified-id=\"신경망-모델-학습-2.6\"><span class=\"toc-item-num\">2.6&nbsp;&nbsp;</span>신경망 모델 학습</a></span></li><li><span><a href=\"#신경망-모델-평가\" data-toc-modified-id=\"신경망-모델-평가-2.7\"><span class=\"toc-item-num\">2.7&nbsp;&nbsp;</span>신경망 모델 평가</a></span></li><li><span><a href=\"#신경망-모델-검증하기\" data-toc-modified-id=\"신경망-모델-검증하기-2.8\"><span class=\"toc-item-num\">2.8&nbsp;&nbsp;</span>신경망 모델 검증하기</a></span></li><li><span><a href=\"#다음-예제에서는\" data-toc-modified-id=\"다음-예제에서는-2.9\"><span class=\"toc-item-num\">2.9&nbsp;&nbsp;</span>다음 예제에서는</a></span></li></ul></li><li><span><a href=\"#참고\" data-toc-modified-id=\"참고-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>참고</a></span></li></ul></div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 필요한 라이브러리 및 옵션\n",
    "\n",
    "### 기본 라이브러리(Library)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:14.716094Z",
     "start_time": "2019-12-22T16:30:14.708252Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import sys\n",
    "import glob\n",
    "import random\n",
    "import warnings\n",
    "import itertools\n",
    "from tqdm import tqdm\n",
    "from pathlib import Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:15.265214Z",
     "start_time": "2019-12-22T16:30:14.718108Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:15.361914Z",
     "start_time": "2019-12-22T16:30:15.266580Z"
    }
   },
   "outputs": [],
   "source": [
    "import cv2\n",
    "from PIL import Image\n",
    "from PIL import ImageFont\n",
    "from PIL import ImageDraw\n",
    "from IPython.display import SVG\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "import xml.etree.ElementTree as ET\n",
    "from xml.etree.ElementTree import Element, ElementTree"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Keras, Tensorflow와 손을 잡다!\n",
    "\n",
    "> Tensorflow + Keras = tf.Keras !!\n",
    "\n",
    "<img src='https://blog.keras.io/img/keras-tensorflow-logo.jpg' width=80%>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.134037Z",
     "start_time": "2019-12-22T16:30:15.362971Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.datasets import mnist,cifar10\n",
    "from tensorflow.keras.preprocessing.image import load_img,img_to_array,ImageDataGenerator\n",
    "from tensorflow.keras.applications import MobileNet, MobileNetV2\n",
    "from tensorflow.keras.models import Model,Sequential\n",
    "from tensorflow.keras.layers import Dense, Conv2D, Flatten, Dropout, MaxPooling2D, Conv2D,GlobalAveragePooling2D\n",
    "from tensorflow.keras.optimizers import RMSprop, Adam\n",
    "from tensorflow.keras.utils import to_categorical,plot_model\n",
    "from tensorflow.keras.losses import categorical_crossentropy\n",
    "from tensorflow.keras.callbacks import Callback\n",
    "from tensorflow.python.client import device_lib"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 신경망 구현 순서¶\n",
    "Keras 를 사용하면 다음과 같은 순서로 신경망을 구성할 수 있다.\n",
    "\n",
    "1. Sequential 모형 클래스 객체 생성\n",
    "\n",
    "\n",
    "2. add 메서드로 레이어 추가.\n",
    "\n",
    "    - 입력단부터 순차적으로 추가한다.\n",
    "    - 레이어는 출력 뉴런 갯수를 첫번째 인수로 받는다.\n",
    "    - 최초의 레이어는 input_dim 인수로 입력 크기를 설정해야 한다.\n",
    "    - activation 인수로 활성화함수 설정\n",
    "\n",
    "\n",
    "3. compile 메서드로 모형 완성.\n",
    "\n",
    "    - loss인수로 비용함수 설정\n",
    "    - optimizer 인수로 최적화 알고리즘 설정\n",
    "    - metrics 인수로 트레이닝 단계에서 기록할 성능 기준 설정\n",
    "\n",
    "\n",
    "4. fit 메서드로 트레이닝\n",
    "\n",
    "    - nb_epoch 로 에포크(epoch) 횟수 설정\n",
    "    - batch_size 로 배치크기(batch size) 설정\n",
    "    - verbose는 학습 중 출력되는 문구를 설정하는 것으로, 주피터노트북(Jupyter Notebook)을 사용할 때는 verbose=2로 설정하여 진행 막대(progress bar)가 나오지 않도록 설정한다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 옵션(Option)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.138373Z",
     "start_time": "2019-12-22T16:30:16.135126Z"
    },
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\"\n",
    "\n",
    "warnings.filterwarnings(action='ignore')\n",
    "warnings.filterwarnings(action='default')\n",
    "\n",
    "%matplotlib inline\n",
    "# print(device_lib.list_local_devices())\n",
    "# keras.backend.tensorflow_backend._get_available_gpus()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## 예제 - MNIST\n",
    "\n",
    "![](https://upload.wikimedia.org/wikipedia/commons/2/27/MnistExamples.png)\n",
    "\n",
    "MNIST 데이터베이스 (Modified National Institute of Standards and Technology database)는 손으로 쓴 **`0~9까지 숫자`들로 이루어진 대형 데이터베이스**이며, 다양한 화상 처리 시스템을 트레이닝하기 위해 일반적으로 사용된다. 이 데이터베이스는 또한 기계 학습 분야의 트레이닝 및 테스트에 널리 사용된다. NIST의 오리지널 데이터셋의 샘플을 재혼합하여 만들어졌다. 개발자들은 NIST의 트레이닝 데이터셋이 미국의 인구조사국 직원들로부터 취합한 이후로 테스팅 데이터셋이 미국의 중등학교 학생들로부터 취합되는 중에 기계 학습 실험에 딱 적합하지는 않은 것을 느꼈다. 게다가 NIST의 흑백 그림들은 28x28 픽셀의 바운딩 박스와 앤티엘리어싱 처리되어 그레이스케일 레벨이 들어가 있도록 평준화되었다. MNIST sample images. MNIST 테스트 데이터셋의 샘플 이미지. MNIST 데이터베이스는 **`60,000개의 트레이닝 이미지와 10,000개의 테스트 이미지를 포함`** 한다. 트레이닝 세트의 절반과 테스트 세트의 절반은 NIST의 트레이닝 데이터셋에서 취합하였으며, 그 밖의 트레이닝 세트의 절반과 테스트 세트의 절반은 NIST의 테스트 데이터셋으로부터 취합되었다.\n",
    "\n",
    "출처 : [위키백과](https://ko.wikipedia.org/wiki/MNIST_%EB%8D%B0%EC%9D%B4%ED%84%B0%EB%B2%A0%EC%9D%B4%EC%8A%A4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### MNIST 데이터 불러오기\n",
    "\n",
    "Keras에서는 MNIST와 같이 보편적으로 많이 쓰이는 데이터를 쉽게 가져올 수 있는 함수가 준비되어있다.  \n",
    "처음 함수 호출시에는 데이터 다운로드를 하기 위해서 어느정도 시간이 소요될 수 있다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.337507Z",
     "start_time": "2019-12-22T16:30:16.139366Z"
    }
   },
   "outputs": [],
   "source": [
    "# Keras에서 제공하는 MNIST 데이터를 불러오는 함수 : mnist.load_data()\n",
    "(X_train, y_train), (X_test,y_test) = mnist.load_data() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.341330Z",
     "start_time": "2019-12-22T16:30:16.339164Z"
    }
   },
   "outputs": [],
   "source": [
    "# Train / Validation 데이터 분리하기\n",
    "# Train 데이터를 Train / Validation 으로 나누어줌으로써 Overfitting 여부를 파악\n",
    "X_train, X_valid = X_train[:50000], X_train[50000:]\n",
    "y_train, y_valid = y_train[:50000], y_train[50000:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.345439Z",
     "start_time": "2019-12-22T16:30:16.342638Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MNIST Train 데이터의 크기 : 50000\n",
      "MNIST Validation 데이터의 크기 : 10000\n",
      "MNIST Test 데이터의 크기 : 10000\n"
     ]
    }
   ],
   "source": [
    "# Train / Test 데이터의 크기 확인\n",
    "print('MNIST Train 데이터의 크기 : {}'.format(len(X_train)))\n",
    "print('MNIST Validation 데이터의 크기 : {}'.format(len(X_valid)))\n",
    "print('MNIST Test 데이터의 크기 : {}'.format(len(X_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### MNIST 데이터 형태 확인하기"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train 데이터셋"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.348849Z",
     "start_time": "2019-12-22T16:30:16.346488Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train Shape :  (50000, 28, 28)\n",
      "y_train Shape :  (50000,)\n"
     ]
    }
   ],
   "source": [
    "print(\"X_train Shape : \",X_train.shape) # 28*28짜리 크기의 숫자 이미지 50000개\n",
    "print(\"y_train Shape : \",y_train.shape) # 각 이미지별 레이블 50000개"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Validation 데이터셋"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.352494Z",
     "start_time": "2019-12-22T16:30:16.349851Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_val Shape :  (10000, 28, 28)\n",
      "y_val Shape :  (10000,)\n"
     ]
    }
   ],
   "source": [
    "print(\"X_val Shape : \",X_valid.shape) # 28*28짜리 크기의 숫자 이미지 10000개\n",
    "print(\"y_val Shape : \",y_valid.shape) # 각 이미지별 레이블 10000개"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Test 데이터셋"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.355804Z",
     "start_time": "2019-12-22T16:30:16.353636Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_test Shape :  (10000, 28, 28)\n",
      "y_test Shape :  (10000,)\n"
     ]
    }
   ],
   "source": [
    "print(\"X_test Shape : \",X_test.shape) # 28*28짜리 크기의 숫자 이미지 60000개\n",
    "print(\"y_test Shape : \",y_test.shape) # 각 이미지별 레이블 60000개"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 데이터 시각화하기\n",
    "\n",
    "그러면 이제 구체적으로 MNIST의 Train 데이터셋의 첫번재 데이터를 확인해보자."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.361356Z",
     "start_time": "2019-12-22T16:30:16.356849Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   3,\n",
       "         18,  18,  18, 126, 136, 175,  26, 166, 255, 247, 127,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,  30,  36,  94, 154, 170,\n",
       "        253, 253, 253, 253, 253, 225, 172, 253, 242, 195,  64,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,  49, 238, 253, 253, 253, 253,\n",
       "        253, 253, 253, 253, 251,  93,  82,  82,  56,  39,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,  18, 219, 253, 253, 253, 253,\n",
       "        253, 198, 182, 247, 241,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,  80, 156, 107, 253, 253,\n",
       "        205,  11,   0,  43, 154,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,  14,   1, 154, 253,\n",
       "         90,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 139, 253,\n",
       "        190,   2,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  11, 190,\n",
       "        253,  70,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  35,\n",
       "        241, 225, 160, 108,   1,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "         81, 240, 253, 253, 119,  25,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,  45, 186, 253, 253, 150,  27,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,  16,  93, 252, 253, 187,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0, 249, 253, 249,  64,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,  46, 130, 183, 253, 253, 207,   2,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  39,\n",
       "        148, 229, 253, 253, 253, 250, 182,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  24, 114, 221,\n",
       "        253, 253, 253, 253, 201,  78,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,  23,  66, 213, 253, 253,\n",
       "        253, 253, 198,  81,   2,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,  18, 171, 219, 253, 253, 253, 253,\n",
       "        195,  80,   9,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,  55, 172, 226, 253, 253, 253, 253, 244, 133,\n",
       "         11,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0, 136, 253, 253, 253, 212, 135, 132,  16,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0],\n",
       "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,\n",
       "          0,   0]], dtype=uint8)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:16.482749Z",
     "start_time": "2019-12-22T16:30:16.362648Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[0], cmap='Greys', interpolation='nearest')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "50000개의 중 랜덤으로 한번 살펴보자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:17.659801Z",
     "start_time": "2019-12-22T16:30:16.483871Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 재연성을 위하여 랜덤시드 고정\n",
    "np.random.seed(1234)  \n",
    "\n",
    "# random 함수를 통해서 임의의 16개 데이터 가져오기\n",
    "samples = np.random.randint(0,len(X_train)+1,size=16)\n",
    "\n",
    "# MNIST를 그릴 Figure 준비\n",
    "plt.figure(figsize=(12,12))\n",
    "\n",
    "# 16개의 이미지 시각화\n",
    "for count, n in enumerate(samples,start=1):\n",
    "    plt.subplot(4, 4, count)\n",
    "    plt.imshow(X_train[n], cmap='Greys', interpolation='nearest')\n",
    "    label_name = \"Label:\" + str(y_train[n])\n",
    "    plt.title(label_name)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 데이터 전처리"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:17.714607Z",
     "start_time": "2019-12-22T16:30:17.660909Z"
    }
   },
   "outputs": [],
   "source": [
    "# 데이터 크기 조정(Data Reshape)\n",
    "X_train = X_train.reshape(X_train.shape[0],28,28,1)\n",
    "X_valid = X_valid.reshape(X_valid.shape[0],28,28,1)\n",
    "X_test = X_test.reshape(X_test.shape[0],28,28,1)\n",
    "\n",
    "# 데이터 포맷 바꾸기\n",
    "# 정수(int)인 데이터에서 실수(float)으로 변환\n",
    "X_train = X_train.astype('float32')\n",
    "X_valid = X_valid.astype('float32')\n",
    "X_test = X_test.astype('float32')\n",
    "\n",
    "# 데이터 정규화(Data Regularization)\n",
    "# 이 과정을 통해서 추후 학습할 신경망이 조금 더 학습이 원할히 될 수 있게함\n",
    "X_train = X_train / 255\n",
    "X_valid = X_valid / 255\n",
    "X_test = X_test / 255\n",
    "\n",
    "# 원-핫 인코딩(One Hot Encoding)\n",
    "# Keras의 to_categorical함수를 통해서 모든 Train 데이터의 레이블을 벡터화(Vectorize)\n",
    "# ex) [3] -> [0 0 0 1 0 0 0 0 0 0]\n",
    "y_train = to_categorical(y_train, 10) \n",
    "y_valid = to_categorical(y_valid,10)\n",
    "y_test = to_categorical(y_test, 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 모델링\n",
    "\n",
    "간단하게 CNN 모델을 이용하여, MNIST 데이터를 분류해보자."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:17.717796Z",
     "start_time": "2019-12-22T16:30:17.715596Z"
    }
   },
   "outputs": [],
   "source": [
    "INPUT_SHAPE = (28,28,1)\n",
    "KERNEL_SIZE = (3, 3)\n",
    "POOL_SIZE = (2,2)\n",
    "DROP_RATE = 0.3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 모델 구성\n",
    "Keras에서는 add함수를 통해서 레고 블록을 조립하듯이, 원하는 Layer를 추가 가능함  \n",
    "여기서는 Conv Layer와 MaxPooling Layer 그리고 Dense Layer를 이용하여 모델을 구성하도록 하자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:17.810273Z",
     "start_time": "2019-12-22T16:30:17.718975Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/minsuk/anaconda3/envs/work/lib/python3.6/site-packages/tensorflow/python/ops/init_ops.py:1251: calling VarianceScaling.__init__ (from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Call initializer instance with the dtype argument instead of passing it to the constructor\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "\n",
    "# 1번쨰 ConvNet\n",
    "# Conv Layer\n",
    "model.add(Conv2D(filters=64,\n",
    "                 kernel_size=KERNEL_SIZE,\n",
    "                 activation='relu',\n",
    "                 input_shape=INPUT_SHAPE\n",
    "                 ))\n",
    "# Max Pooling Layer\n",
    "model.add(MaxPooling2D(pool_size=POOL_SIZE))\n",
    "\n",
    "# 2번쨰 Conv Layer\n",
    "# Conv Layer\n",
    "model.add(Conv2D(filters=128,\n",
    "                 kernel_size=KERNEL_SIZE,\n",
    "                 activation='relu',\n",
    "                 ))\n",
    "# Max Pooling Layer\n",
    "model.add(MaxPooling2D(pool_size=POOL_SIZE))\n",
    "\n",
    "# 분류를 하기 위해 shape을 1차원으로 맞추는 Layer\n",
    "model.add(Flatten())\n",
    "model.add(Dense(units=1024, activation='relu'))\n",
    "model.add(Dropout(rate=DROP_RATE))\n",
    "\n",
    "# 마지막 분류를 하기 위한 Fully Connected Layer\n",
    "# One Hot Encoding을 했으므로 activation function은 softmax로 처리함\n",
    "model.add(Dense(units=10, activation='softmax'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 모델 컴파일"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:17.851928Z",
     "start_time": "2019-12-22T16:30:17.811386Z"
    }
   },
   "outputs": [],
   "source": [
    "model.compile(loss=categorical_crossentropy,\n",
    "              optimizer=Adam(),\n",
    "              metrics=['acc'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 모델 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:17.925135Z",
     "start_time": "2019-12-22T16:30:17.853010Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 26, 26, 64)        640       \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 13, 13, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 11, 11, 128)       73856     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 5, 5, 128)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 3200)              0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1024)              3277824   \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 1024)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 10)                10250     \n",
      "=================================================================\n",
      "Total params: 3,362,570\n",
      "Trainable params: 3,362,570\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.summary()\n",
    "\n",
    "plot_model(model, to_file='./img/model/mnist_cnn_model.png', show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 신경망 모델 학습\n",
    "\n",
    "Keras의 `fit`의 메소드를 통해서 간단하게 학습가능하다. 이번 예제에서는 여기서는 10번의 epoch만으로 학습을 진행하도록 한다. 그리고 앞서 준비한 검증 데이터(Validation Set)을 통해서 신경망의 오버피팅 여부를 판단하도록 하자."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:17.928266Z",
     "start_time": "2019-12-22T16:30:17.926298Z"
    }
   },
   "outputs": [],
   "source": [
    "EPOCHS = 3\n",
    "BATCH_SIZE = 64"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:49.075226Z",
     "start_time": "2019-12-22T16:30:28.321623Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 50000 samples, validate on 10000 samples\n",
      "Epoch 1/3\n",
      "50000/50000 [==============================] - 7s 141us/sample - loss: 0.1354 - acc: 0.9584 - val_loss: 0.0571 - val_acc: 0.9838\n",
      "Epoch 2/3\n",
      "50000/50000 [==============================] - 6s 125us/sample - loss: 0.0430 - acc: 0.9864 - val_loss: 0.0468 - val_acc: 0.9861\n",
      "Epoch 3/3\n",
      "50000/50000 [==============================] - 7s 132us/sample - loss: 0.0298 - acc: 0.9906 - val_loss: 0.0361 - val_acc: 0.9889\n",
      "CPU times: user 26.3 s, sys: 3.28 s, total: 29.6 s\n",
      "Wall time: 20.7 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "history = model.fit(X_train,  # 학습할 데이터\n",
    "                    y_train,  # 학습할 레이블\n",
    "                    epochs=EPOCHS,  # 전체 학습할 횟수\n",
    "                    batch_size=BATCH_SIZE,  # 배치 사이즈\n",
    "                    use_multiprocessing=True,\n",
    "                    validation_data=(X_valid, y_valid) # 검증 데이터로 확인\n",
    "                   )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:49.272708Z",
     "start_time": "2019-12-22T16:30:49.079208Z"
    }
   },
   "outputs": [],
   "source": [
    "# 학습된 모델 weight 저장 \n",
    "model.save('./bin/mnist_cnn.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 신경망 모델 평가\n",
    "\n",
    "위에서 학습된 결과를 시각화 해보자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:52.955617Z",
     "start_time": "2019-12-22T16:30:49.275034Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Loss : 0.017212021363414823\n",
      "Train Accuracy : 0.9953399896621704\n",
      "Validation Loss : 0.036114234882034364\n",
      "Validation Accuracy : 0.9889000058174133\n"
     ]
    }
   ],
   "source": [
    "# Train 데이터로 평가하기\n",
    "train_loss, train_acc = model.evaluate(X_train,y_train,verbose=0)\n",
    "print('Train Loss : {}'.format(train_loss))\n",
    "print('Train Accuracy : {}'.format(train_acc))\n",
    "\n",
    "# Validation 데이터로 평가하기\n",
    "valid_loss, valid_acc = model.evaluate(X_valid,y_valid,verbose=0)\n",
    "print('Validation Loss : {}'.format(valid_loss))\n",
    "print('Validation Accuracy : {}'.format(valid_acc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:53.172424Z",
     "start_time": "2019-12-22T16:30:52.958502Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3deXxU9bn48c+TdbKHLUDYwubCTgJxR6i1xd0qKlgRYhGL1Xqv1Vv016rX6r1ea+tWtbZUQIvElWoVpS7gUhUIqKwCYYcgYUtIQgJZnt8f5ySZhMmeyWR53q/XeTFz1mdODvPM93yXI6qKMcYYU11QoAMwxhjTOlmCMMYY45MlCGOMMT5ZgjDGGOOTJQhjjDE+WYIwxhjjkyUI41ciEiwi+SLStznXDSQRGSQifmkfXn3fIvIvEfmpP+IQkd+KyJ8bu71p/yxBmCrcL+jyqUxECr3e+/yiqo2qlqpqtKruas51WysR+UhE7vMx/2oR2SsiDfo/p6o/UtUFzRDXD0VkR7V9/05Vf97Uffs41gwRWdbc+zUtzxKEqcL9go5W1WhgF3CZ17yTvqhEJKTlo2zV5gFTfcyfCvxdVctaNhxjGs8ShGkQEXlIRF4RkYUikgfcICJnichXIpIjIvtE5CkRCXXXDxERFZEk9/3f3eXviUieiHwpIv0buq67/CIR2SwiuSLytIj8W0Sm1xB3fWK8RUQyReSIiDzltW2wiDwuIodEZCswsZZT9CbQQ0TO9tq+C3Ax8KL7/nIR+cb9TLtE5Le1nO/Pyz9TXXG4v9w3uvvdKiIz3PlxwD+Bvl6lwQT3bznPa/srRWS9e44+FpFTvZbtEZE7RWSte74Xikh4Leehps/TW0TeEZHDIrJFRG7yWnamiKwWkaMisl9Efu/OjxSRl93PnSMiK0Ska0OPbRrOEoRpjJ8ALwNxwCtACXAH0BU4B+eL65Zatr8e+C3QGaeU8ruGrisiCcCrwN3ucbcDqbXspz4xXgykAKNxEt8P3fmzgB8BI91jXFvTQVS1AHgduNFr9mRgjaqud9/nAzfgnL/LgDtE5NJaYi9XVxz7gUuAWOBm4GkRGaGque5xdnmVBrO9NxSR04G/A7cD3YAPgX+WJ1HXtcCFwACc8+SrpFSXV3D+VonAdcCjInK+u+xp4PeqGgsMwjmPAGlAJNAb6ALcChQ14timgSxBmMb4XFX/qaplqlqoqitVdbmqlqjqNuAvwPm1bP+6qmaoajGwABjViHUvBb5R1bfcZY8DB2vaST1j/F9VzVXVHcAyr2NdCzyuqntU9RDwSC3xAswHrvX6hX2jO688lo9VdZ17/r4F0n3E4kutcbh/k23q+Bj4CDivHvsFJ4m97cZW7O47FjjDa50nVPV799jvUPvf7SRu6S8VmK2qRaq6GphLZaIpBgaLSBdVzVPV5V7zuwKD3HqqDFXNb8ixTeNYgjCNsdv7jYicJiLvisj3InIUeBDnP3RNvvd6fQyIbsS6id5xqDPq5J6adlLPGOt1LGBnLfECfALkApeJyCk4JZKFXrGcJSLLROSAiOQCM3zE4kutcYjIpSKy3L19k4NT2qjvrZhE7/25dSV7gF5e6zTk71bTMQ66paxyO72OkQYMATa5t5EudufPwynRvCpORf8jYnVfLcIShGmM6k0rnwfW4fzCiwXuA8TPMezDueUAgIgIVb/MqmtKjPuAPl7va22G6yarl3BKDlOBxarqXbpJB94A+qhqHDCnnrHUGIeIRODckvlfoLuqxgP/8tpvXc1hs4B+XvsLwjm/e+sRV31lAV1FJMprXt/yY6jqJlWdDCQAfwDeEBGPqp5Q1QdU9XTgXJxbnA1uUWcazhKEaQ4xOL+YC9x72bXVPzSXd4BkEbnM/TV5B869c3/E+CrwHyLSy61w/nU9tpmPU89xE163l7xiOayqRSJyJs7tnabGEQ6EAQeAUrdO4wKv5ftxvpxjatn35SIy3q13uBvIA5bXsH5dgkTE4z2p6nYgA/gfEQkXkVE4pYYFACIyVUS6uqWXXJykViYiPxCRYW7SOopzy6m0kXGZBrAEYZrDr4BpOF8oz+NURPqVqu7HqeT8I3AIGAh8DRz3Q4zP4dzPXwuspLLytLb4tgIrAA/wbrXFs4D/FacV2L04X85NikNVc4D/BBYBh4FJOEm0fPk6nFLLDrclUEK1eNfjnJ/ncJLMROBytz6iMc4DCqtN4PzNBuPcrnoduFdVl7rLLgY2uuflMeA6VT2Bc2vqTZzksB7ndlPFLTvjP2IPDDLtgYgE49zCmKSqnwU6HmPaAytBmDZLRCaKSJzbWui3OE1ZVwQ4LGPaDUsQpi07F9iG07x1InClqtZ0i8kY00B2i8kYY4xPfi1BuLcANokzfMFsH8vHuV3rS0Rkko/lsW675z/5M05jjDEn81tnE7fS8Bmcrvl7gJUi8raqbvBabRcwHbirht38DqfTUZ26du2qSUlJjY63oKCAqKiouldsYRZXw1hcDWNxNUx7jGvVqlUHVdV3E3FV9csEnAUs8Xp/D3BPDevOw2l94j0vBadD0XTgT3UdLyUlRZti6dKlTdreXyyuhrG4Gsbiapj2GBeQoTV8r/qtDsK9ZTRRVctHlJwKnKGqt/lYdx7wjqq+7r4PAj7G6YV6ATCmhu1mAjMBunfvnpKent7oePPz84mObujIAf5ncTWMxdUwFlfDtMe4JkyYsEpVx/ha5s/xTHwNHVDfbHQrzvAEu50RFHxT1b/gDLrGmDFjdPz48Q2NscKyZctoyvb+YnE1jMXVMBZXw3S0uPyZIPZQddyY3jgdmerjLOA8EbkVZ0CwMBHJV9WTKrqNMcb4hz8TxEqcoXv74wzGNRlnbP86qWrFQFziPCxljCUHY1qH4uJi9uzZQ1GR/x7JEBcXx8aNG/22/8Zqy3F5PB569+5NaGhoret581uCUNUSEbkNWAIEAy+o6noReRCnUuRtERmLM3ZMJ5yhkf9bVYf6KyZjTNPt2bOHmJgYkpKSqO0WcFPk5eURE1PTuIKB01bjUlUOHTrEnj176N+/f43rVefXMdVVdTGwuNq8+7xer8RryOYa9jEPp5WTMaYVKCoq8mtyMM1PROjSpQsHDhxo0HY21IYxpsEsObQ9jfmbdfgEUVxaxv8s3sihwrJAh2KMMa1Kh08Qe48UsnDFLv6wqojcY40d+t4Y01IOHTrEqFGjGDVqFD169KBXr14V70+cOFGvfaSlpbFp06Za13nmmWdYsGBBc4TMueeeyzfffNMs+2pJHf65rkldo3h+agpT5yxn5ksZvPizVMJDggMdljGmBl26dKn4sn3ggQeIjo7mrruqjtZT3hM4KMj3b+C5c+fWeZxf/OIXTQ+2jevwJQiAswd2ZcbwcJZvP8zdr62hrMxGuDWmrcnMzGTYsGH8/Oc/Jzk5mX379jFz5kzGjBnD0KFDefDBByvWLf9FX1JSQnx8PLNnz2bkyJGcddZZZGdnA/Cb3/yGJ554omL92bNnM378eE499VS++OILwBkD6eqrr2bkyJFMmTKFMWPG1LukUFhYyLRp0xg+fDjJycl8+umnAKxdu5axY8cyatQoRowYwbZt28jLy+Oiiy5i5MiRDBs2jNdfr/Ohhs2iw5cgyp2VGEJcYhKPvr+JxPgIZl90WqBDMqbV++9/rmdD1tFm3eeQxFjuHN+3Udtu2LCBuXPn8uc//xmARx55hM6dO1NSUsKECROYNGkSQ4YMqbJNbm4u559/Po888gh33nknL7zwArNnn9ztSlVZtmwZS5cu5cEHH+T999/n6aefpkePHrzxxht8++23JCcn1zvWp556irCwMNauXcv69eu5+OKL2bJlC88++yx33XUX1113HcePH0dVeeutt0hKSuK9996riLklWAnCy6zzB3LDmX358ydbefHLHYEOxxjTQAMHDmTs2LEV7xcuXEhycjLJycls3LiRDRs2nLRNREQEF110EQApKSns2LHD576vuuqqk9b5/PPPmTx5MgAjR45k6ND6d+P6/PPPmTp1KgBDhw4lMTGRzMxMzj77bB566CEeffRRdu/ejcfjYcSIEbz//vvMnj2bf//738TFxdX7OE1hJQgvIsIDlw3l+9wiHnh7PT3jIrhwSPdAh2VMq3X/Zf7p15qXl9eo7byHvN6yZQtPPvkkK1asID4+nhtuuMFn7++wsLCK18HBwZSUlPjcd3h4+EnrNGWw05q2nTp1KmeddRbvvvsuF154IfPnz2fcuHFkZGSwePFi7r77bi699FLuvffeRh+7vqwEUU1IcBBPTRnN8F5x3L5wNV/vOhLokIwxjXD06FFiYmKIjY1l3759LFmypNmPce655/Lqq68CTt2BrxJKTcaNG1fRSmrjxo3s27ePQYMGsW3bNgYNGsQdd9zBJZdcwpo1a9i7dy/R0dFMnTqVO++8k9WrVzf7Z/HFShA+RIaF8LfpY7nq2S+YMT+DN2adTVLX1veQEGNMzZKTkxkyZAjDhg1jwIABnHPOOc1+jNtvv50bb7yRESNGkJyczLBhw2q8/fPjH/+4Yhyk8847jxdeeIFbbrmF4cOHExoayosvvkhYWBgvv/wyCxcuJDQ0lMTERB566CG++OILZs+eTVBQEGFhYRV1LH5X04Mi2trkjwcGbc3O01H/vUTPf/RjPZhX1KT9N1Z7fECJP1lcDdOYuDZs2ND8gVRz9OhRvx+jMarHVVxcrIWFhaqqunnzZk1KStLi4uKAx1UTX387anlgkN1iqsWAbtHMmTaGfblFzHgxg8ITpYEOyRjTiuTn53POOecwcuRIrr76ap5//nlCQtrPjZn280n8JKVfZ56cPJpZC1ZxR/rXPHdDCsFBNg6NMQbi4+NZtWpVoMPwGytB1MPEYT2479Ih/GvDfh785/omtVwwxpi2wkoQ9ZR2Tn/2Hilkzufb6dUpgpnjBgY6JGOM8StLEA1w78Wns+9oEf+z+Dt6xkVw2cjEQIdkjDF+YwmiAYKChD9cM5IDR4/zq1e/JSEmnDMGdAl0WMYY4xdWB9FAntBg/nJjCn06R3Dzixls2d+4Hp/GmMYZP378SZ3ennjiCW699dZat4uOjgYgKyuLSZMm1bjvjIyMWvfzxBNPcOzYsYr3F198MTk5OfUJvVYPPPAAjz32WJP305wsQTRCfGQY89JSCQ8NZvrclWQf9d/D240xVU2ZMoX09PQq89LT05kyZUq9tk9MTGzSaKjVE8TixYuJj49v9P5aM0sQjdSncyRzp4/lyLETpM1bSf5x3+O3GGOa16RJk3jnnXc4fvw4ADt27CArK4tzzz2X/Px8LrjgApKTkxk+fDhvvfXWSdvv2LGDYcOGAc6Q25MnT2bEiBFcd911FBYWVqw3a9asiqHC77//fgCee+45srKymDBhAhMmTAAgKSmJgwcPAvDHP/6RYcOGMWzYsIqhwnfs2MHpp5/OzTffzNChQ/nRj35U5Th18bXPgoICLrnkkorhv9944w0AZs+ezZAhQxgxYsRJz8hoDKuDaIJhveJ45qfJzJifwS8WrGbOtDGEBlvONR3Ie7Ph+7XNu88ew+Hc/1fj4i5dupCamsr777/PFVdcQXp6Otdddx0igsfjYdGiRcTGxnLw4EHOPPNMLr/88hqfx/zcc88RGRnJmjVrWLNmTZXhuh9++GE6d+5MaWkpF1xwAWvWrGHWrFk8++yzLF26lK5du1bZ16pVq5g7dy7Lly9HVTnjjDM4//zz6dSpE1u2bGHhwoX89a9/5dprr+WNN97ghhtuqPNU1LTPbdu2kZiYyLvvvgvAnj17OHz4MIsWLeK7775DRJrltpd9mzXRhFMTePjKYXyy+QD/b9Fa6yNhTAvwvs3kfXtJVbn33nsZMWIEP/zhD9m7dy/79++vcT+ffvppxRf1iBEjGDFiRMWyV199leTkZEaPHs369evrHIjv888/5yc/+QlRUVFER0dz1VVX8dlnnwHQv39/Ro0aBdQ+pHh99zl8+HA+/PBDfv3rX/PZZ58RFxdHbGwsHo+HGTNm8OabbxIZGVmvY9TGShDNYHJqX7JyCnnq40x6xUdyxw8HBzokY1rGRY/4Z791DPd95ZVXVoxqWlhYWPHLf8GCBRw4cIBVq1YRGhpKUlKSzyG+vfkqXWzfvp3HHnuMlStX0qlTJ6ZPn17nfmr7cVg+VDg4w4XX9xZTTfs85ZRTWLVqFYsXL+aee+7h/PPP5+GHH2bFihV89NFHpKen86c//YmPP/64XsepiZUgmsl/XngKVyf35vEPN/Naxu5Ah2NMuxYdHc348eO56aabqlRO5+bmkpCQQGhoKEuXLmXnzp217sd7yO1169axZs0awBkqPCoqiri4OPbv31/xJDeAmJgYn8+rGDduHP/4xz84duwYBQUFLFq0iPPOO69Jn7OmfWZlZREZGckNN9zAXXfdxbfffkt+fj65ublcfPHFPPHEE/V+9GltrATRTESE/71qOPuPFnHPm2vpHuth3CndAh2WMe3WlClTuOqqq6q0aPrpT3/KZZddxpgxYxg1ahSnnVb7o4NnzZpFWloaI0aMYNSoUaSmpgLO0+FGjx7N0KFDTxoqfObMmVx00UX07NmTpUuXVsxPTk5m+vTpFfuYMWMGo0ePrvftJICHHnqooiIanLoFX/tcsmQJd999N0FBQYSGhvLYY4+Rl5fHFVdcQVFREarK448/Xu/j1qimYV7b2uSP4b4b42jhCf3x45/okN++p+v25jR5f+1pmOiWYHE1jA333TBtPS4b7jvAYjyhzEtLJTYilLS5K9mbU//mbMYY05pYgvCDHnEe5qWlUlhcyvQXVpB7rDjQIRljTIP5NUGIyEQR2SQimSIy28fycSKyWkRKRGSS1/xRIvKliKwXkTUicp0/4/SHU3vE8PzUFHYcKmDmSxkcL7GHDZn2Q605d5vTmL+Z3xKEiAQDzwAXAUOAKSIypNpqu4DpwMvV5h8DblTVocBE4AkRaXN92c8e2JXHrhnJ8u2Hufu1NZSV2X8q0/Z5PB4OHTpkSaINUVUOHTqEx+Np0Hb+bMWUCmSq6jYAEUkHrgAqepuo6g53WZn3hqq62et1lohkA92ApncNbGFXjOrF3pxCHn1/E4nxEcy+qPZWFca0dr1792bPnj0cOHDAb8coKipq8JdZS2jLcXk8Hnr37t2g/fozQfQCvDsE7AHOaOhORCQVCAO2+lg2E5gJ0L17d5YtW9aoQMF5tmxTtq/N6ar8oE8If/5kKwUHdnNB39BWEVdTWFwNY3E1TH5+fsXoq61JW4+rrn4hJ6mpeVNTJ+AaYI7X+6nA0zWsOw+Y5GN+T2ATcGZdx2stzVxrUlxSqjfNXaH9Z7+j/1r/fb23a0/NI1uCxdUwFlfDtMe4CFAz1z1AH6/3vYGs+m4sIrHAu8BvVPWrZo6txYUEB/H09aMZ3iuO2xeu5utdRwIdkjHG1MqfCWIlMFhE+otIGDAZeLs+G7rrLwJeVNXX/Bhji4oMC+Fv08eSEONhxvwMdh4qCHRIxhhTI78lCFUtAW4DlgAbgVdVdb2IPCgilwOIyFgR2YNzO+p5EVnvbn4tMA6YLiLfuNMof8XakrpGhzMvbSxlqkx7YQWH8o8HOiRjjPHJr/0gVHWxqp6iqgNV9WF33n2q+rb7eqWq9lbVKFXtok6zVlT176oaqqqjvKamjzzVSgzoFs2caWPYl1vEjBczKDxhfSSMMa2P9aQOkJR+nXly8ii+2Z3DHelfU2p9JIwxrYwliACaOKwn9106hH9t2M+D/1xvHY+MMa2KDfcdYGnn9GfvkULmfL6dXp0imDluYKBDMsYYwBJEq3Dvxaez72gR/7P4O3rGRXDZyMRAh2SMMZYgWoOgIOEP14zkwNHj/OrVb0mICeeMAV0CHZYxpoOzOohWwhMazF9uTKFP5whufjGDLftrfyavMcb4myWIViQ+Mox5aamEhQQzfe5Kso/W/pB0Y4zxJ0sQrUyfzpHMnT6WI8dOkDZvJYUl1rLJGBMYliBaoeG943jmp8l8930ez35znOLSsro3MsaYZmYJopWacGoCD185jLUHS/nNonXWR8IY0+IsQbRik1P7ctnAUF7J2M1TH2UGOhxjTAdjzVxbuasGhRIal8DjH24mMd7DNWP61L2RMcY0A0sQrZyI8MhVI8g+epx73lxL91gP407pFuiwjDEdgN1iagPCQoJ47oZkBiVEM+vvq1iflRvokIwxHYAliDYixhPKvLRUYiNCSZu7kr05hYEOyRjTzlmCaEN6xHmYl5ZKYXEpaXNXkFtYHOiQjDHtmCWINubUHjE8PzWF7QcLuOWlDI6X2MOGjDH+YQmiDTp7YFd+P2kkX207zN2vraHMHjZkjPEDa8XURl05uhdZuYU8+v4mEuMjmH3RaYEOyRjTzliCaMNmnT+QrJxC/vzJVnp1imDqmf0CHZIxph2xBNGGiQgPXDaUfTlF3P/WOnrEerhwSPdAh2WMaSesDqKNCwkO4unrRzO8Vxy3L1zN17uOBDokY0w7YQmiHYgMC+Fv08eSEONhxvwMdh4qCHRIxph2wBJEO9E1Opx5aWMpU2XaCys4lH880CEZY9o4SxDtyIBu0cyZNoZ9uUXMeDGDwhPWR8IY03iWINqZlH6deXLyKL7ZncMd6V9Tan0kjDGNZAmiHZo4rCf3XTqEf23Yz+/e2WAPGzLGNIo1c22n0s7pz94jhcz5fDu94iO4edyAQIdkjGlj/FqCEJGJIrJJRDJFZLaP5eNEZLWIlIjIpGrLponIFnea5s8426t7Lz6dS4b35OHFG/nnt1mBDscY08b4rQQhIsHAM8CFwB5gpYi8raobvFbbBUwH7qq2bWfgfmAMoMAqd1tr5N8AQUHCH64dSXZeEb969VsSYsI5Y0CXQIdljGkj/FmCSAUyVXWbqp4A0oErvFdQ1R2qugYoq7btj4EPVPWwmxQ+ACb6MdZ2yxMazF9vHEOfzhHc/GIGmdl5gQ7JGNNG+DNB9AJ2e73f487z97ammvjIMOalpRIWEsy0F1aSfbQo0CEZY9oA8VcLFxG5Bvixqs5w308FUlX1dh/rzgPeUdXX3fd3A+Gq+pD7/rfAMVX9Q7XtZgIzAbp3756Snp7e6Hjz8/OJjo5u9Pb+0pxx7cgt5X9XFNEjKojZqR4iQqRVxNWcLK6Gsbgapj3GNWHChFWqOsbnQlX1ywScBSzxen8PcE8N684DJnm9nwI87/X+eWBKbcdLSUnRpli6dGmTtveX5o7r4+/264B73tUb/7ZcT5SUNno/HeV8NReLq2EsroZpSlxAhtbwverPW0wrgcEi0l9EwoDJwNv13HYJ8CMR6SQinYAfufNME004NYGHrxzGJ5sP8JtF66yPhDGmRn5LEKpaAtyG88W+EXhVVdeLyIMicjmAiIwVkT3ANcDzIrLe3fYw8DucJLMSeNCdZ5rB5NS+3P6DQbySsZunPsoMdDjGmFbKrx3lVHUxsLjavPu8Xq8Eetew7QvAC/6MryO788JT2JtTyOMfbiYx3sM1Y/oEOiRjTCtjPak7KBHhkatGkH30OPe8uZbusR7GndIt0GEZY1oRG4upAwsLCeK5G5IZlBDNrQtWsz4rN9AhGWNaEUsQHVyMJ5R5aanEeEJIm7uSvTmFgQ7JGNNKWIIw9IjzMC8tlcLiUtLmriC3sDjQIRljWgFLEAaAU3vE8PzUFLYfLOCWlzI4XmIPGzKmo7MEYSqcPbArv580kq+2Hebu19ZQZg8bMqZDs1ZMpoorR/ciK7eQR9/fRGJ8BLMvOi3QIRljAsQShDnJrPMHkpVTyJ8/2UqvThFMPbNfoEMyxgSAJQhzEhHhgcuGsi+niPvfWkePWA8XDuke6LCMMS3M6iCMTyHBQTx9/WiG94rj9oWr+XqXPavJmI7GEoSpUWRYCHOmjSUhxsOM+RnsPFQQ6JCMMS3IEoSpVbeYcOaljaVMlelzV3K44ESgQzLGtBBLEKZOA7pFM2faGLJyCvnZ/JUUnrA+EsZ0BJYgTL2k9OvMk5NH8c3uHO5I/5oye46EMe2eJQhTbxOH9eS+S4fwrw37eXnjCXvYkDHtnCUI0yBp5/Rnxrn9+XBXCXM+2x7ocIwxfmQJwjTYvRefztgewTy8eCP//DYr0OEYY/zEOsqZBgsKEm4eHg6ecH716rckxIRzxoAugQ7LGNPMrARhGiUsWPjrjWPo0zmCm1/MIDM7L9AhGWOaWb0ShIgMFJFw9/V4EfmliMT7NzTT2sVHhjEvLZWwkGCmvbCS7KNFgQ7JGNOM6luCeAMoFZFBwN+A/sDLfovKtBl9Okcyd/pYjhw7Qdq8leQfLwl0SMaYZlLfBFGmqiXAT4AnVPU/gZ7+C8u0JcN7x/HM9cl8930ev1iwmuLSskCHZIxpBvVNEMUiMgWYBrzjzgv1T0imLZpwWgIPXzmMTzYf4DeL1lkfCWPagfq2YkoDfg48rKrbRaQ/8Hf/hWXaosmpfdmbU8jTH2fSq1MEv7xgcKBDMsY0Qb0ShKpuAH4JICKdgBhVfcSfgZm26c4LT2FvTiF//GAzPeM8XDOmT6BDMsY0Un1bMS0TkVgR6Qx8C8wVkT/6NzTTFokIj1w1gnMHdeWeN9fy6eYDgQ7JGNNI9a2DiFPVo8BVwFxVTQF+6L+wTFsWFhLEczckMyghmlsXrGZD1tFAh2SMaYT6JogQEekJXEtlJbUxNYrxhDIvLZUYTwhp81awN6cw0CEZYxqovgniQWAJsFVVV4rIAGCL/8Iy7UGPOA/z0lI5dqKUtLkryC0sDnRIxpgGqFeCUNXXVHWEqs5y329T1avr2k5EJorIJhHJFJHZPpaHi8gr7vLlIpLkzg8VkfkislZENorIPQ37WKa1OLVHDM9PTWH7wQJueSmD4yX2sCFj2or6VlL3FpFFIpItIvtF5A0R6V3HNsHAM8BFwBBgiogMqbbaz4AjqjoIeBz4P3f+NUC4qg4HUoBbypOHaXvOHtiV308ayVfbDnP3a2soK7M+Esa0BfW9xTQXeBtIBHoB/3Tn1SYVyHRLGyeAdOCKautcAcx3X78OXCAiAigQJSIhQMT+JNYAAB7gSURBVARwArCazjbsytG9uPvHp/L2t1k8umRToMMxxtSD1KfHq4h8o6qj6ppXbfkkYKKqznDfTwXOUNXbvNZZ566zx32/FTgDyAVeAi4AIoH/VNW/+DjGTGAmQPfu3VPS09Pr/Cw1yc/PJzo6utHb+0t7iktVeXHDCZbuLuHGIWH8oG/zd8ZvT+erJVhcDdMe45owYcIqVR3jc6Gq1jkBHwI3AMHudAPwUR3bXAPM8Xo/FXi62jrrgd5e77cCXYBzgAU4w3kkAJuAAbUdLyUlRZti6dKlTdreX9pbXMUlpXrT3BXaf/Y7+q/13zdvUNr+zpe/WVwN0x7jAjK0hu/V+t5iugmniev3wD5gEs7wG7XZA3h3o+0NVH/8WMU67u2kOOAwcD3wvqoWq2o28G/Ad4YzbUpIcBBPXz+a4b3iuH3har7ZnRPokIwxNahvK6Zdqnq5qnZT1QRVvRKn01xtVgKDRaS/iIQBk3HqMby9jTMAIDhJ52M3o+0CfiCOKOBM4Lt6fibTykWGhTBn2li6xYTzs3kr2XmoINAhGWN8aMoT5e6sbaE6w4PfhtN/YiPwqqquF5EHReRyd7W/AV1EJNPdX3lT2GeAaGAdTqKZq6prmhCraWW6xYQzPy2VMlWmz13J4YITgQ7JGFNNU55JLXWtoKqLgcXV5t3n9boIp66i+nb5vuab9mVAt2jmTBvD9X9dzoz5K3n55jPxhAYHOixjjKspJQhrzG6aLKVfZ56cPIqvd+dwR/rXlFofCWNajVoThIjkichRH1MeTp8IY5ps4rCe3HfpEJas38/v3tlgDxsyppWo9RaTqsa0VCCmY0s7pz97jxQy5/Pt9IqP4OZxAwIdkjEdXlPqIIxpVvdefDr7cot4ePFGesR5uGykFVKNCSRLEKbVCAoS/nDtSLLzivjVq9+SEBPOGQO6BDosYzqsplRSG9PsPKHB/PXGMfTpHMHNL2aQmZ0X6JCM6bAsQZhWJz4yjHlpqYSFBDPthZVkHy0KdEjGdEiWIEyr1KdzJHOnj+XIsRPcNH8l+cdLAh2SMR2OJQjTag3vHccz1yezcV8ev1iwmuLSskCHZEyHYgnCtGoTTkvgoSuH8cnmA/xm0TrrI2FMC7JWTKbVm5Lal6ycQp7+OJNenSL45QWDAx2SMR2CJQjTJtx54SnszSnkjx9spmech2vG9Kl7I2NMk1iCMG2CiPDIVSPIPnqce95cS/dYD+NO6RbosIxp16wOwrQZYSFBPHdDMoMSorl1wWo2ZNljyo3xJ0sQpk2J8YQyLy2VGE8IafNWsDenMNAhGdNuWYIwbU6POA9z08Zy7HgpaXNXkFtYHOiQjGmXLEGYNum0HrE8PzWF7QcLuOWlDI6XlAY6JGPaHUsQps06e1BXfj9pJF9tO8x/vb6GMnvYkDHNyloxmTbtytG92JtTyO+XbCIxPoIzPIGOyJj2wxKEafNuHT+QvTmFPLdsKwVDwhgf6ICMaSfsFpNp80SEBy8fygWnJfDShhN8sGF/oEMypl2wBGHahZDgIJ6+fjRJcUHcvnA13+zOCXRIxrR5liBMuxEZFsJ/JHvoFhPOz+atZOehgkCHZEybZgnCtCtx4cL8tFTKVJk+dyWHC04EOiRj2ixLEKbdGdAtmjnTxpCVU8iM+SspKrY+EsY0hiUI0y6l9OvMk5NH8fXuHO5I/5pS6yNhTINZgjDt1sRhPfntJUNYsn4/v3tngz1syJgGsn4Qpl276dz+ZOUUMufz7fSKj+DmcQMCHZIxbYZfSxAiMlFENolIpojM9rE8XERecZcvF5Ekr2UjRORLEVkvImtFxPrImka59+LTuWR4Tx5evJF31mQFOhxj2gy/JQgRCQaeAS4ChgBTRGRItdV+BhxR1UHA48D/uduGAH8Hfq6qQ4HxgA3ZaRolKEj4w7UjGZvUiTtf+Zbl2w4FOiRj2gR/liBSgUxV3aaqJ4B04Ipq61wBzHdfvw5cICIC/AhYo6rfAqjqIVW1piim0Tyhwfz1xjH06RzBzS9mkJmdF+iQjGn1xF8VdyIyCZioqjPc91OBM1T1Nq911rnr7HHfbwXOAG4AUoAEoBuQrqqP+jjGTGAmQPfu3VPS09MbHW9+fj7R0dGN3t5fLK6GqSuuA8fK+N1XRYQGwW/P9BDvaZl2Gm31fAWKxdUwTYlrwoQJq1R1jM+FquqXCbgGmOP1firwdLV11gO9vd5vBboAdwHbga5AJPAlcEFtx0tJSdGmWLp0aZO29xeLq2HqE9ea3Tl6+m/f00ue+lTzi4r9H5S27fMVCBZXwzQlLiBDa/he9efPpz1AH6/3vYHqNYQV67j1DnHAYXf+J6p6UFWPAYuBZD/GajqQ4b3jeOb6ZDbuy+PWBaspLi0LdEjGtEr+TBArgcEi0l9EwoDJwNvV1nkbmOa+ngR87Ga0JcAIEYl0E8f5wAY/xmo6mAmnJfDQlcP4ZPMBfrNonfWRMMYHv/WDUNUSEbkN58s+GHhBVdeLyIM4RZq3gb8BL4lIJk7JYbK77RER+SNOklFgsaq+669YOZ7vt12b1mtKal+ycgp5+uNMenWK4JcXDA50SMa0Kn7tKKeqi3FuD3nPu8/rdRFOXYWvbf+O09TVvwpz4P/6cU5IDGweBPF9Ib5f5b+d+kFcHwiL9HsopuXdeeEp7M0p5I8fbKZnnIdrxvSpeyNjOgjrSS0CF9zPgQ1fkRhRAvvXw6b3ofR41fWiEtyk0ddJGhWJpB/E94GQ8MDEb5pERHjkqhFkHz3OPW+upXush3GndAt0WMa0CpYgPHFw3p1sLl1G4vjxzryyMsjfDzm73GmH8++RnZD1NWx8G8pKqu4npmfV0kdFEunrlECCQ1v6k5l6CgsJ4tkbkrn2z19y64LVvHrLWQxJjA10WMYEnCUIX4KCILanM/U94+TlZaWQt68yaeTsghz3391fwbo3wLtfnwRBTGK1kodXaSQmEYLtTxFIsZ5Q5qWl8pNn/03avBUsuvUcEuMjAh2WMQFl30qNERQMcb2dqd/ZJy8vLYajWZVJwzuRbP/UWYZXq5mgEIjtVbXew30dXnTQSUhBwS328TqqHnEe5qaN5ZrnvmT63BW89vOziYuwkp/puCxB+ENwqPMl36mf7+UlJyB3t9ctrJ2VSSTzQ8j/vmLVswBW/Nyp56hyCyupshQS3d2pSzFNdlqPWJ6fmsK0uSu45aUM5t+USniIJWfTMVmCCISQMOgy0Jl8KS5yE8hONi//gFMSPJVJZNNiKDhQbX8ep57Du96jvAK9Uz+I7GIJpAHOHtSV308ayX+88g3/9foaHr92FEFBdv5Mx2MJojUK9UDXwdB1MFl7QjilvPK83IkCyNntdQtrZ+UtrL2roPBItf1FeSWO6q2w+kJEJ0sg1Vw5uhd7cwr5/ZJNJMZH8OuJpwU6JGNanCWItigsChJOcyZfio563b6qdgtr15dw/GjV9cNjq5U8qrXC8sT5/zO1QreOH8jenEKeW7aVxPgIpp5Zwy1DY9opSxDtkScWegxzJl8Kc6omjYqK9B2w7RMoLqi2v3ivpOEkji4Hj8J+t29IeOsb3bI5iAgPXj6U/blF3P/WOnrEerhwSPdAh2VMi7EE0RFFxDtTz5EnL1OFY4fdBFKtFdaBzbDlQygpZDjAuoecbSK7+Ch9eHUiDG27zUVDgoN4+vrRTPnLV9y+cDXpM89iVJ/4QIdlTIuwBGGqEoGoLs7Uy8cAuqpQcIBVHy8ipX+Xqrew9q9zKtFLT1TdJiqhhj4gSU5T4VbeCz0yLIQ508Zy1XP/5mfzVvLmrWfTr0tUoMMyxu8sQZiGEYHoBPJiT4Xh409eXtEL3fsWllcF+oa3qvVCl8pe6Ce1wurrJJBW0Au9W0w489JSufq5L5g+dyVvzDqbzlFhgQ7LGL+yBGGaV5Ve6GeevLys1O1EuOvkW1g7v4S1r4F6PZ9Bgqp2IqzeCis2scU6EQ7sFs3fpo3h+r8uZ8b8lbx885l4Qq2PhGm/LEGYlhUU7Hb66wOcc/Ly0mI4utf3MCbbP6m5F7qbNPodUfh2f2Uiie7hJK1mktKvM09OHsWsBau5I/1rnv1pCsHWR8K0U5YgTOsSHOrUTXRKgv4+lpcch9w9vlthbfmA/vn7YccCr/2FOZ0Iq/QBqWyNRXRCg/uATBzWk99eMoQH39nA797ZwP2XDUGsH4lphyxBmLYlJLzWXuiffrSEcSP6e5U8vBLJd+/CsYPV9ueppQ9Izb3Qbzq3P1k5hcz5fDu94iO4edwAf3xaYwLKEoRpV8qCw6HbKc7ky4mCagMoVqtEr6kXuo9WWPdO6Me+nEIeXryRnvEeLh2R6P8PaEwLsgRhOpawKEg43Zl8KcqtOoyJ9y2snV9U6YUeBPwpPIa7o7uR+Xpnvt84nB79Tq2aSDz2XAnTdlmCMMabJw56xPnuha4KRTlVkobk7KTXoR3o9u+I3fgybKz2JMKIThDfl6HFkXDiI3eU3yTo1N+pGwmxprKm9bIEYUx9iThf+BGdIHFUxexQIOTwMcY9828SQgr4+9Xd6Vz8fZVWWFF7N8LyVdU6EYrbAivJnfpVvo7v16gKdGOakyUIY5pBn86RzE1L5bq/fMmNS0p4ZeZlRIVX/vdasWwZ48eNc571cWSHO+2sfL31I+cphd5CIyuHbPdOHOXJJMx6cxv/sgRhTDMZ3juOZ65PZsaLGdy6YDVzpo0hNNirD0ZQkNOxLzbR95MIiwsr6z3KE0eO+3rH53Aiv+r65UOYnJQ8klq0A6FpvyxBGNOMJpyWwENXDuOeN9fym0XreOTq4fXvIxEaAd1OdabqVOHYITd5bK+aPHYvh3VvVn0OelCo0xmxeuIoTygRnZr6UU0HYAnCmGY2JbUvWTmFPP1xJr06RfDLCwY3faciENXVmXqnnLy8tNjpQFi95HFkB2R9A4WHq67viYNOSV6V50lu8rDKc1PJEoQxfnDnhaewN6eQP36wmcT4CLr6+4DBodC5vzP5UpRb2e/Dqw4kau+GhlWed0qCqG5Wed5BWIIwxg9EhEeuGkH20ePMfmMNl/QP4Xi37xmcEE3fzpGEBDff+FD14omDniOcyYvvyvMdlfUgmR86y7xVVJ4n+agDscrz9sQShDF+EhYSxLM3JDNjXgZvbT3MW1tXOfODg+jfNYpBCdEMSohmcHfn3/5dowgPCUDFcr0rz3ecXIG+4zOrPG/HLEEY40exnlBe/flZvPfhUhJPHc2W7Hy2ZOexNTufdVm5LF63D3UHpw0S6NfFSRyDy5NHQgwDE6KIDAvgf9V6VZ7vqDrl7HQrz9+oOny7d+W5rwp0qzxvVSxBGNMCIkKEkX3iGVntcaVFxaVsO1BQkTS2ZOeTmZ3P0u+yKSmrHNa8V3yEU9LoVl7iiGFQQjRxEQF+mFKVyvMxJy+3yvM2za8JQkQmAk8CwcAcVX2k2vJw4EUgBTgEXKeqO7yW9wU2AA+o6mP+jNWYQPCEBjMkMZYhiVXHbCouLWPnoQIys/PZsj+fzAPOv19uPcTxkspf5Akx4ZUlju4xFQmkS1RY6xiCvL6V59WSR409z+N6+262a5XnfuG3BCEiwcAzwIXAHmCliLytqhu8VvsZcERVB4nIZOD/gOu8lj8OvOevGI1prUKDg9xSQgwTvYaFKi1T9h4pZEt2XkVpY0t2Pm+s3kv+8cpHucZHhrq3qWIqEsjg7tH0iPW0jsRRrq7K87x9J7W8anDleackZ+BEqzxvMH+WIFKBTFXdBiAi6cAVOCWCclcAD7ivXwf+JCKiqioiVwLbgAI/xmhMmxIcJPTtEknfLpFccHr3ivmqyvdHi5zShps0tmbn8966feQcK65YLzo8hIEJlbeqirJL6H+ogN6dIlvfk/GCgiCulzPVWXm+o2oF+vZPobjaV0dUQg0tr5Ks8rwGoqp1r9WYHYtMAiaq6gz3/VTgDFW9zWudde46e9z3W4EzgELgQ5zSx11Avq9bTCIyE5gJ0L1795T09PRGx5ufn090dHSjt/cXi6thLK6qVJW8E5BVUEZWvjsVlJGVr+Qcr/y/HxoEPaKC6BUtJEYH0TMqiF7RQSRECiEBSBxNPl+qhBYfxVP0PRGF+/EU7Sei8Hs8Rdl4ir7HU3QQofJWXZmEUOTpRpGnO4URPSjydK/yuiQ0unni8pOmxDVhwoRVquqjAsm/JQhfV1X1bFTTOv8NPK6q+bUVh1X1L8BfAMaMGaPjx49vXKTAsmXLaMr2/mJxNYzFVX+5hcW89v6nxPQeXFHqyMzO56t9hRXrhAQJSV2jKlpVlU8Du0XjCfXfL26/n6/SYsjdXVHqCDqyg8icnUQe2QFHlp/84Ci38jy7JIqEwWOq3r5qBZXn/jpf/kwQe4A+Xu97A1k1rLNHREKAOOAwTilikog8CsQDZSJSpKp/8mO8xnQocRGhDOoUzPixfavMP3aihK3ZBWQeyKu4ZbXp+zyWrP+e8oZVItC3cySDukUzqLvTHLc8eUSHt4HGkcGh0HmAM/lSQ+V59N4NsDyjauW5BDk9z9th5bk//5IrgcEi0h/YC0wGrq+2ztvANOBLYBLwsTr3vM4rX0FEHsC5xWTJwZgWEBkWwvDecQzvHVdl/vGSUnYcPMaW7LzKEsf+fD7bcpATpZW3a3rGeSr6cFR0BOwWTaeoNtREtT6V59Wb7R7ZWY/K86ST60DCIlviEzWK3xKEqpaIyG3AEpxmri+o6noReRDIUNW3gb8BL4lIJk7JYbK/4jHGNE14SDCn9ojh1B4xVeaXlJax6/CxiltU5dPCFbsoLK4cYbZrdFhl73Gv1lXdYsJbV8uqunhXnnPOyctPqjzfUVkaqW/leXnyCHDluV/Lgqq6GFhcbd59Xq+LgGvq2McDfgnOGNMsQoKDGNAtmgHdovnx0Mr5ZWXK3pxCMg84JY1Mtxf5W99kkVdU2SQ31hNSkTQGd49mYEI0hwrLKCtTglpby6r6aEzP84ph2331PO9b89AlEfEnHaI5tYGbhcaYtigoSOjTOZI+nSOZcGpCxXxV5UDeca9+HE5dx0ff7eeVjN0V6/32iyUVpYyBXsOPBGSww+ZSr57nu6s22S2/lZX1dY2V54OC+kAbq6Q2xpiTiAgJsR4SYj2cM6jqQOhHCk6QeSCfdz9bhcQnkpmdz5fbDvHm13sr1gkLDmJAt6gqSWNwQgxJXSMDM9hhc2pI5blXHUhwfpFfwrEEYYxpNTpFhTE2qjMFO0IZP77yflVeUTFbDxSwZX9exS2rtXtyWby2crDD4CChX+fIKiPkDk6IYUC3AA922JxqqDzftGwZPf1wuHZy1owx7VmMJ5RRfeIZ5WOww60HKivGy8et+rjaYIe9O0VUGyE3unUMdtjKWYIwxrRZntBghibGMTSxapPcEyVl7DpcwJb9+VXGrPr31kOc8BrssHtseJWkMdidukSHt/RHaZUsQRhj2p2wkMrBDi/yml9apuw5cqzKCLmZ2Xm8lrGbghOVTXI7RYZWTRruLSt/DU3UWlmCMMZ0GMFBQr8uUfTrEsUPqTrY4b7cIq8hR5zOgO+t28dCr8EOPcFw2oZ/n/RQp96dItpmk9w6WIIwxnR4IkJifASJ8RGMO6VbxXxV5VDBiYqSxrKvN1EYGswnmw/w+qo9Fet5QoMY0DX6pIc69esSSWhbbZKLJQhjjKmRiNA1Opyu0eGcNbALfY7vYPz4MwHIPVZM5oG8isrxLdn5ZOw4wlvfVA45FxosJHk/RtZ9qNOAblF+HeywuViCMMaYRoiLDCWlX2dS+nWuMr/geEnFY2TLK8i/qzbYYVD5YIfVHuo0sJUNdth6IjHGmHYgKtz3YIdFxaXsOFRQMUJu+fTJ5gMUl1ZWfifGeao8Pra8riM+suUHO7QEYYwxLcATGsxpPWI5rUfV54+XlJax8/Axr74cTmfAl7cfoqi4sklu1+hwBiVEVYxZVT7Uejc/Nsm1BGGMMQEUEhzEwG7OQ5h8DnbojldV3sLqH9/sPWmww9Pi1R9DMVmCMMaY1qjKYIenVR3sMDvveEXLqi3Z+eQe2OeXGCxBGGNMGyIidI/10D3Ww7mDncEOly075Jdjtd0GusYYY/zKEoQxxhifLEEYY4zxyRKEMcYYnyxBGGOM8ckShDHGGJ8sQRhjjPHJEoQxxhifpL08IUlEDgA7m7CLrsDBZgqnOVlcDWNxNYzF1TDtMa5+qtrN14J2kyCaSkQyVHVMoOOozuJqGIurYSyuhulocdktJmOMMT5ZgjDGGOOTJYhKfwl0ADWwuBrG4moYi6thOlRcVgdhjDHGJytBGGOM8ckShDHGGJ/afYIQkRdEJFtE1tWwXETkKRHJFJE1IpLstWyaiGxxp2ktHNdP3XjWiMgXIjLSa9kOEVkrIt+ISEYLxzVeRHLdY38jIvd5LZsoIpvcczm7heO62yumdSJSKiKd3WX+PF99RGSpiGwUkfUicoePdVr0GqtnTIG6vuoTW4tfY/WMq8WvMRHxiMgKEfnWjeu/fawTLiKvuOdkuYgkeS27x52/SUR+3OAAVLVdT8A4IBlYV8Pyi4H3AAHOBJa78zsD29x/O7mvO7VgXGeXHw+4qDwu9/0OoGuAztd44B0f84OBrcAAIAz4FhjSUnFVW/cy4OMWOl89gWT3dQywufrnbulrrJ4xBer6qk9sLX6N1SeuQFxj7jUT7b4OBZYDZ1Zb51bgz+7rycAr7ush7jkKB/q75y64Icdv9yUIVf0UOFzLKlcAL6rjKyBeRHoCPwY+UNXDqnoE+ACY2FJxqeoX7nEBvgJ6N9exmxJXLVKBTFXdpqongHSccxuIuKYAC5vr2LVR1X2qutp9nQdsBHpVW61Fr7H6xBTA66s+56smfrvGGhFXi1xj7jWT774NdafqLYuuAOa7r18HLhARceenq+pxVd0OZOKcw3pr9wmiHnoBu73e73Hn1TQ/EH6G8wu0nAL/EpFVIjIzAPGc5RZ53xORoe68VnG+RCQS50v2Da/ZLXK+3KL9aJxfed4Cdo3VEpO3gFxfdcQWsGusrnPW0teYiASLyDdANs4PihqvL1UtAXKBLjTD+QppbNDtiPiYp7XMb1EiMgHnP/C5XrPPUdUsEUkAPhCR79xf2C1hNc7YLfkicjHwD2AwreR84RT9/62q3qUNv58vEYnG+cL4D1U9Wn2xj038fo3VEVP5OgG5vuqILWDXWH3OGS18jalqKTBKROKBRSIyTFW96+L8dn1ZCcLJqn283vcGsmqZ32JEZAQwB7hCVQ+Vz1fVLPffbGARDSw2NoWqHi0v8qrqYiBURLrSCs6XazLViv7+Pl8iEorzpbJAVd/0sUqLX2P1iClg11ddsQXqGqvPOXO1+DXm7jsHWMbJtyErzouIhABxOLdjm36+mrtSpTVOQBI1V7peQtUKxBXu/M7AdpzKw07u684tGFdfnHuGZ1ebHwXEeL3+ApjYgnH1oLKDZSqwyz13ITiVrP2prEAc2lJxucvL/2NEtdT5cj/7i8ATtazTotdYPWMKyPVVz9ha/BqrT1yBuMaAbkC8+zoC+Ay4tNo6v6BqJfWr7uuhVK2k3kYDK6nb/S0mEVmI0yqiq4jsAe7HqehBVf8MLMZpZZIJHAPS3GWHReR3wEp3Vw9q1SKlv+O6D+c+4rNOfRMl6ozW2B2nmAnOf5iXVfX9FoxrEjBLREqAQmCyOldjiYjcBizBaW3ygqqub8G4AH4C/EtVC7w29ev5As4BpgJr3fvEAPfifAEH6hqrT0wBub7qGVsgrrH6xAUtf431BOaLSDDOHZ9XVfUdEXkQyFDVt4G/AS+JSCZO8prsxrxeRF4FNgAlwC/UuV1VbzbUhjHGGJ+sDsIYY4xPliCMMcb4ZAnCGGOMT5YgjDHG+GQJwhhjjE+WIIypgztq5zdeU3OOIpokNYxQa0ygtft+EMY0g0JVHRXoIIxpaVaCMKaR3GcA/J87Xv8KERnkzu8nIh+J86yFj0Skrzu/u4gscgeh+1ZEznZ3FSwif3XH+/+XiES46/9SRDa4+0kP0Mc0HZglCGPqFlHtFtN1XsuOqmoq8CfgCXfen3CG9x4BLACecuc/BXyiqiNxnm1R3gt4MPCMqg4FcoCr3fmzgdHufn7urw9nTE2sJ7UxdRCRfFWN9jF/B/ADVd3mDvT2vap2EZGDQE9VLXbn71PVriJyAOitqse99pGEM4TzYPf9r4FQVX1IRN4H8nFGM/2HVj4XwJgWYSUIY5pGa3hd0zq+HPd6XUpl3eAlwDNACrDKHanTmBZjCcKYprnO698v3ddf4A6YBvwU+Nx9/REwCyoeAhNb005FJAjoo6pLgf8C4oGTSjHG+JP9IjGmbhFeI3wCvK+q5U1dw0VkOc6PrSnuvF8CL4jI3cAB3NFbgTuAv4jIz3BKCrOAfTUcMxj4u4jE4QxF/bg6zwMwpsVYHYQxjeTWQYxR1YOBjsUYf7BbTMYYY3yyEoQxxhifrARhjDHGJ0sQxhhjfLIEYYwxxidLEMYYY3yyBGGMMcan/w8xMw+b12ntRwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Train / Validation 데이터에 대해서 Loss 시각화\n",
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "epochs = range(1,len(loss)+1)\n",
    "\n",
    "plt.plot(epochs,loss,label='Training Loss')\n",
    "plt.plot(epochs,val_loss,label='Validation Loss')\n",
    "plt.title('Training and Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:53.302155Z",
     "start_time": "2019-12-22T16:30:53.173478Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Train / Validation 데이터에 대해서 Accuracy 시각화\n",
    "acc = history.history['acc']\n",
    "val_acc = history.history['val_acc']\n",
    "epochs = range(1,len(loss)+1)\n",
    "\n",
    "plt.plot(epochs,acc,label='Training Accuarcy')\n",
    "plt.plot(epochs,val_acc,label='Validation Accuarcy')\n",
    "plt.title('Training and Validation Accuarcy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuarcy')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 신경망 모델 검증하기\n",
    "\n",
    "MNIST 데이터셋을 학습한 모델을 놀랍게도 정확도 99% 이상의 성능을 보인다.  \n",
    "그렇다면 이러한 신경망이 학습에 전혀 사용되지 않는 데이터인 Test 데이터셋에서는 어떤 결과를 보이는지 살펴보자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:53.774630Z",
     "start_time": "2019-12-22T16:30:53.303204Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Validation Loss : 0.03085688237536233\n",
      "Validation Accuracy : 0.9898999929428101\n"
     ]
    }
   ],
   "source": [
    "test_loss, test_acc = model.evaluate(X_test,y_test,verbose=0)\n",
    "print('Validation Loss : {}'.format(test_loss))\n",
    "print('Validation Accuracy : {}'.format(test_acc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test 데이터에 대해서도 99%의 정확도를 보여준다.  \n",
    "\n",
    "그러면 과연 학습된 신경망이 어떠한 데이터를 잘 못 예측했는지 살펴보자."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-22T16:30:55.882964Z",
     "start_time": "2019-12-22T16:30:53.776502Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArsAAAK7CAYAAADyY0eCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdd5wV1f0+8OcRlI6IIKGvXYEo6irGgihBsWBLUWxoJGhE/ZIoBo35qiEF840m0Z8lEJGiBFAU0djQhGCNggERsQCCUoRFQJoN+Pz+mNl43XOWnb11ZvZ5v177Yvdz78ycuTw7e+7cOXNoZhARERERSaOdSt0AEREREZFCUWdXRERERFJLnV0RERERSS11dkVEREQktdTZFREREZHUUmdXRERERFIrVZ1dkjNIDiz2sqVE8mKSL2b8vInkXlms53ySz+a3dRJVHc1uL5LLMn6eT7JXFus5luS7eW2cRFZHs6vjbsLV0dzW2WNuLDu7JJeQ/G6p21GJgV+Q/JDkBpITSTaPuGwZSQsPhpvCfRtWqLaaWVMzWxyxTfUzlnvQzE7Md3vCX4pNVb6M5Pfyva04iFt2AYDkVSQ/CLM7i+QxtVjWSG4O/9+Wk7ydZL1CtNPMuprZjIht2idjuRfMbP9CtInkUSRfI7mR5Ju1ee2SJm7Z1XE3P0gOCLebuM5ZFHHLLaBjbi7C35N/ktxC8p18/d/GsrMbQxcBuBDA0QDaAWgE4M5arqOFmTUF0B/A/5LsW/UJmQfBtAh/KZpWfgE4DcAmAE+XuGl1AskeAEYA+D6AXQHcB+DRWh48Dw7/73oDOA/Ajz3bSV12SbYEMA3A/wFoAeD3AB4nuVtJG1Z36LibozCr1wOYX+q21BU65ubsbwD+A2B3AL8A8DDJ1rmuNFGdXZK7kXyCZAXJdeH3Hao8be/wTMynJB8L/2BVLn8kyZdJric5txan7/sBuM/MPjKzTQBuBXAOyca13QczewXBgadb2CYjOZjk+wDeD2sHkJxOci3Jd0n+MGMfdic5LXzH+BqAvTPXn/kOjGQjkreRXBq+Hi+SbARgZvj09eG7x+/Q/VjuKJKvh8u9TvKojMdmkBxO8qXwjNezJFtFfAkGAHjYzDbX9rVLshJmtwzAfDObbcF0ieMAtAKwR233wczeAfACvs7uEpI/J/kmgM0k65NsR3JKuJ8fkLw6Yx8akRwT7v/bAA7PXH/mGRqS9UjeQHJRmLHZJDuSrMzu3DC759D9aO7AMKPrGXxMd3rGY2NI3kXy7+F6/03yG79DGY4CsMrMHjKzbWb2AIAKAGfX9rVLMh13E33c/R2AOwCsqe1rlnQ65ibvmEtyPwCHArjJzD4zsykA5gHI+ZPgRHV2EbT3fgCdAXQC8BmA/1flORcB+BGCMwFbEfyig2R7AH8H8GsALQFcC2AKPe8YSHYK/9M6VZbCL2T83ADAvrVpPANHA+iK4J1LpTMB9ADQhWQTANMBTEDwy9EfwN0ku4bPvQvA5wDahvv5ox1s8g8ADkPwR7slgOsAbAfQM3y8RXjG9ZUq7WyJ4LW6A8G7q9sB/J3k7hlPOw/AJWEbd0HwelYu/ybJ8zz73xjBu92xO2hzWpUqu08BqEeyB4MzCz8CMAfAx7XdAZJdAByLb2a3P4BTEZz53A7gcQBzAbRHcFZiCMmTwufehKCTsDeAkxC88anOz8J1nwKgedjuLWZWmd2Dw+xOqtLGncM2PIsgm1cBeJBk5kdu/QHcAmA3AAsB/CZj+Sf49cfdVX/vK2vddtDuNNJxN4HHXZJHACgHcO8O2ppmOuYm75jbFcBiM9uYsezcsJ4bM4vdF4AlAL4b4XndAazL+HkGgBEZP3cB8CWAegB+DmB8leWfATAgY9mB1WxnIID3ELxj2xXBR5sG4DsR2lgWPnc9gHUAFgC4OuNxA3BCxs/nAHihyjr+giC09QB8BeCAjMd+C+DFKuvbB8Ev+mcIAlpdm+pn1C6uXA+Cjw5fq7LMKwAuznitbsx47AoAT0d4LS4E8AEAljpjdSi7BHBDmJutCM7wHF6L/TEAG8LsLkJw8N8pY19/lPHcHgA+rLL89QDuD79fDKBvxmODACzzvXYA3gVwxg7atE/Gz70q14PgD8PHlW0Ma38DcHP4/RgAf8147BQA71Sznd0R/N72B7Azgj8U2wH8pdQ5qyPZ1XE3y+Nu2OZZla/Vjl7npH/FMLc65mZ/zL0QwKtVar8BMCbXnCTqmo/wzOAfAfRF8A4BAJqRrGdm28KfP8pYZCmCP1KtELy7+wHJfhmP7wzgnxE2PRpARwQBrw/gNgQfsS3bwTJVtTKzrdU8ltnmzgB6kFyfUasPYDyA1uH3VffRuz0ADRH8stRWO896lyJ451gp813qFgBNI6x3AIBxFia4LilhdgcieIfeFcE76hMBPEHyEDNbEbH5h5rZwmoeq5rddlWyWw/Bx3BAkKso2QWC37dss/uRmW2vsp1aZ9fMPiF5BoIzdXch+GP3HGr3e594Ou4m8rh7BYA3rcrZ47pEx1wACTvmIhjPU3UQanMAGz3PrZWkXcZwDYD9AfQws+b4+mOhzI+6OmZ83wnBu6s1CP7Dx5tZi4yvJmY2oqaNmtl2M7vJzMrMrAOCa7+Wh1/5kNn5+wjAv6q0s6mZ/QTB9YJbPfvoswbBx26+a2Nq6myuQPBLlKkTcthfkh0RvBscl+06Eq4k2QVwMIDHzey9MMdPA1iJ4CPWfKia3Q+qtLOZmZ0SPr4S0bJbua7qrqXdkRUAOpLMPLZlnV0z+5eZHW5mLRGcddgfwGvZrCvBdNxN3nG3N4CzSH5M8mMEv++3kaz6MX6a6ZibvGPufAB7kWyWUTsYeRhgGefO7s4kG2Z81QfQDMFHROvD65tu8ix3Acku4bu6XyEYDLUNwAMA+pE8KbwQu2F4kXXVC9YdJFuS3Du89qsLgmupflX5TobkzSRn5Ge38QSA/UheSHLn8OtwkgeG+/EIgJtJNg7b4r0GJ2zbaAC3M7iAvR6DARENEBy8twOo7r6QT4ZtOI/BBfDnIPiI54kc9utCAC+bWTbvHJMmNtkF8DqAU0nuFea3D4D9ALwF/Pd+oUty32UAQSdwA4MBFI3CtnYjWTkoYjKA6xkMHOmA4Nqu6vwVwHCS+4btPijj2sVVqD67/wawGcB14e9OLwRnAydms0MkDwnX0xzBGd5lZvZMNutKiNhkV8fdnI67FwM4EMFH990RXNJwC4LR7WkUm9xCx9xeyPKYa2bvIbi++abwNT8LwEEAptR2XVXFubP7JIKgVn7dDOBPCG4/swbAq/Dfvmo8gmtEPkbwcdLVAGBmHwE4A8G1NBUI3sUMhec1YHDB+SZ+fcF5q7A9mxFcfD7azEZmLNIRwEtZ72kGCy7MPhHAuQjeMX2MYBRyg/ApVyL4CODjcD/v38HqrkUwkvF1AGvD9exkZlsQXAfzEoML64+s0oZPENwi7BoAnyAYYHGamUUa0ctgNOb5VcoXoe4MTItTdschOOjMQHAd2B0ALrNglC+Q3+xuQ3CQ647g2uw1CA6gu4ZPuQXBx1sfIBjMMH4Hq7sdwYH62bDd9yF4/YDg9RwbZveHmQuZ2ZcATgdwcrj9uwFclLG/O0TyKZI3ZJSuw9dnetoCOCvKehIsTtnVcTfL466ZrTezjyu/EFyLusHMPo2yrgSKU251zM3tmHsugoGV6xDews3MKqKsa4fbqYOXT+YdyTkAeocHK5HEYDB70/+Y2YJSt0WkNnTclSTSMbc01NkVERERkdSK82UMIiIiIiI5UWdXRERERFJLnV0RERERSa2cJpUg2RfAnxHcxPivNd2DrlWrVlZWVpbLJqUOmj179hozc6ZpzIWyK4W2ZMkSrFmzpup0wzmrTXaVW8mGjrmSVNVlN+vOLoM5n+8C0AfBjDavk5xmZm9Xt0xZWRlmzZqV7SaljiK5o1lfslmfsisFV15envd11ja7yq1kQ8dcSarqspvLZQxHAFhoZovD+6xNRHBfOpG4U3YlqZRdSSLlVkoql85ue3xzzuVl+OZcyAAAkoNIziI5q6Ii5/sCi+SDsitJVWN2lVuJIR1zpaRy6ez6rkVzbtprZiPNrNzMylu3zuslQCLZUnYlqWrMrnIrMaRjrpRULp3dZQimvavUAcE0iyJxp+xKUim7kkTKrZRULp3d1wHsS3JPkrsgmM94Wn6aJVJQyq4klbIrSaTcSkllfTcGM9tK8koAzyC4lchoM5uft5aJFIiyK0ml7EoSKbdSajndZ9fMngTwZJ7aIlI0yq4klbIrSaTcSilpBjURERERSS11dkVEREQktdTZFREREZHUUmdXRERERFJLnV0RERERSS11dkVEREQktdTZFREREZHUUmdXRERERFJLnV0RERERSS11dkVEREQktdTZFREREZHUql/qBoiIiIik1aZNm5zaxIkTndqQIUOc2meffebUjjjiCKd2xx13OLXDDz88ahNTT2d2RURERCS11NkVERERkdRSZ1dEREREUkudXRERERFJrZwGqJFcAmAjgG0AtppZeT4aJVJoyq4klbIrSVRXcrt27VqndtVVVzm1SZMmRVofSaf2+uuvO7XevXs7tenTpzu1Hj16RNpu2uTjbgzHm9maPKxHpNiUXUkqZVeSSLmVktBlDCIiIiKSWrl2dg3AsyRnkxzkewLJQSRnkZxVUVGR4+ZE8kbZlaTaYXaVW4kpHXOlZHLt7B5tZocCOBnAYJI9qz7BzEaaWbmZlbdu3TrHzYnkjbIrSbXD7Cq3ElM65krJ5HTNrpmtCP9dTfJRAEcAmJmPhhXa559/7q0/99xzTu20004rdHPw5ZdfOrXOnTs7tQkTJji1d99916kNHDjQqdWvrwnzKiU5u7Wxfft2pzZlyhSntnz58qy3cd111zm1efPmObV99tnHqdWrVy/r7dZVdSW7hbZ+/Xqn5puZ6osvvnBqs2fPdmqtWrXKT8NSKsm5NTNvfeZMt/k//vGPndqiRYucmm/gWS42b97s1MaNG+fU6uoAtazP7JJsQrJZ5fcATgTwVr4aJlIoyq4klbIrSaTcSqnlcqqvDYBHw3cn9QFMMLOn89IqkcJSdiWplF1JIuVWSirrzq6ZLQZwcB7bIlIUyq4klbIrSaTcSqnp1mMiIiIiklp1dsRSw4YNvfWePZ0BopgzZ45T+/a3v+3Uog622bRpk1PzXUh+7LHHOrXbbrvNqf3rX/9yaueee65Ta9GiRaT2STJNmzbNqU2dOtWpjR07Nq/b9Q3e6NKli1N76KGHnNrZZ5+d17aI+PzjH/9waoMHD3ZqCxcujLQ+X259g5UkHXx/nwHg0ksvjbS8r79w4403OrUOHTpEWt/555/v1JYsWeLUnnnmGafmG3DZoEGDSNtNMp3ZFREREZHUUmdXRERERFJLnV0RERERSS11dkVEREQktdTZFREREZHUqrN3Y9i6dau3PmjQIKc2efJkp/bKK684Nd80fL5piZ999lmnNmrUKKd25ZVXOrULLrjAqX3yySdO7c4773Rqv/zlL52aJE91U/sOGTLEqS1dujTSOn1T+UYdofvWW9EmQrr++uud2qmnnhppuxs2bHBqO+3kvldv2rRppLZIOqxZs8ap/elPf3JqI0aMcGq+qbQ7duzo1Hx53LZtW9QmSsL4pjr39Quq4+sHPPfcc06tUaNGtWtYBt/U1L67MfhqkyZNcmq77rqrU/PdUWffffeN1sAY0pldEREREUktdXZFREREJLXU2RURERGR1FJnV0RERERSq84OUNuyZYu3/vjjjzu1X/ziF07tiCOOiLSdl156yalNmTLFqY0ZM8apHXTQQU6NpFPzDVD77W9/69T69evn1Lp37+7UJD58U0v7ppEG/IPRfFNE33fffU7tlFNOcWq77LJLlCZ6p0n1TVftm4rVN2Wrb2DE//3f/zm1K664wqlpEGZ6+Qaj/eEPf3Bqv//97yOtzzdo7eqrr3Zqf//7353aoYceGmkbEm++gerjx4+P9LzqnHjiiU4tl8FoX375pVPz/V3wTdvuc8kll2Tdlosvvtip3XPPPd7nRv37USw6sysiIiIiqaXOroiIiIikljq7IiIiIpJaNXZ2SY4muZrkWxm1liSnk3w//He3wjZTpPaUXUkqZVeSSLmVuIoyQG0MgP8HYFxGbRiA581sBMlh4c8/z3/zCqe62aHmzJnj1HKZNaR169ZOzTfDysEHHxxpfb4Bb8cdd5xT880OdNtttzk138X4KTIGCc/uF1984dSizooG+GcoO/PMM3NqU1U9e/Z0ar4ZBn0D6+6///5I2zjhhBOc2nXXXRdp2YQag4RnNxcVFRVO7ZhjjnFq77//fqT13XrrrU7tZz/7mVOrV6+eU/ve977n1D777DOn5huU+fLLLzu1q666yqkNHTrUqSXUGCQot3PnznVqt99+u1Pr37+/d/mJEyc6taiZ9PENhPPl5d1333VqvsHrvloufAPpqxuI5pvFtX790t0TocYzu2Y2E8DaKuUzAIwNvx8LIL9/PUXyQNmVpFJ2JYmUW4mrbK/ZbWNmKwEg/HeP/DVJpKCUXUkqZVeSSLmVkiv4ADWSg0jOIjnL99GUSFwpu5JEyq0klbIrhZJtZ3cVybYAEP67uronmtlIMys3s3Lf9asiRabsSlJFyq5yKzGjY66UXLZXC08DMADAiPDfx/LWoiKpboBa1MFomzdvdmq+mal8g9EuuuiiSNvwDTLzDcrxPc834C1FgyBykajs+i7+79Chg/e5y5Ytc2orV650ar5Bb9X9PmSrc+fOTu3ZZ591al27do20Pt/gtny3OQESld2ofDOj5Xsw2k9/+lOn5huMFtWrr77q1CZPnhxp2bvuusuppfzYHNvc7rGHe0WFb5awP//5z97lfQPUfLUmTZo4tcaNG0daNpcz3Mcff7xT883g6huYP3DgQKe2aNEipzZq1KjI7alutrViiHLrsb8BeAXA/iSXkbwUQWj7kHwfQJ/wZ5FYUXYlqZRdSSLlVuKqxjO7Zua/5wbQO89tEckrZVeSStmVJFJuJa40g5qIiIiIpJY6uyIiIiKSWqWbziLhxo4d69QefPBBpzZo0CCn1rBhw0jb+Otf/+rUfDP3+Oy1115OrUuXLpGWlfho1qyZUxsxwn/J27Bhw5zaP/7xD6c2ePBgp3b33Xc7tepmxslW1NHVbdq0cWo/+clP8toWiQ/frJBRB6Odd955Ts03GC2XmZt+/nN3sq9cBtoMHz4862Ulv3zH1/Ly8rxvx/e3PN+zm/Xp08epPfLII07NNzDONwumbxDmYYcd5tQ++ugjb3sef/xxpxbrAWoiIiIiIkmlzq6IiIiIpJY6uyIiIiKSWursioiIiEhqaYBalj788EOn1rFjR6c2Y8YMp9a7t3vLQd9AIt+sK76ZTnwDHt5++22nlssgDYmP/v39t7L0DXi49tprndr999/v1GbOnOnUTjrpJKfmm8HPl/tcTJo0yalp6lDxad68uVPL5Tjn+z34y1/+4tQ2bdoUaX2jR492at///vdr3zApiBYtWji1yy+/3Kn5ZikFgN/97ndO7frrr8+9YRk6derk1AYMGODUfMfmRo0aZb3dli1bOrU77rjDqZ111lne5X0zv02YMMGp+QaZFoLO7IqIiIhIaqmzKyIiIiKppc6uiIiIiKSWOrsiIiIikloasZSlc845x6n9+c9/dmpXXHGFU5s7d65TO/roo53atm3bnFrXrl2dmm92lnHjxjk1Sbdzzz3XqX3nO9+JVFu0aJFT882qNn78eKfmy9rpp59ebTtrsuuuu2a9rCSPbwCOb5bJzz//3Kk99NBDTu2dd97Jui1vvPGGU9uwYUOkZZs2berUjjzySKeWy6AhKY2ddvKfFxw6dKhTmzx5slObPXt2pO34Brf9+te/jrRsMRx++OFOzcy8z926datT883opgFqIiIiIiI5UmdXRERERFJLnV0RERERSS11dkVEREQktWocoEZyNIDTAKw2s25h7WYAPwZQOUXGDWb2ZKEaGUeHHHKIUxszZoxT88204xvI4DNx4kSnNn/+fKd2xBFHODXfIKS6RtkFOnfu7NSWLl3q1D766COnduONNzo13+xmZ555plNr166dU7vqqquqbWemF154wakddNBBkZZNi7qUXd+x1DfIxzco+JNPPnFqvlkr8813DJ82bZpTO+CAAwreljipS7mtjm8wm292y7PPPtup3XLLLQVpU74sX77cqfn2rTrLli1zal988YVTa9CgQe0aFkGUM7tjAPT11P9oZt3Dr9QGVxJtDJRdSaYxUHYlecZAuZUYqrGza2YzAawtQltE8krZlaRSdiWJlFuJq1yu2b2S5JskR5PcrbonkRxEchbJWRUVFdU9TaSYlF1Jqhqzq9xKDOmYKyWVbWf3HgB7A+gOYCWA26p7opmNNLNyMytv3bp1lpsTyRtlV5IqUnaVW4kZHXOl5LKaQc3MVlV+T3IUgCfy1qKUiToY7b333nNqP/rRj5yab7YS36Ch+vU1OZ6PsgvsvPPOTm2vvfZyahMmTHBqw4cPd2oHHnigU1uxYoVT880O5Mvzscce69SkbmX3tNNOc2q+Abv/+7//69R8M1Tmonnz5k5t6tSpTq1Xr1553W5a1KXc1sYee+zh1OrVq1eClkT3+OOP57T8p59+6tR8M8UWQlZndkm2zfjxLABv5ac5IoWl7EpSKbuSRMqtxEGUW4/9DUAvAK1ILgNwE4BeJLsDMABLAFxWwDaKZEXZlaRSdiWJlFuJqxo7u2bW31O+rwBtEckrZVeSStmVJFJuJa40g5qIiIiIpJZGMcWEb/adzz//3Km1adPGqflmqxIphL333tupvfWWewleeXm5U9u8ebNT882+s2XLlixbJ2nWr18/p3b88cc7tXHjxjm1K6+8Muvtjh492qlpMJrURb4Z0GqjW7duTq1x48Y5rTMqndkVERERkdRSZ1dEREREUkudXRERERFJLXV2RURERCS1NECtBHwDz373u99FWvaKK65war4ZsUSKZb/99nNqQ4cOdWo333xzpPWNGDHCqflmrBLxzVA5b968SMv6Zu877LDDnNpJJ51U+4aJJNwLL7zg1O6//36n5htkXB3frLDFojO7IiIiIpJa6uyKiIiISGqpsysiIiIiqaXOroiIiIiklgaolYBvxql169Y5tY4dOzo138AfkTR57bXXnNq7777r1Pbff/9iNEdibPHixU5t0qRJkZb1DW7zDRRu0qRJ7Rsmddb111/v1M4++2yndu+99zq1a6+91qntueee+WlYaPv27U5twYIFTu373/9+pGV32sl/zvRb3/qWUzv55JOjNLEgdGZXRERERFJLnV0RERERSS11dkVEREQktdTZFREREZHUqnGAGsmOAMYB+BaA7QBGmtmfSbYEMAlAGYAlAH5oZu4oK3HMnTvXqQ0cONCp+Wb48Q1ka9SoUX4aljLKbnF89dVXTm38+PFZr2/VqlVO7Te/+Y1TGzduXNbbiDPlNrq3337bqa1fvz7Ssv3793dqffr0yblNdZmyC+y9995OzTfLmK/WtWtXp3b33Xc7Nd/gXF9/4Z133nFqjz76qFN76qmnnJqPbzBadTOo+WbCLKUoZ3a3ArjGzA4EcCSAwSS7ABgG4Hkz2xfA8+HPInGi7EoSKbeSVMquxFKNnV0zW2lmb4TfbwSwAEB7AGcAGBs+bSyAMwvVSJFsKLuSRMqtJJWyK3FVq2t2SZYBOATAvwG0MbOVQBBwAHtUs8wgkrNIzqqoqMittSJZUnYliZRbSSplV+IkcmeXZFMAUwAMMbMNUZczs5FmVm5m5a1bt86mjSI5UXYliZRbSSplV+Im0gxqJHdGENwHzeyRsLyKZFszW0myLYDVhWpkkn3xxRdObePGjU5t2rRpTq1evXpO7c4778xPw+oIZbfwVq92X75FixY5tbKyMqd26KGHOrVHHnnEqT3zzDNO7ZNPPnFqu+++e3XNTBTl1uUbCHnddddFWrZhw4ZO7cYbb8y5TeKq69n99re/7dTOPNO9auOxxx5zal9++aVT8w1e9/ENUKtu8Fi2mjdv7tTuuOMO73MvuOCCvG47VzWe2WXwat0HYIGZ3Z7x0DQAA8LvBwBw/+dESkjZlSRSbiWplF2Jqyhndo8GcCGAeSTnhLUbAIwAMJnkpQA+BPCDwjRRJGvKriSRcitJpexKLNXY2TWzFwFUdy68d36bI5I/yq4kkXIrSaXsSlxpBjURERERSS11dkVEREQktSLdjUGyt337dqf23nvvObUuXbo4tZ/97GdOzTeqWKSUHnrooUjP27p1q1PbtGlTpGV9uffdrUTS66WXXnJqvulQfSZOnOjUOnXqlHObRKKYMGGCU7vkkkuc2uTJk4vRnEiuvfZap+a7+0nLli2L0Zyc6cyuiIiIiKSWOrsiIiIiklrq7IqIiIhIaqmzKyIiIiKppQFqBdaoUSOnNmzYMKf28MMPO7W+ffsWpE0ipbBs2bJINZ9HH33UqbVo0SLnNklyTJkyJetlu3fvnseWiNROgwYNnNr999/v1HxTAw8ZMsSpNWvWzKm9+uqrTs038P3CCy90ahdddJFT8029Xr9+cruMOrMrIiIiIqmlzq6IiIiIpJY6uyIiIiKSWursioiIiEhqJfdq44R48sknndoFF1zg1KZOnerUknwxuEi2TjzxRKd28MEHl6AlEifPP/98qZsgkje+QWsnnHCCU3vzzTeL0ZzU05ldEREREUktdXZFREREJLXU2RURERGR1Kqxs0uyI8l/klxAcj7J/wnrN5NcTnJO+HVK4ZsrEp2yK0mk3EpSKbsSV1FGQG0FcI2ZvUGyGYDZJKeHj/3RzP5QuObF1/Lly53aokWLnNqKFSuc2t133+3UevbsmZ+GSSZltwguu+wyp3bNNddEWrZ9+/ZOzTezUL169WrfsORSbj18A9QOOeQQp7Z69epiNEf8lF2JpRo7u2a2EsDK8PuNJBcAcP9CicSMsitJpNxKUim7Ele1umaXZBmAQwD8OyxdSfJNkqNJ7pbntoVUbTsAACAASURBVInkjbIrSaTcSlIpuxInkTu7JJsCmAJgiJltAHAPgL0BdEfwTu62apYbRHIWyVkVFRV5aLJI7Si7kkTKrSSVsitxE6mzS3JnBMF90MweAQAzW2Vm28xsO4BRAI7wLWtmI82s3MzKW7duna92i0Si7EoSKbeSVMquxFGN1+ySJID7ACwws9sz6m3D63MA4CwAbxWmifHkG1jjq2ngWekou8XRqFEjp7Zt27YStCQdlFu/tm3bOrWPP/64BC2R6ii7EldR7sZwNIALAcwjOSes3QCgP8nuAAzAEgDukGyR0lJ2JYmUW0kqZVdiKcrdGF4EQM9DT+a/OSL5o+xKEim3klTKrsSVZlATERERkdRSZ1dEREREUkudXRERERFJLXV2RURERCS11NkVERERkdRSZ1dEREREUkudXRERERFJLZpZ8TZGVgBYCqAVgDVF23BhaV8Kr7OZlXTuSGU31uK6H3HKLRDf16m20rIfQHz3RdktjLTsBxDfffFmt6id3f9ulJxlZuVF33ABaF/qljS9RmnZl7TsR6Gl5XVKy34A6dqXQkrL65SW/QCSty+6jEFEREREUkudXRERERFJrVJ1dkeWaLuFoH2pW9L0GqVlX9KyH4WWltcpLfsBpGtfCiktr1Na9gNI2L6U5JpdEREREZFi0GUMIiIiIpJa6uyKiIiISGoVvbNLsi/Jd0kuJDms2NvPBcnRJFeTfCuj1pLkdJLvh//uVso2RkGyI8l/klxAcj7J/wnriduXYlJ2S0/ZrT3lNh6U3dpTdksvLbktameXZD0AdwE4GUAXAP1JdilmG3I0BkDfKrVhAJ43s30BPB/+HHdbAVxjZgcCOBLA4PD/IYn7UhTKbmwou7Wg3MaKslsLym5spCK3xT6zewSAhWa22My+BDARwBlFbkPWzGwmgLVVymcAGBt+PxbAmUVtVBbMbKWZvRF+vxHAAgDtkcB9KSJlNwaU3VpTbmNC2a01ZTcG0pLbYnd22wP4KOPnZWEtydqY2UogCAWAPUrcnlohWQbgEAD/RsL3pcCU3ZhRdiNRbmNI2Y1E2Y2ZJOe22J1demq691mJkGwKYAqAIWa2odTtiTllN0aU3ciU25hRdiNTdmMk6bktdmd3GYCOGT93ALAiXysnOYPkwCIvu4pk23AdbQGszmb72SJ5MckXM37eRHKvCMvtjCC4D5rZIyTPB7BzKfcl5pTdPFN2i6KguQVKkt2S5jbc7s0kHwi/7xRmt16E5apm94awruy6dMzNs7qc22J3dl8HsC/JPUnuAuBcANOqPonkEpLfLXLbqsXAL0h+CGAegA4km4cPTwMwIPx+AIDHqixbRtLCUG0K961gF3KbWVMzW7yj54QfRXwJ4B0zuz1c7kEAd2MH+5ItkiPDEbXbSV6cj3WWQFKzezzJeSTXA/gPgE4kKz8KVHZrEO7/5ozX4K/5WG8RRcotEL/sAgDJ8wC8CKALyakkW6KG3IbL9QqPN5tIbgyPP5cUoo1m9mGY3W07eh7JXgDWA1iQkd3fAhiFPGeX5PkZmd1EckuY5cNyXXcRJfWY+9/sAZgP4ECSlf+/NR1z63puu5CcRXJd+PUc8zQosaidXTPbCuBKAM8guMh5spnNL2YbsnQRgAsRdBY2A2gOYCXJSwGMANCH5PsA+oQ/+7Qws6YA+gP4X5JVR2mCZP1CNN6jPPz3eJJzwq9TEH1famsugCsAvJGn9RVdgrP7NoCTADwF4HMATQAsVHZr5eDwj0JTM8vqTFCpJDi3INkVwcCXRgC2ATgRwNOI/n+9IsxtcwA/BzDK94eziNn9NoDGAE4odHbN7MGMzDZFcPxdjAQdg5OcXQRnoB8HsAHAdgC/qcUxt87mFsHr9n0ALQG0QvDmYGIe1guYWey+ACwB8F1PfTcATwCoALAu/L5DxuMzAPwOwGsAPkXwTqNlxuNHAngZwbuUuQB6VVl2YDXteRjA0Iyfj0LQcWgcYV/KEFxnVD+j9jqAa8PvDcBgAO8D+CCsHQBgOoKRnO8C+GHGsruHAdgQ7udwAC9mPG4A9gm/bwTgNgBLw9fjxbD2Yfi8TeHXdwBcXGU9R4Xt/DT896gqr9VwAC8B2AjgWQCtIrwWLwK4uNT5qkvZrdKGBuE23o64L3U+u5ltSvtX3LIL4LcAJmT8vDeCs/rNIuxLLwDLqtQqEPwhrcz1pWGeZkZo554A/hVmZjqA/wfgAd/vCYI/1Pcj+MO9DsBUBG8yP0PQ8anMbjsAN1euJ1z2dARnA9eHr82BVf5/rgXwZvg6TwLQMOL/7T8B3FTqjNWR3DrZq8W+KLdfL1cfwd+XLXnJSamDWsvw7g7gewjeZTQD8BCAqVUCuBxAt/A/aUrGf2x7AJ8AOAXBGe0+4c+tq4YXQKfwP61T+PMUANdlbOfoMCQHR9iX/wYKwQX3RwPYAqB3+LiFIWyJ4I95EwQjUC8JlzkUwBoAXcPnTwQwOXxet3B/q+sw3BXuV3sA9RB0AhpUDXn43Isr1xO2ZR2Cs9n1EZzRWwdg94zXahGA/cI2zwAwImNdbwI4z/Na1OXObkmyW6W2HcBXUf8PlN3/tmkFgI8BPAKgrNQZqyvZRdD5+HmVtmwCcFiEfemFsNMQbvesMPv7Z2RoXNjeRhHa+QqA28MM9kTQeaiu0/B3BH/QdwOwM4DjqrYpo503Z6xnPwSfHPYJl7sOwEIAu2T8/7yGoLPREsHZzssz1rUewDGe16IzgjPje5Y6Y3Ukt70QvClbBeADAH8E0CTivii3X9e2IvibdWNeclLqoNYmvJ7ndQewrkp4M/9wdQlDVw/BxwHjqyz/DIABVcPr2c5AAO+F4dgVwdkpA/CdCG2sDNR6BH90FwC4OuNxA3BCxs/nAHihyjr+AuCmcD++AnBAxmO/hafDEAb/M3g65FVDHtYuxtcdhgsBvFZlmVcQdpLC1+rGjMeuAPB0hNeiznZ2S5XdKsu0DNd1ZMR9qfPZRfAHYhcALRCcFXkrc9tp+opbdhHcqP7yKrXlyDhztYM29kLwh3I9gk8Z5gA4t0qG9sp4frXtRNCZ2YqMDguACfB0GgC0Dbe7WzVt2lGn4ZcIPqqvfGynzP0N/38uyHj89wDujfBa/BLAjFLnqw7l9lvhunZCcGZ1JoC/RNwX5fbr5zVBcHw+NR85KdY1H3lBsjGCd0l9Ebz7AIBmJOvZ1xdZZ96XbymCdxqtELy7/QHJfhmP74zg452ajEYwKnQGgmDcBqAfgtGiUbWy4Bokn8w2dwbQIxxQVKk+gPEAWoffV91H7/YANERwFqu22nnWuxTfvMfhxxnfbwHQNIvt1BklzO5/mdlakmMBzCXZfgd5rKrOZteCG8MDwJcMpsncAOBABANV64QSZncTgusWMzVHcHYqihVm1mEHj1fNbnXtbIegk7Q547Gl+OadAip1BLDWzNZFbGOmb2TXzLaT/Ag7zm67COu9CMEbyzqlVLk1s4/x9f/TBySvQ3DW9LKITVdug/VsJnkvgAqSB5pZTnd7SFRnF8A1CE7n9zCzj0l2RzBoLPN+fJn/kZ0QnE1agyAg483sx7XdqJltR3B26iYAIHkigncuy7PZCd8mMr7/CMC/zKxP1ScxuEXIVgT7+E5Y7lTNOtcguK54bwTX8VS3PZ8VCH6JMnVCMDhEslOS7HrUR3Dz7+ZwZ/fJRl3LrsF//880K1V25wM4uPIHBrela4DgU7Z8qJpdbztJdgawG8kmGR2HTvBn8SMALUm2MLP1VR6Lkt1vZ2yXCF7XrP/OkDwaQcfi4WzXkWBxOebm+5iR+txm2AnBZSjtkeOtzYp967Ha2Jlkw4yv+giuu/kMwPrwFjQ3eZa7ILx9RWMAvwLwcPgu7gEA/UieRLJeuM5eJHf0DgoAQLIlyb0Z6ILgGphfhZ3gynvXzcjPbuMJAPuRvJDkzuHX4eE7m20Irhu8mWTjsC0DfCsJ2zYawO0k24X7/B2SDRBc8L4dQHX3NH0ybMN5JOuTPAfBxzJPZLNDJHch2RDBL3zl/2ucs5erOGX3bJL7k9yJZGsE2f2Pma0NH1d2q0GyK8nu4fabIvhEZzmCyznSKjbZBfBguOyxJJuE633EgilLQXIMyTG57zKwo3aa2VIAswDcEh7LjkHwyZ7DgpmkngJwN8ndwt+BnuHDqwDsTnLXatowGcCpJHszuK/pNQC+QDD4KFsDAEypfM1SLDa5DZ/XKewvdERwl4LHMh5XbqtBsg/JQ8J9aY7g71XlJXQ5iXOH40kEQa38uhnAnxBclL0GwKvwn60ZD2AMgtPmDQFcDQBm9hGCuZxvQPAH8yMAQ+F5Dfj1zZYrzzy1CtuzGUEgRpvZyIxFOiIY3Z2z8KB0IoJ7ClYOjLkVwRkNILgVS9OwPgbB6MnqXIvg49bXEZzFuxXATma2BcBvALxEcj3JI6u04RMApyEI7ScILjg/zczWRNkHkvMZ3Oi/0rMI/g+PAjAy/L6nb9mUiFN224fb2oggC9sRDHqopOxmqJLdNggGbGxAcNumsnBdX0VZV0LFJrsW3GbqcgSd3tUIOi9XZCySz+zW1M7zAPRAkMWbEAwSqs6FCM4QvhO2e0i4jXcA/A3A4jC73/go18zeBXABgDsRvNb9APQzsy+j7EP42h2b8XNDAD9EcPu2tItNbhEMzH0FQX/hZQTX+V+dsYhym6FKbluE2/oUwWVs+wDoa2afR1nXDrcTXggsOSA5B8EI9U9K3RaR2lB2JYkYTDIwF8BBKX/zISmi3JaOOrsiIiIiklpxvoxBRERERCQn6uyKiIiISGqpsysiIiIiqZVTZ5dkX5LvklxIcli+GiVSaMquJJWyK0mk3EopZT1AjcFN4t9DMB/yMgS3COpvZm9Xt0yrVq2srKwsq+1J3TV79uw1ZtY6X+tTdqUYlixZgjVr1uR1AoraZle5lWzomCtJVV12c5lB7QgAC81sMQCQnIjgfm/VhresrAyzZs3KYZNSF5GsblrZbCm7UnDl5eWFWG2tsqvcSjZ0zJWkqi67uVzG0B7fnKN5Gb45F3LlhgeRnEVyVkVFRQ6bE8kbZVeSqsbsKrcSQzrmSknl0tn1fTznXBNhZiPNrNzMylu3ztunIiK5UHYlqWrMrnIrMaRjrpRULp3dZQimvavUAcEUoSJxp+xKUim7kkTKrZRULp3d1wHsS3LPcAq8cwFMy0+zRApK2ZWkUnYliZRbKamsB6iZ2VaSVwJ4BkA9AKPNbH7eWiZSIMquJJWyK0mk3Eqp5XI3BpjZkwCezFNbRIpG2ZWkUnYliZRbKSXNoCYiIiIiqaXOroiIiIikljq7IiIiIpJa6uyKiIiISGqpsysiIiIiqaXOroiIiIikVk63HhMRERGR0lqzZo1Tu+yyy5zawIEDndrJJ59ckDbFic7sioiIiEhqqbMrIiIiIqmlzq6IiIiIpJY6uyIiIiKSWhqgJiIiIpIQvsFoxx13nFN79913ndq8efOcmgaoiYiIiIgkmDq7IiIiIpJa6uyKiIiISGqpsysiIiIiqZXTADWSSwBsBLANwFYzK89Ho5Jg8eLFTm3w4MGRlr3oooucWv/+/XNuk0SX5Oxu2rTJWz///POd2rRp05zaTjtFe4/7wAMPOLU99tgj0rI9evRwak2bNo20rOxYkrMrdZdym52HHnrIqd14441ObdGiRZHW17Vr15zblET5uBvD8WbmDg0UiT9lV5JK2ZUkUm6lJHQZg4iIiIikVq6dXQPwLMnZJAflo0EiRaLsSlIpu5JEyq2UTK6XMRxtZitI7gFgOsl3zGxm5hPCUA8CgE6dOuW4OZG8UXYlqXaYXeVWYkrHXCmZnDq7ZrYi/Hc1yUcBHAFgZpXnjAQwEgDKy8stl+3FiW8w2tNPPx1p2ajP06C1wklydpcvX+6tP/nkk07NNxgt6gA130BKn+3btzu1Y445xqldeeWVTq1Pnz5OrXHjxk5tl112idSWuqCm7MY1t9WZOXOmU7v99tud2tlnn531Nq6++mqn1rFjR6c2dOhQp9a3b1+nFnWgpnwtycfcYvEdS+fMmePUFi5c6NRIOrWDDjrIqY0dOzbL1iVb1pcxkGxCslnl9wBOBPBWvhomUijKriSVsitJpNxKqeVyZrcNgEfDdxP1AUwws2inLEVKS9mVpFJ2JYmUWymprDu7ZrYYwMF5bItIUSi7klTKriSRciulpluPiYiIiEhq5WNSidR7+eWXnZpvkNmECROcmm/wxb333uvUxo0b59Q0QE182rRp463feeedWa/zH//4h1N79NFHs16f73fGV/O5+OKLndqgQe6dig4//PBat0uK45NPPnFqJ510kve5b775plPbunWrU/PNBpiL+fPnOzVf9po3b+7UGjZsGGkbM2bMcGoHHHBApGWl7vFl/NZbb816fe3bt3dqvjzXBTqzKyIiIiKppc6uiIiIiKSWOrsiIiIiklrq7IqIiIhIaqmzKyIiIiKppbsxRHDBBRdEep7v7gmnnnqqU/PdjcF3d4cNGzY4tbo6klK+1qJFC2/98ssvz3qd/fr1c2ovvfSSU1u9enXW24hqzJgxTs23z7obQ3y98cYbkWrVadKkiVPz3YVk8eLFtWtYDevzWbVqlVPzHZt9fNNmd+rUyak98cQT3uXbtWsXaTuSPBs3bnRqf/jDH5yamTtrsm9aYd808HfccUeWrUsfndkVERERkdRSZ1dEREREUkudXRERERFJLXV2RURERCS1NECtwHwDyvbcc0+n9sEHH0RaViRXvgFg999/v1MrxmA0n5///OdO7ZJLLilBS6QYzjvvPKd25plnOrWf/exnTq2srMypPfXUU07NN3inadOmkdp39913O7VXX33VqT3//PNObe3atU7NN5Xyueee6932b3/7W6fmG/QmyVNRUeHUfLki6dR8eR4wYIBT8w2GrKt0ZldEREREUkudXRERERFJLXV2RURERCS11NkVERERkdSqcYAaydEATgOw2sy6hbWWACYBKAOwBMAPzWxd4ZpZWvvvv79T8w0o882q8/e//z3Ssj5/+9vfnJpvljbxq2vZXbhwoVPbd999nZpvcEMufvrTnzo13+xUQ4cOzet20yzp2W3WrJlT69Chg/e5rVu3dmqPPPKIU1u2bJlT+8tf/uLUfMfrXAwfPtyp+WawuuWWW5zayJEjnZpvRrYXX3zRu+3LLrvMqflmomvQoIF3+WJLem6LqXHjxk5tjz32cGpRBwr7ZmH1DYL71re+FWl9aRPlr94YAH2r1IYBeN7M9gXwfPizSNyMgbIryTQGyq4kzxgotxJDNXZ2zWwmgKr3TzkDwNjw+7EA3PvEiJSYsitJpexKEim3ElfZfp7ZxsxWAkD4r3vuPURyEMlZJGf5TqmLFJmyK0kVKbvKrcSMjrlScgUfoGZmI82s3MzKfddmicSVsitJpNxKUim7UijZzqC2imRbM1tJsi2A0ky1VCQXXXSRU/NdDL7rrrvmdbvjxo1zahqglrPUZtc3WMc3GC2XAWq+2c1+/etfZ70+qZXEZPfII490alOnTvU+t7y8PNI6GzVq5NQaNmxYu4blie93yDdA7YorrnBq9957r1PzDYIDgAULFjg136A136yIMZKY3BaTb6CYb6bIW2+9NdL6fAPZ7rnnHqfmy2ldkO1fvWkAKuemGwDgsfw0R6TglF1JKmVXkki5lZKrsbNL8m8AXgGwP8llJC8FMAJAH5LvA+gT/iwSK8quJJWyK0mk3Epc1XgZg5lV97l57zy3RSSvlF1JKmVXkki5lbjSDGoiIiIiklrZDlCrU3yDwpYuXerUrr/+eqd2+eWXO7WePXs6tfPOO8+p+QbBiZSSb7BEu3btnJovzy1atChIm6RumDRpklM77bTTStCS6HwzCd50001O7aWXXvIu/9xzzzm1KVOmOLVf/OIXTs03e6LEm28AcNQBaj6+Qe6+AY6+Y3ja6MyuiIiIiKSWOrsiIiIiklrq7IqIiIhIaqmzKyIiIiKppQFqWRo2bFikWlS+AT0itdGtW7eSbPeqq65yao8++qhTmz59ejGaIzHw8ccfO7WzzjrL+9wOHTo4tRtuuMGpnXLKKbk3LKZ69/bfmcs3QG3z5s1O7dVXX3VqGqCWPM2bN3dqvhkqfYPhfbP6ffjhh07t2GOPdWqLFi2K2sTE0pldEREREUktdXZFREREJLXU2RURERGR1FJnV0RERERSSwPUEubll192akcddVQJWiJxM2DAAKfWsWNHp0bSqU2YMMGpjRkzJuu2zJ4926nNmjXLqZWXl2e9DYmvefPmOTVf7gBg5syZTq1z5855b1OcVTd4zzcQSeqWIUOGODXf33zfbIJbtmxxasuWLXNqv/rVr5zapZde6tTat29fbTvjTmd2RURERCS11NkVERERkdRSZ1dEREREUkudXRERERFJrRoHqJEcDeA0AKvNrFtYuxnAjwFUhE+7wcyeLFQjRbKh7AInnHBCpOcdf/zxTm3UqFFO7emnn3ZqJ598slP79NNPnVqPHj2c2j//+U+n1rNnz2rbWVekMbsHHHCAt96mTZsityR+ysrKvPXTTz/dqU2bNq3ArcleGnNbag0bNnRqvmPkD37wA6c2depUp+Y7NvsGqN13331ObcGCBU6tcePGTi2OopzZHQOgr6f+RzPrHn4puBJHY6DsSjKNgbIryTMGyq3EUI2dXTObCWBtEdoiklfKriSVsitJpNxKXOVyze6VJN8kOZrkbtU9ieQgkrNIzqqoqKjuaSLFpOxKUtWYXeVWYkjHXCmpbDu79wDYG0B3ACsB3FbdE81spJmVm1l569ats9ycSN4ou5JUkbKr3ErM6JgrJZfVDGpmtqrye5KjADyRtxbVUZdffrlTu/fee53a+PHjnZpmUItO2c1N377u5Xht27Z1alHPyvgGVTz00ENOTYPWkpXd3XZzT96df/75JWhJMvgG/gDAq6++6tR8r22/fv3y3qZ8SVJuk8w3oMw3M9+ZZ54ZaX2+mda2bdtW+4bFRFZndklm/nU7C8Bb+WmOSGEpu5JUyq4kkXIrcRDl1mN/A9ALQCuSywDcBKAXye4ADMASAJcVsI0iWVF2JamUXUki5VbiqsbOrpn195Td8+UiMaPsSlIpu5JEyq3ElWZQExEREZHUymqAmuSfbwCOb4Car3bPPfcUpE0iUbz++utOrU+fPk7t/fffd2pr17q35Ozdu7dT++qrr7JsnRTaG2+84dRWrlzp1Hyz9AH+GaLqmqFDh3rrq1evdmq+GedatGiR9zZJ8nXp0qXUTYgNndkVERERkdRSZ1dEREREUkudXRERERFJLXV2RURERCS1NEAtJjp37lzqJohkpX379k7tmWeecWp77bVXMZojRfbOO+84tTlz5ji1OM/yVUwffvihU3v77be9z23Xrp1TGzVqVN7bJNl57733Ij1vv/32K3BLgKeeesqpDRkyxKmZWaT1bd++Pec2xYnO7IqIiIhIaqmzKyIiIiKppc6uiIiIiKSWOrsiIiIikloaoBYT3bp1y3rZl19+2akdddRRuTRH6hjfLGiTJ0+OtKxvwMPmzZtzbpMkw3nnnefUli9f7tSqG4SV5lmexo4d69SuueYap+abSRAAbrnlFqd28skn594wqZWNGzd660ceeaRTO/30053aZZddlvW2J02a5NSmTZvm1Hy/c9u2bXNqJJ3ascceG2kbTZo0qbadcaczuyIiIiKSWursioiIiEhqqbMrIiIiIqmlzq6IiIiIpFaNA9RIdgQwDsC3AGwHMNLM/kyyJYBJAMoALAHwQzNbV7implvz5s2zXnb48OFOzXdRey7bSCJl12/9+vVOzTfQIirfTDs77aT30dlKWm7vvPPOSM8rxixSxeLL/E9+8hOnFnW2sxEjRnjr1157be0aVmJJy25U1c06tmHDBqc2fvx4p/bAAw8UvD2+gWe77rqrU/O1r2fPnk6tadOmWbYunqL8RdoK4BozOxDAkQAGk+wCYBiA581sXwDPhz+LxImyK0mk3EpSKbsSSzV2ds1spZm9EX6/EcACAO0BnAGg8r4qYwGcWahGimRD2ZUkUm4lqZRdiatafdZIsgzAIQD+DaCNma0EgoAD2KOaZQaRnEVyVkVFRW6tFcmSsitJpNxKUim7EieRO7skmwKYAmCImbkXqlTDzEaaWbmZlbdu3TqbNorkRNmVJFJuJamUXYmbSDOokdwZQXAfNLNHwvIqkm3NbCXJtgBWF6qRddWee+7p1D744AOn9vTTTzu1u+++26kNG1b3LpNSdl2+gQe+WZ3++Mc/Zr2NXAaotWnTJutl0yJJuR04cKBTW7hwoVP74osvvMvXrx/viTyXLFni1P70pz85taiD0X7/+987tSFDhnifm8SBnknKblS+wV+AfwCYbwBwvpWVlTm14447zqldd911Tu2AAw4oRJNir8bfJAb/y/cBWGBmt2c8NA3AgPD7AQAey3/zRLKn7EoSKbeSVMquxFWUt9RHA7gQwDySc8LaDQBGAJhM8lIAHwL4QWGaKJI1ZVeSSLmVpFJ2JZZq7Oya2YsA/Ofwgd75bY5I/ii7kkTKrSSVsitxlbwLgkREREREIlJnV0RERERSK97DYOu4QYMGObXrr78+0rJLly7Nd3MkJXyj330j6seOHevU1q5dm9e2+KaQnT59el63IYXVqFEjp9ahQwen1qRJk2I0x2vdOndm2k8//dSp+Y65c+fOdWpR7wHrmwbYd+eFuN+Roq5r1qyZt/7KK684tf/85z+R1nnzzTc7tX79+jm1ww47zKmdc845kbYhX9OZXRERERFJLXV2RURERCS11NkVERERkdRSZ1dEREREUktXxcdYz549S90EqSN8A8VOPPFEpzZx4sS8bvcHP3DvLd+uXbu8bkOKb86cOU5tLxcxUgAAIABJREFUwoQJ3ucOHjzYqbVt29apPfaYO+nW7rvv7tSOOeYYp+bLWcOGDZ3ac88959R8g8d69Ojh1O655x6ndtBBBzm1JE4BLH6+46av5qNBZsWl3zoRERERSS11dkVEREQktdTZFREREZHUUmdXRERERFJLA9Ri7Kijjor0vL59+zo132AJkdrwZWjjxo1O7fHHH3dqbdq0cWrjx493akcffXSWrZM4O/74453a8OHDvc/t3bu3U9uyZYtT++KLL3JvWA0OPfRQp3bHHXc4tajHZhGJB53ZFREREZHUUmdXRERERFJLnV0RERERSS11dkVEREQktWocoEayI4BxAL4FYDuAkWb2Z5I3A/gxgIrwqTeY2ZOFaqgEzKzUTUgMZTc3TZs2dWpTp04tQUvqlrTm9uGHH/bWfTOUXXPNNU7tvvvuy2t7unXr5tR8M6i1aNEir9tNs7RmV5Ivyt0YtgK4xszeINkMwGyS08PH/mhmfyhc80RyouxKEim3klTKrsRSjZ1dM1sJYGX4/UaSCwC0L3TDRHKl7EoSKbeSVMquxFWtrtklWQbgEAD/DktXknyT5GiSu1WzzCCSs0jOqqio8D1FpOCUXUki5VaSStmVOInc2SXZFMAUAEPMbAOAewDsDaA7gndyt/mWM7ORZlZuZuWtW7fOQ5NFakfZlSRSbiWplF2Jm0gzqJHcGUFwHzSzRwDAzFZlPD4KwBMFaaFIDpRdSaI05rZly5aRnztq1KhINYmfNGZXkq/GM7skCeA+AAvM7PaMetuMp50F4K38N08ke8quJJFyK0ml7EpcRTmzezSACwHMIzknrN0AoD/J7gAMwBIAlxWkhSLZU3YliZRbSSplV2Ipyt0YXgRAz0O6R57EmrIrSaTcSlIpuxJXmkFNRERERFJLnV0RERERSS11dkVEREQktdTZFREREZHUUmdXRERERFJLnV0RERERSS2aWfE2RlYAWAqgFYA1RdtwYWlfCq+zmZV07khlN9biuh9xyi0Q39epttKyH0B890XZLYy07AcQ333xZreond3/bpScZWblRd9wAWhf6pY0vUZp2Ze07EehpeV1Sst+AOnal0JKy+uUlv0AkrcvuoxBRERERFJLnV0RERERSa1SdXZHlmi7haB9qVvS9BqlZV/Ssh+FlpbXKS37AaRrXwopLa9TWvYDSNi+lOSaXRERERGRYtBlDCIiIiKSWursioiIiEhqFb2zS7IvyXdJLiQ5rNjbzwXJ0SRXk3wro9aS5HSS74f/7lbKNkZBsiPJf5JcQHI+yf8J64nbl2JSdktP2a095TYelN3aU3ZLLy25LWpnl2Q9AHcBOBlAFwD9SXYpZhtyNAZA3yq1YQCeN7N9ATwf/hx3WwFcY2YHAjgSwODw/yGJ+1IUym5sKLu1oNzGirJbC8pubKQit8U+s3sEgIVmttjMvgQwEcAZRW5D1sxsJoC1VcpnABgbfj8WwJlFbVQWzGylmb0Rfr8RwAIA7ZHAfSkiZTcGlN1aU25jQtmtNWU3BtKS22J3dtsD+Cjj52VhLcnamNlKIAgFgD1K3J5aIVkG4BAA/0bC96XAlN2YUXYjUW5jSNmNRNmNmSTnttidXXpquvdZiZBsCmAKgCFmtqHU7Yk5ZTdGlN3IlNuYUXYjU3ZjJOm5LXZndxmAjhk/dwCwoshtyLdVJNsCQPjv6hK3JxKSOyMI7oNm9khYTuS+FImyGxPKbq0otzGi7NaKshsTachtsTu7rwPYl+SeJHcBcC6AaflaOckZJAcWedlpAAaE3w8A8Fg2288WyYtJvpjx8yaSe9WwDAHcB2CBmd0e1s4H0AAl3JeYU3bzjOTNJB8Iv+8UZrdeDcv4snsDgC+g7PoUNLdASbJb0twCOu4WiY65eVaXc1vUzq6ZbQVwJYBnEFzkPNnM5ld9HsklJL9bzLbtCMnjSc4j+SWAxQC6kFxB8lIAIwD0Ifk+gD7hz5nLlpG0MFSbwn0r2KhFM2tqZotreNr3AFwI4ASSc0jOAbAOwGHYwb5kg+SxGfte+WUkv5fruospwdntRXI7ya8QZLcrybURs1u57CaSGxncAuiSQrTTzD4Ms7uthqdeCTe7cwAcv6N9yQbJLiRnkVwXfj3HZI0Gj5xbIJbZJcm5ABYhyO0WkoNRQ27DZev0cRcAwv3fnPEa/DUf6y2WBB9zSfIXJDfj6/7CcvUXoilUbuvnYyW1YWZPAniy2NvN0dsATjKzFSQbABgO4AAzuy98vHeEdbQws60kvwPgeZJzzOzpzCeQrB/+ghfarPDfQz3bi7IvkZnZCwCaVv5MsheAxwE8Xd0ycZXQ7ALACjPrUM1jNf1/rzCzDuG7+zMAPEzy32b2duaTipjdeQCWm9lBnsfyml0EH5l+H8BSBCcGBiMYEe7bdmwlOLcXITh7VIbgj+uDAI4ws7sQ/f+6Th53MxxsZgsLtO6CS2h2L0LQOTwAX+d2vfoLtZL33CZqBjWSu5F8gmRFeKblCZJV/4jvTfI1kp+SfIxky4zljyT5Msn1JOeGHa8amdkqM8u8VmgbgH2y2QczewXAfADdwjYZycHhu6P3w9oBDG7SvDY8m/bDjH3YneQ0khtIvgZg78z1h+vbJ/y+EcnbSC4NX48XSTYCMDN8+vrwndN36H68cRTJ18PlXid5VMZjM0gOJ/lSeMbvWZKtIr4EAwA8bGaba/vaJVmpspsvFpiK4ODdJeMMxKUkPwTwj5rayeDjyH+FmZkOoFXGY5Xrqx/+3JLk/Qw+QVlHcirJJgCeAtAu411/O2ZcDhEuezqDm5+vD7N6YMZjS0heS/LN8HWeRLJhNfu83syWmJkhGCyT9e99kpUwu/0A3GdmH5nZJgC3AjiHZOPa7oOOu3WPcqvcfoOZxe4LwBIA3/XUd0dwSr0xgGYAHgIwNePxGQCWIwhGEwQXVD8QPtYewCcATkHQye8T/tw6Y9mB4fedAKwH0Clj3ZW17QC+AnBxxH0pQzCCtD6CP5hHA9gCoHf4uAGYDqAlgEZhuz8CcEm4zKEA1gDo+v/bu/M4vcb7/+PvjywSEksktliC4ttQWyNqqcZaVWppKbGX6oKK2mmLony/tlJVgkikCH6oWFoUaayVRFMSobZYKiKhJNaIfH5/nJO6zXVN5sw993LOmdfz8ZjHzLznPudc557PnLnuc5/rXOnjx0i6KX3c+un+PlyxPZf0pfTr36f71V9SF0lb6PMzJS6pa8VyBy9cT9qW/yh5ddpV0r7p98tVPFcvSlonbfM4SedWrOspSUMjz8USkuZKGtLsGusstStpiKR5kmZKelnSRZKWzLgvQyS9nn69mKQ90tpft6KGrk3b2zNDOx+TdGFag1untbBwH79Qk5LuknSjpGUldZP0jZZtqmjn6RXrWUfSB+m2u0k6QdILkrpX/H6ekLRyWufTJP24Yl3vStqqxfrfVXJj9QWSftHsGutEtXuLpBMqtrNlWiMbZtiX/9aTOulxN93WG5LelHSrpAHNrjHqlrptVt02vVDbU7yRx20k6T8tirfySRyo5B99F0knShrdYvl7JB3Usnjb2GafdF1fy7gvCwvl3bQApkn6WYtf7LYV339f0kMt1nGFpNPS/fhUySUUC3/2m1jxKvkD/UiRP7AMxXuApCdaLPOY0g5++lz9ouJnP5X0lwzPxQFKOlzW7BrrLLUracV0XYtJWkPJq/QrMu7LECUdvHeV3Bx9sqR9WtTQmhWPb7WdSv4hzFdFR1vS9Yp0diWtlG532VbatKjO7i+VXNu38GeLKTnAD6n4/exf8fP/k3R5hudiybTOv93sGutEtXuYpH+ltbG0ksE9LmnzDG1cWE+d9rir5AVld0nLSLpU0pTKbZflg7qlbrN8NPya3Y5I3wa4SMkUfAvnYe5tZl3884EtlTehfkXJ2Z2+klaXtJeZ7Vrx826SHmxPG9z9HTMbJemfZtbfs18z03cRj61s8+qSNjOzdyuyrpJGS+qXft1yH6Pbk9RDySuq9lo5st5X9MUber9Z8fWHqrgudxEOknStpxXdmTSrdt39TX3+u3rZzE5Qctb0RxmbvqjrfVu2eVHtXFnJP5rKy1de0RdvLbTQqpLecff/ZGxjpS/UrrsvMLPXtOjaXbmtlbr7B2Z2uaRZZvZld8/1bXZqqYnH3RFKamGckuPeBUreIn69Hc3vtMddT2bwkqR5Zna0pDmSvqzkuvfSo26p20qFumZX0rFK3kLdzN2XUvIKQPrizacr/3mupuSVzWwlv/DR7r5MxceS7l7NCMKuSmYLWaqKZWMqO3+vSfpbi3b2cvefSJql5OxYy32MmS3pY7W4RieyvZg3lPwRVVpNyRmyqpjZqkrOyl1b7ToKLi+16y222VEta7e1ds6QtKwl190u1Frtviapj5kt08b2Yr5Qu2ZmSp7Xqmu3wmJK3hIt+ixO7dWU2nX3Be5+mrsPSF9wTVXye6zF71LqBMfdyPZr+befd9Qtdftfee7sdjOzHhUfXZVcd/ORkgul+yg5Vd/S/pbcMmgJSb9WMhjqM0l/lLSrmX3TzLqk6xwSuWA9YGZ7mtm6ZraYmfVTct3hP9z9nfTnp5vZuNrstu6UtI6ZHWBm3dKPTdOzSZ8puYbldDNbwpLbIB0UW4m7L1DyCvNCSwbxdEkvLF9cyR/BAkmt3V/v7rQNQ82sq5l9X8lbPHd2YL8OkPSou1fzyrFo8lS7Qyy5h62lLzjOVcX9EM1spJmN7PguS4tqp7u/omRU7xlm1t3MtlJytiPgydSTf5Z0mSWDTLqZ2cJ/VDMlLWdmS7fShpskfdvMtrPkRujHKrkH76Pt3Rkz28HMNk73ZSklf/cL31osqzzVbh8zWyut3YFKnv9fp8c2jruLYGbrmdlG6fZ7KTm7+G+Vt3apW+p2kfLc2b1bSaEu/Dhd0m+VXOA8W9Ljit++arSkkUpOm/eQ9DNJcvfXlNw66RQlv7zXJB2vyHNgn9/gfuGroP7ptuYqOZW+QMlgnYVWlfRItTtayd3nStpRyQ20F16k/b9KLhSXkvsO9krzkZKuWcTqjkvbO0HJNZf/K2kxd/9Q0tmSHrFkpOnXWrThbUm7KOkovK1kkM8u7j47yz5YMhJ+vxbxgZJGZVm+BPJUu5souX7qAyUdvikL15uqZe221c6hkjZTUounadFn+Q9QcpblWSUz8wxLt/GspBskvZTW7hcuQXD35yTtL+l3Sp7rXSXt6u7zsuxD+tx9Pf12mXRb7yl5e+9LknZy94+zrKug8lS7fdP2fKDkxc8Idx9esQjH3QotjrsrKBngOUfJvV4HpOv6NMu6Coi6pW4XvZ1OePlkzVlyk+Xt0l86UAiWzEr0T0kblPifIEqK4y6KiLptDjq7AAAAKK08X8YAAAAAdAidXQAAAJRWhzq7ZraTJdPTvWBmJ9WqUUC9UbsoKmoXRUTdopmqvmbXzLoomSVkByU3S54gaV93f6a1Zfr27esDBgyoanvovCZNmjTb3fvVan3ULhph+vTpmj17dk3va9re2qVuUQ2OuSiq1mq3IzOoDZb0gru/JElmNkbJrTpaLd4BAwZo4sSJHdgkOiMza23Wl2pRu6i7QYMG1WO17apd6hbV4JiLomqtdjtyGUN/fXEautcVmVnIzA43s4lmNnHWrFkd2BxQM9QuiqrN2qVukUMcc9FUHensxt6eC66JcPfh7j7I3Qf161ezd0WAjqB2UVRt1i51ixzimIum6khn93V9cc7lVZTM4AHkHbWLoqJ2UUTULZqqI53dCZLWNrM10pmY9pE0tjbNAuqK2kVRUbsoIuoWTVX1ADV3n29mR0q6R1IXJfM/T61Zy4A6oXZRVNQuioi6RbN15G4Mcve7Jd1do7YADUPtoqioXRQRdYtmYgY1AAAAlBadXQAAAJQWnV0AAACUFp1dAAAAlBadXQAAAJQWnV0AAACUFp1dAAAAlBadXQAAAJQWnV0AAACUFp1dAAAAlBadXQAAAJQWnV0AAACUVtdmNwAA2jJmzJggmzdvXubl33jjjSB76623guzCCy9sX8MAoA0ffvhhkP3mN78JshEjRgTZ4MGDM21jzpw5QfbAAw9kWrYz4MwuAAAASovOLgAAAEqLzi4AAABKq0PX7JrZdElzJX0mab67D6pFo4B6o3ZRVNQuioi6RTPVYoDaNu4+uwbrARqN2m2nRx55JMhigy9i1l9//SA75ZRTguzuu+8Osthgso7ab7/9ar7OBqJ2UUSlr9vYsWqrrbYKshdffDHIzCzI7rjjjkyPW2KJJYLs6quvDrJDDz00yDoDLmMAAABAaXW0s+uS7jWzSWZ2eC0aBDQItYuionZRRNQtmqajlzFs6e5vmNnyku4zs2fdfXzlA9KiPlySVltttQ5uDqgZahdFtcjapW6RUxxz0TQdOrPr7m+kn9+SdJuk4O7H7j7c3Qe5+6B+/fp1ZHNAzVC7KKq2ape6RR5xzEUzVX1m18yWlLSYu89Nv95R0q9r1jKgTqjdbDbbbLMge/LJJ4Ns/vz5mda34oorBtmbb76ZadmhQ4cGWdeu2Q9fRxxxRJBtsMEGmZfPC2q3bbGZpK699tog++Mf/xhkEyZMqHq7Dz30UJBtscUWVa+vTMpat7HBuaeeemqQxQajrbHGGkG26667ZnrcMcccE2SDBoU3t+isg9FiOnIZwwqSbktHBXaVdL27/6UmrQLqi9pFUVG7KCLqFk1VdWfX3V+StGEN2wI0BLWLoqJ2UUTULZqNW48BAACgtOjsAgAAoLRqMYNarsQuGD/zzDOD7MEHH4wuv8kmmwTZscceG2TuHmQrrLBCkPXu3Tu6HSDvYqOht9lmmyBbb731guxLX/pSkB155JGZthsbfHH++ecH2WKL8Vq9M4kNPBs+fHiQXXTRRUE2c+bMIIsdw2MzU8Vq/ve//32QDRgwIMhQbhdffHGQjRgxIshidfXUU08FWWwWtEsuuSTT+o477rggi80SeeWVVwbZuHHjgiw24G355ZcPsqLgvwUAAABKi84uAAAASovOLgAAAEqLzi4AAABKi84uAAAASiuXd2OYPHlykP3whz8Msueffz7IPv300yD76KOPgmzllVeObvuJJ54IsquuuirIFixYEGSLL754kHXr1i3IYlP4HXLIIdH2tBQb5d6jR49MywLtceedd2Z6XOxvbrfddqt6u4MHDw4y7rzQucT+Bxx22GGZHpdVz549g+zjjz8Osp133jnI1llnnaq3i+J57733ovnvfve7IIvd5eO3v/1tkMXuvBAzcuTIIPvBD34QZL/4xS+C7J///GeQ3XDDDUEWu7vDHXfcEWSxv4Wi4D8IAAAASovOLgAAAEqLzi4AAABKi84uAAAASiuXA9ROPvnkIJs0aVKQbbrppkG25pprBllsCuDWLrSODbZZaqmlgmz+/PlB9vjjjwdZ7CLvV199Nci+/vWvB1nsovjYdH3nnXdekO20005BFpv+FWiPV155JchiAyP+/Oc/B1lsEESsnvfZZ58qW4ciuu+++4Ls5z//eZBNmzatptt99tlng+y6664Lsl/+8pdBNnDgwCCLTSvcvXv3KluHPIlNVS1Jb731VpDFjnMdERtkFptqODYwLjaoMzbgPtbm2N/lkCFDgizrQLtm48wuAAAASovOLgAAAEqLzi4AAABKq83OrpmNMLO3zGxKRdbHzO4zs+fTz8vWt5lA+1G7KCpqF0VE3SKvsgxQGynpUknXVmQnSbrf3c81s5PS70+sVaNiM4YMHTo0yDbaaKMgu+CCC2rVjHaLzapz4IEHZlo2NnPP66+/HmRjxowJsmHDhgVZbKBebPBFyWcCGqkG126ZvPTSS0F21llnBdkf//jHTOtbbbXVgmz69OntblcnMVIlq93YgBdJ2n333YMsdjxcb731guzll18OstiMmZtttlmQrbLKKkEW+z8Tq/lvf/vbQTZ+/Pgg22KLLYKs5EaqZHUrSVdccUU0jw0Ki2WjRo0KspkzZwbZiBEjMq1vhRVWCLLYbIKxAcCxv6NjjjkmyC6++OIg69WrV5CdeeaZQZZHbZ7Zdffxkt5pEe8maeFvb5Sk8GgFNBm1i6KidlFE1C3yqtprdldw9xmSlH4OXz6kzOxwM5toZhNnzZpV5eaAmqF2UVSZape6Rc5wzEXT1X2AmrsPd/dB7j6I+7yiSKhdFBF1i6KidlEv1XZ2Z5rZSpKUfg7vrAzkE7WLoqJ2UUTULZqu2hnUxko6SNK56efba9YixS++vuuuu4KsS5cutdxsU/Xo0SPI1lhjjSD78Y9/HGSxQRrXXHNNkJ1xxhlBtuuuuwbZbrvtFm1jz549o3nB1LV2y2T99dcPstjgn6xigzTQLoWp3dhb0LFZHVsTO87FZoOKzTwZm9EpNpNlzKqrrhpkEyZMCLLYIJ9YW2KzasZm/iy5wtRta77//e9H83PPPTfT8rHBY//4xz+CLDaTWez/8SWXXBJkscFoMT/5yU+CbPTo0UH25JNPBtmUKVOCrCiy3HrsBkmPSVrXzF43s0OVFO0OZva8pB3S74FcoXZRVNQuioi6RV61eWbX3fdt5Ufb1bgtQE1RuygqahdFRN0ir5hBDQAAAKVFZxcAAAClVe0AtYaLDeAqk9gF7LGZhd5+++0g++CDDzJt49577w2yG2+8Mciefvrp6PJf/vKXM20H+fb8888H2fbbbx9kHRmMFvPd7343yJZccslMjzvvvPOCrEwDVMvmN7/5TZDFBt+0Jjbw5+abb860bGymtj59+mTedksDBgwIstjg4dhsgNttF757P23atCDr379/VW1DY6y11lrRfOuttw6ycePGZVrnN77xjSDbcMMNg+zkk08OsqyD0WK6desWZL179w6y2MxtY8eOrXq7zcaZXQAAAJQWnV0AAACUFp1dAAAAlBadXQAAAJRWYQaold2MGTOC7NVXX8207GabbRZkL730UpBdeeWVQbbMMssEGQPRiunjjz8OshNPPDHIrr/++iCbPXt2XdpUKTa4MpZddNFFQfbII48EWWwg0lJLLVVl61CtN998M8hiMzi25rrrrguy3/72t0F2zjnnBFls4FnseNgRscHRjz76aJBtvPHGQRZ7bmIDmPbbb7/qGoeGiM3KJ0l33nlnkMWOaTGx2fqaJVZ/f/vb34KsPYNM84YzuwAAACgtOrsAAAAoLTq7AAAAKC06uwAAACgtBqjlxI477hhkP/vZz4Js+PDhQXb55ZcHWWzGl169elXZOhTBL3/5yyC75JJLql7fiiuuGGStzSSUxd577x1ksZm2Zs6cGWRPPPFEkH322WdVtwW18/LLLwfZ+++/n3n5s88+O8imTp0aZLHBMccee2zm7dRSbAar2PH6lFNOCbI//elPQbbPPvtEt8MsgfkWG7jW2mC2PNt1112DLDaDWsywYcOieWyQaTNxZhcAAAClRWcXAAAApUVnFwAAAKXVZmfXzEaY2VtmNqUiO93M/m1mk9OPnevbTKD9qF0UFbWLIqJukVdZBqiNlHSppGtb5Be5+/k1b1EnFRuIEJsxaOmllw6y7373u0E2ZMiQILvsssuCrHv37hlbWEgjVcLajQ3gkqSbb7450/KxWacOO+ywIDvyyCODrCOz/ixYsCDIXnvttSA7//zC/mpqaaQKUrurrLJKkPXs2TPIPvroo+jyzzzzTKbtnHvuuUF29NFHZ1q2EbIOULvtttuCLDZoTYof23NupApSt/hcbMBlbEBoLHvqqafq0qZaa/PMrruPl/ROA9oC1BS1i6KidlFE1C3yqiPX7B5pZk+lb1ss29qDzOxwM5toZhNnzZrVgc0BNUPtoqjarF3qFjnEMRdNVW1n9w+S1pK0kaQZki5o7YHuPtzdB7n7oH79+lW5OaBmqF0UVabapW6RMxxz0XRVdXbdfaa7f+buCyRdKWlwbZsF1Ae1i6KidlFE1C3yoKoZ1MxsJXefkX67h6Qpi3o8qhMb5HHaaacF2Z577hlkW265ZZC99957QXbRRRcFWWzASVmUoXZXWGGFaB4b0BgbPBAbSLPSSit1vGFtmDNnTpAxGC27vNZubNDiO++El23+6Ec/ii6/+eabB9kuu+wSZI2o0Y6IHa9jsxqeeeaZQRb7m5QKOUAtkNe6xaJlnUHt1VdfjeYffvhhkDVzdrk2O7tmdoOkIZL6mtnrkk6TNMTMNpLkkqZLih/FgCaidlFU1C6KiLpFXrXZ2XX3fSPx1XVoC1BT1C6KitpFEVG3yCtmUAMAAEBp0dkFAABAaVU1QA3Ns9hi4euTDTfcMMhefPHFIIvNqnb66acH2SWXXBLddjMvLkfb9t03fAcxljXC3/72tyA78cQTq17f1ltvHWSxAUHIh27dugXZiBEjmtCS/Mk6MxXQTFnrdLXVVosun7f+Amd2AQAAUFp0dgEAAFBadHYBAABQWnR2AQAAUFoMUCuBefPmBdkzzzwTZG+99VaQXX11eAvE+++/P7qdl19+uYrWoewee+yxIBs2bFiQTZ48OdP6YoPR7rnnniDr0aNHpvUBANrn5JNPDrLYLJ3vvvtudPlPP/00yGIDVxuFM7sAAAAoLTq7AAAAKC06uwAAACgtOrsAAAAoLQao5djcuXODbOzYsUH2m9/8JsimTZuWaRuxWaiGDh2aaVmU29tvvx1ko0ePDrLYzGixQZOxGXUOOuigILvwwguDjMFoAPImNug79j/6+uuvD7IHHnigLm2qlSlTpgRZbAa12KyueVSMVgIAAABVoLMLAACA0qKzCwAAgNJqs7NrZqua2YNmNs3MpprZ0Wnex8zuM7Pn08/L1r+5QHbULoqIukVRUbvIqywD1OZLOtbdnzSz3pImmdl9kg6WdL+7n2tmJ0k6SVI4UgWB2ExksQvYzz///CB77733qt7ussuGx5cxY8bjm8c0AAAgAElEQVQE2Q477FD1NnKmULU7bty4IDv77LOD7Oabb44uv8wyy2TazowZM4LswAMPDLKXXnopUxaz2WabBdlZZ50VZNtvv32m9XUyhapboEKnqt1jjjkmyGL/y2ODuP72t78F2Zprrhlkq666aqa2xPoGc+bMCbLLL788yK655poge/PNN4MsNkBt3XXXjbanmbOlxbR5ZtfdZ7j7k+nXcyVNk9Rf0m6SRqUPGyVp93o1EqgGtYsiom5RVNQu8qpd1+ya2QBJG0v6u6QV3H2GlBS4pOVr3TigVqhdFBF1i6KidpEnmTu7ZtZL0i2Shrl7eG689eUON7OJZjZx1qxZ1bQR6BBqF0VE3aKoqF3kTabOrpl1U1K417n7rWk808xWSn++kqTw7sqS3H24uw9y90H9+vWrRZuBzKhdFBF1i6KidpFHbQ5Qs+SK5KslTXP3yqmNxko6SNK56efb69LCAhk/fnyQ3XXXXUF2xRVXBFnsQvKYr371q0G23XbbBdnaa68dZPvuu2+QxWa1Koui1e6rr74aZH/961+D7JBDDoku36tXr0zbefTRR4Ms68Cz2CC4PfbYI8guvfTSICtzrdVS0eoWiQkTJgTZGWecEWSxwUoXX3xxXdrUaJ2tdvfbb78giw36jg3s2nbbbTM9buutt87UlmeffTbIYjO8uXum7cayTTbZJMiuvPLKTO1rtix3Y9hS0gGSnjazyWl2ipKivcnMDpX0qqS96tNEoGrULoqIukVRUbvIpTY7u+7+sKSwi58ITykCOUHtooioWxQVtYu8YgY1AAAAlBadXQAAAJQWnV0AAACUVpYBasgoNkVqbErA2HS8AwcODLI999wzyFZcccUgW2GFFbI2ETm25JJLBlnv3r2D7E9/+lOHthMbZRuroVGjRgVZbDrL2J0/gDKL3b1kn332CbLYnRdix/Xdd2dCsSLaeeedgyx2t6Wf//znQfb+++9n2kZsGvnYMTzrXRZij4sd/2+77bYgi/VxioIzuwAAACgtOrsAAAAoLTq7AAAAKC06uwAAACgtBqjV0OKLLx5kv/zlL5vQEhTRd7/73SDbYostgiw2ZaMkvfnmm0HWt2/fILvsssuCbK+9mNAInUvs7yA2HfsjjzwSZLHj+ieffBJka621VpAde+yxQdalS5dW24liOeyww4Jst912C7IHHnggyGID2WLH9ayDjA899NAgW3755YPswAMPDLKll146yIqMM7sAAAAoLTq7AAAAKC06uwAAACgtOrsAAAAoLQaoATm20korBdmMGTOa0BKguGJ/M0cddVSQxQb+ZLX99tsH2Zlnnhlkm266adXbQDH169cvyL7//e9nylAbnNkFAABAadHZBQAAQGnR2QUAAEBptdnZNbNVzexBM5tmZlPN7Og0P93M/m1mk9OPnevfXCA7ahdFRN2iqKhd5FWWAWrzJR3r7k+aWW9Jk8zsvvRnF7n7+fVrHtAh1C6KiLqtsdhAzz/96U9BduONNwbZKqusEmTrrLNOkMVmoeratdONAad2kUtt/iW6+wxJM9Kv55rZNEn9690woKOoXRQRdYuionaRV+26ZtfMBkjaWNLf0+hIM3vKzEaY2bKtLHO4mU00s4mzZs3qUGOBalG7KCLqFkVF7SJPMnd2zayXpFskDXP3OZL+IGktSRspeSV3QWw5dx/u7oPcfVDsXnNAvVG7KCLqFkVF7SJvMnV2zaybksK9zt1vlSR3n+nun7n7AklXShpcv2YC1aF2UUTULYqK2kUetXnNriVTylwtaZq7X1iRr5RenyNJe0iaUp8mAtWhdlFE1G1j7LrrrpkyZEftIq+yDBXdUtIBkp42s8lpdoqkfc1sI0kuabqkH9WlhUD1qF0UEXWLoqJ2kUtZ7sbwsKTYhOF31745QO1Quygi6hZFRe0ir5hBDQAAAKVFZxcAAAClRWcXAAAApUVnFwAAAKVFZxcAAAClRWcXAAAApUVnFwAAAKVl7t64jZnNkvSKpL6SZjdsw/XFvtTf6u7e1InSqd1cy+t+5Klupfw+T+1Vlv2Q8rsv1G59lGU/pPzuS7R2G9rZ/e9GzSa6+6CGb7gO2JfOpUzPUVn2pSz7UW9leZ7Ksh9SufalnsryPJVlP6Ti7QuXMQAAAKC06OwCAACgtJrV2R3epO3WA/vSuZTpOSrLvpRlP+qtLM9TWfZDKte+1FNZnqey7IdUsH1pyjW7AAAAQCNwGQMAAABKq+GdXTPbycyeM7MXzOykRm+/I8xshJm9ZWZTKrI+ZnafmT2ffl62mW3MwsxWNbMHzWyamU01s6PTvHD70kjUbvNRu+1H3eYDtdt+1G7zlaVuG9rZNbMukn4v6VuSBkra18wGNrINHTRS0k4tspMk3e/ua0u6P/0+7+ZLOtbdvyzpa5KOSH8PRdyXhqB2c4PabQfqNleo3XagdnOjFHXb6DO7gyW94O4vufs8SWMk7dbgNlTN3cdLeqdFvJukUenXoyTt3tBGVcHdZ7j7k+nXcyVNk9RfBdyXBqJ2c4DabTfqNieo3XajdnOgLHXb6M5uf0mvVXz/epoV2QruPkNKikLS8k1uT7uY2QBJG0v6uwq+L3VG7eYMtZsJdZtD1G4m1G7OFLluG93ZtUjG7SCaxMx6SbpF0jB3n9Ps9uQctZsj1G5m1G3OULuZUbs5UvS6bXRn93VJq1Z8v4qkNxrchlqbaWYrSVL6+a0mtycTM+umpHCvc/db07iQ+9Ig1G5OULvtQt3mCLXbLtRuTpShbhvd2Z0gaW0zW8PMukvaR9LYWq3czMaZ2WENXnaspIPSrw+SdHs126+WmR1sZg9XfP++ma3ZxjIm6WpJ09z9wjTbT9LiauK+5By1W2NmNsTMXq/4fqqZDWljmVjtfl1SP1G7MXWtW6kptdvUupU47jYIx9wa68x129DOrrvPl3SkpHuUXOR8k7tPbfk4M5tuZts3sm2LYmbbmNnTZjZP0kuSBprZG2Z2qKRzJe1gZs9L2iH9vnLZAWbmaVG9n+5b3UYtunsvd3+pjYd9V9IBkrY1s8lmNlnSfyR9VYvYl2ql+/9BxXNwVS3W20hFrV1JMrOjzOx9fV67b2Wp3XTZyt/dv83swnSUdM25+3ruPq6Nh22ppHZ3qqjd3pLWVn1qd4Alt9350Myezdvvti1Z61bKbe3eqqTd65nZAjN7XNnqtlMfd81sHTO73cxmmdk7ZnaPma3b0fU2UsGPuf3MbLqSY+566fGD/kIGZjbcktvNLTCzg2uxTknqWqsVZeXud0u6u9Hb7aBnJH3T3d8ws8UlnSnpf9z96vTn22VYxzLuPt/MNpd0v5lNdve/VD7AzLqmf+D1NjH9vElke1n2pRobuvsLdVp3QxSxds1sMyUHoa0lPSnpx5J+LWmku3+mbL/vDd39BTP7H0njJP1L0uUtttOQ2nX3h5MTDdolUk/1qN0bJD0maef04/+Z2druPqsO26qLItZthX6S/iDpDEkfSlrf3d9W9t91Zz3uLqPkLOIhkuZK+pWSM2//U+Pt1FWBa/fW9KOybv+R/oz+wqL9U9KNkv63list1AxqZrasmd2Zvlr9T/r1Ki0etpaZPWFm76WvbPtULP81M3vUzN41s39aG2+ZLuTuM9298lqhzyR9qZp9cPfHJE2VtH7aJjezI9JXR8+n2f9YcpPmd9JXOHtX7MNyZjbWzOaY2ROS1qpcf7q+L6Vf9zSzC8zslfT5eNjMekoanz783fTV4+YWvr2xhZlNSJebYGZbVPxsnJmdaWaPmNlcM7vXzPpW83x0Fs2qXUkDJE1190mezA1+raS+qmLkrLs/K+khfV67083sRDN7StIHZtbVzFY2s1vS/XzZzH5WsQ89zWxkuv/PSNq0cv2VZ2jMrIuZnWJmL6Y1NsmSm5svrN1/prX7fQsvh/hyWqPvWnJpxHcqfjbSzH5vZnel6/27mX3hb6jisetI2kTSae7+kbvfIulpJWc6Oo1m1a6Z7ajkms3j3f09d/+0osPQLp3tuOvuT7j71e7+jrt/KukiSeua2XLVPH9FRN0Wr27Tff69u98v6eNqnrNWuXvuPiRNl7R9JF9OyT+aJZS8dXmzpD9V/HycpH8rKYwllVxQ/cf0Z/0lva3k7MxiSk67vy2pX8Wyh6VfrybpXUmrVax7YbZA0qeSDs64LwOUjCDtqmR06ZZKXultl/7cJd0nqY+knmm7X1Pyiryrkn+2syWtlz5+jKSb0setn+7vwxXbc0lfSr/+fbpf/SV1kbSFkuts/tumiuUOXrietC3/UfLWRVdJ+6bfL1fxXL0oaZ20zeMknVuxrqckDW3Rpjckvank1e6AZtdYZ6ldSUtJmiRps7QGjpL0D0mWcX8q62lg+js8tGJfJys5sPdM2zZJyVmk7pLWVPI23jfTx5+rpLPcJ11miqTXY8+dpOOVdCzXVfJ3s2FF/f23Ten3QxauR1I3SS9IOiVtw7ZKzmytm/58pJJ7Xw5Oa/s6SWMq1nWnpJPSr/dQcp1a5fNxqaTfNbvOOknt/krJW9h/TJeZIOkbGfdlgDr5cbfF87G7pBnNrjHqlrrNWreSHlbGflam57bZhdqe4o08biNJ/2lRvJVP4kBJ89Jf3ImSRrdY/h5JB7Us3ja22Sdd19faWbzvpgUwTdLPWhTbthXff1/SQy3WcYWk09L9+FTJJRQLf/abWPEq+QP9SMlb0K3+QbVSvAdIeqLFMo8tLLz0ufpFxc9+Kukvi3gOtlbS8VhGSWdhSuW2y/SRt9pVcsA8Ja2b+UoOhJu2Y39c0py0dl+UdJakxSr29QcVj91M0qstlj9Z0jXp1y9J2qniZ4er9c7uc5J2W0SbWuvsfl1Jh3yxip/fIOn09OuRkq6q+NnOkp5tZTsHSHq8RXa2kktAml5rnaB2h6e/60OVvIjZR8lxtG+GNg5QJz/uVjxuFSWdnH2bXWPULXXbjrqtaWe34dfsdoSZLaHk7ZidJC2ch7m3mXXx5PpD6Ys3oX5FSbH1lbS6pL3MbNeKn3eT9GB72uDu75jZKCVvo/b37NfM9F3EYyvbvLqkzczs3Yqsq6TRSq5f66pwH6Pbk9RDSQelvVaOrPcVffGG3m9WfP2hpF6trcyTmWQkaZ4l82rPkfRlJWfuOoUm1u5hkn4gaT0lZzx3lHSnmW3sX7w0Z1E28davt25Zuyu3qN0uSs7mSkldZaldKTnzW23tvubuC1psp5rafV/JmfFKSyk5U9xpNLF2P5I03T8fGzHGzE5VcrYr68jvTnvclZKBUpLulXSZu99QRZsKi7otbt3WQ6Gu2ZV0rJK3NTdz96WUnDGUvnjz6cr78q2m5JXNbCW/8NHuvkzFx5LuXs0Iwq5Krnls+Y+wWl7x9WuS/tainb3c/SeSZik5O9dyH2NmK7nmJXY9okeySm8o+SOqtJqSswO14IrfMLzMmlW7G0q6w93/5e4LPBnkMEPJW1S10LJ2X27Rzt7uvnP68xnKVrsL1xW9lrYNb0ha1cwqj23V1u5USWuaWe+KbMM070yaVbtPqb6TCJT6uGtmyyrp6I5197OrWUfBUbcFrNt6yXNnt5uZ9aj46KrkupuPlFwo3UfJqfqW9jezgemrul9L+n/pq7g/StrVzL5pyeCXHpYMbGl5wXrAzPY0s3XNbLH0lfKFkv7h7u+kPz/dzMbVZrd1p6R1zOwAM+uWfmxqZl9O9+NWSaeb2RJmNlCf3+fuC9IzWyMkXWjJoKEu6YXliyv5I1ig5JrKmLvTNgy1ZNDR95W8xXNne3fGzNYzs43S7feSdIGSP4Jp7V1XgeSmdpVcL/ZtM1vTEjsouXZqivTf+y5O7/guS5KekDTHkkFrPdO2rm9mCwei3STpZEsGjqyi5Prh1lwl6UwzWztt9wb2+eCamWq9dv8u6QNJJ6R/O0Mk7ark2rV2cfd/Kbkm+bT0Od9D0gZKru0rqzzV7m2SljWzg9Jlv6fkbNEjEsfdRTGzpZS87f6Iu9ft1lU5Qt2WoG4lycy6m1kPJS9KFv5eO9xXzXNn924lhbrw43RJv1VygfNsSY9L+ktkudFKrst7U8lp+Z9Jkru/Jmk3JdcvzlLyiuh4RZ4DM1vNklGHC18F9U+3NVfJW+8LlAxeWWhVpYXcUe4+V8lbzfvo80Fd/6vkQnEpue9grzQfKemaRazuuLS9E5QMyvlfJdcyfqjk2sNHLBlp+rUWbXhb0i5KXhm/LekEJbd6mp1lHywZAb9f+u0KSm4jMkfJNZsD0nV9mmVdBZWn2r1WSUdvnJLfwSWSfuTJnRWk2tbuZ0o6lhtJelnJvl4laen0IWcoeXvrZSVnnEYvYnUXKukc35u2+2olz5+UPJ+j0trdu3Ihd58n6TuSvpVu/zJJB1bs7yKZ2Z/N7JSKaB9Jg5RcP3eupO95gW47VoXc1G56MuE7So5j70k6Scl13AuPQxx3K7Q47u6h5G4nh9jn92ytPC6UDXVbjrqVkmP+R0refRyefr11bNn2sPRCYHSAJTdZ3i79pQOFYWb3Sjra3ct8ph0lxHEXRUTdNgedXQAAAJRWni9jAAAAADqEzi4AAABKq0OdXTPbyZLp6V4ws84w4hMlQe2iqKhdFBF1i2aq+ppdM+si6V9KptF7XckIvn3d/ZnWlunbt68PGDCgqu2h85o0adJsd+9Xq/VRu2iE6dOna/bs2TW9n3R7a5e6RTU45qKoWqvdjsygNljSC+7+kiSZ2Rglt+potXgHDBigiRMndmCT6IzMbFEzbVWD2kXdDRo0qB6rbVftUreoBsdcFFVrtduRyxj664vT0L2uL04PB+QVtYuionZRRNQtmqojnd3Y23PBNRFmdriZTTSzibNmlfle7CgQahdF1WbtUrfIIY65aKqOdHZf1xfnXF5FyQweX+Duw919kLsP6tevZpcAAR1B7aKo2qxd6hY5xDEXTdWRzu4ESWub2Rpm1l3JdHVja9MsoK6oXRQVtYsiom7RVFUPUHP3+WZ2pKR7JHWRNMLdp9asZUCdULsoKmoXRUTdotk6cjcGufvdku6uUVuAhqF2UVTULoqIukUzMYMaAAAASovOLgAAAEqLzi4AAABKi84uAAAASovOLgAAAEqLzi4AAABKi84uAAAASqtD99kFAKDMLrvssiC7+eabg+zBBx9sRHMAVIEzuwAAACgtOrsAAAAoLTq7AAAAKC06uwAAACgtBqgBncz7778fZGPGjAmyYcOGBdlHH30UZIMHDw6ySy65JMg23XTTrE0EmmLevHlBdumllwbZzJkzg2z69OlBtvzyywfZEkssUV3jgAb65JNPguzOO+8Msquuuiq6/HXXXRdkffr06XjDqsSZXQAAAJQWnV0AAACUFp1dAAAAlBadXQAAAJRWhwaomdl0SXMlfSZpvrsPqkWjyu7FF1/M9Lh77703yI477rgg22uvvYJs//33D7JBg8JfzzLLLJOpLWXTWWr3nXfeCbKjjjoqyG688cZM6zOzIJswYUKQbbfddkF23333Bdlmm22Wabv4XGep3WaYNm1akH3jG98IsiuuuCLI1lxzzSB74okngix2HO4MqNv8+vjjj4Ps29/+dpDFZglceeWVo+ucM2dOkDVzgFot7sawjbvPrsF6gEajdlFU1C6KiLpFU3AZAwAAAEqro51dl3SvmU0ys8NjDzCzw81soplNnDVrVgc3B9QMtYuiWmTtUrfIKY65aJqOdna3dPdNJH1L0hFmtnXLB7j7cHcf5O6D+vXr18HNATVD7aKoFlm71C1yimMumqZD1+y6+xvp57fM7DZJgyWNr0XDmuWzzz4LsthMIl27hk/diBEjgmz48OFBNnny5CCLDfzJavTo0UF27bXXBtn5558fZD//+c+r3m6RFbl23T2ajx8fNv+HP/xhkMUGSHak/mI++OCDIIvVJAPU2q/ItZt3b7/9dpCNGzcu07JDhw4Nss46GC2Gus2H2CyBv/vd74LsgQceCLLu3btnepwkDRgwoP2Nq6Oqz+ya2ZJm1nvh15J2lDSlVg0D6oXaRVFRuygi6hbN1pEzuytIui09I9RV0vXu/peatAqoL2oXRUXtooioWzRV1Z1dd39J0oY1bAvQENQuioraRRFRt2g2bj0GAACA0qrFpBKlcsoppwRZbGDXWmutFWRZZ0briO985ztBNnbs2LpvF/kRG+glSYceemim5b/yla8E2S9+8YsgW2WVVTKtb7/99guy6dOnB9k999wTZLHBn4svvnim7QK1Fjv+P/fcc0HWo0ePIDvhhBPq0iaglmbPDuf0OOecc4Js5MiRQRabVa1v3741aVe9cWYXAAAApUVnFwAAAKVFZxcAAAClRWcXAAAApUVnFwAAAKXVae/GcPbZZ0fz8847L8hiU6nG7rwQG+V+1VVXBdmqq66aaRtXXnllkE2YMCHIYrbYYosg22WXXTIti/x4+umng+zwww/PvHxsOt6//vWvQdazZ8/2NaxCbDRu7G4MsezGG28MsqWXXjrIBg4cGGRrr712tgYCEbFp2//5z39mWvayyy4Lsg022KDDbQJqae7cuUG27rrrBlmsng844IC6tKlZOLMLAACA0qKzCwAAgNKiswsAAIDSorMLAACA0uq0A9T69+/foeVj0/bGpnHt3bt3pvXFLhD/3e9+F2SzZs0KstjAiNjUrEsuuWSmtqA55s+fH2SjR4/O9LjW7LjjjkHWkcFo8+bNC7L3338/yNw90/oOOeSQqtty8MEHB9kf/vCH6GO7d+9e9XZQLLG/jwcffDDIjj/++CCLTV/905/+NMgOPPDAKlsH1MdHH30UZD/+8Y+DbPnllw+yPffcsy5tyhPO7AIAAKC06OwCAACgtOjsAgAAoLTa7Oya2Qgze8vMplRkfczsPjN7Pv28bH2bCbQftYuionZRRNQt8irLALWRki6VVDn66iRJ97v7uWZ2Uvr9ibVvXv3EBrdI2QfMbLjhhkF21llnBdmzzz4bZGPHjg2y2Axqm2yySZD96le/CrLYAApIKljtxmZvuvDCC4Ns3333jS4/ZsyYIHv++eerbk9soM9RRx0VZM8991yQxeo5lnXEyJEjg6y1gWixwZ5du+Z6fO5IFah282SxxcJzOLEBwE899VSQxY65sZqPbQOSqNuGiA0Ajs24+sgjjwTZpEmTgqwzDF5v8y/W3cdLeqdFvJukUenXoyTtXuN2AR1G7aKoqF0UEXWLvKr25ekK7j5DktLP4b0sUmZ2uJlNNLOJsdtmAQ1G7aKoMtUudYuc4ZiLpqv7ezHuPtzdB7n7oH79+tV7c0DNULsoIuoWRUXtol6q7ezONLOVJCn9/FbtmgTUFbWLoqJ2UUTULZqu2hEaYyUdJOnc9PPtNWtRk/3rX/8Ksq9//etBduaZZ1a9jfPPPz/Ivve97wXZSiutFGTMBNVhua3d2Mw2sYGUF198cXT52AC1WBYbjLDEEktkWrYjby1us802QXbLLbcE2eTJk4PssMMOC7IXX3wxyGKDNFrT2mxrOZbb2s2T2O/19tvDp6pbt25BFjuur7vuurVpWOdF3dbY7Nmzg2zYsGFBFptJdbnllqtLm/Iuy63HbpD0mKR1zex1MztUSdHuYGbPS9oh/R7IFWoXRUXtooioW+RVm2d23T1+nyNpuxq3BagpahdFRe2iiKhb5BU3CwQAAEBp0dkFAABAaeV6CqFmWHvttYNs2223DbIbb7yx6m28807Le25L7777bpCtvvrqVW8DxdO7d+8gGzRoUM23c9VVVwVZrWc322GHHYLs1ltvDbLYwLitt946yB5//PEg++pXvxpkr732WrQ9d9xxR5AVcIBapxYbHHnMMccE2V/+8pdM6xs3blyQbb755u1uF1BPc+bMCbK99947yE444YQgi/VdOivO7AIAAKC06OwCAACgtOjsAgAAoLTo7AIAAKC0GKCWwahRo4LsiCOOCLLrrrsuyO66664gO+ecc4Ls//7v/4Jsq622yrRsbKBObHYg5NsyyywTZD/+8Y+DbMGCBdHlY7Vx8sknd7xhFVZbbbUgO+igg4IsNliiZ8+eVW+3T58+QXbJJZcE2R577BFdPja46frrrw+yoUOHVtE6NMJzzz0XZLHfYUzs+MpgNBTB6NGjg6xHjx5B9qtf/SrIunTpUpc2FRFndgEAAFBadHYBAABQWnR2AQAAUFp0dgEAAFBaDFDLIDbYa8stt8yUffDBB0E2derUINt5552DbPz48Zm2cfjhh2fKNt544yBD8Sy2WPw16vHHHx9kN910U5BNmjQp03Zig9vOOuusTMs2wqabbhpk7h597Pz584MsNqMbA9Ty4fnnnw+yv//970E2bNiwINt+++2D7Jvf/GZtGrYIH374YZBNmTIlyJZddtkgW2ONNaLr7NqVf9Gdyeuvvx5k5513XpCNHTs2yGKzUeJznNkFAABAadHZBQAAQGnR2QUAAEBptdnZNbMRZvaWmU2pyE43s3+b2eT0I7zgFGgyahdFRe2iiKhb5FWWq99HSrpU0rUt8ovc/fyat6hkllxyySAbPHhwkM2ePTvInnnmmSDba6+9guyKK64IsssvvzzIrrrqqiA79NBDg6xERqqT125sMJuZBdmee+4ZZGeccUZd2lQr//73v4Mstm+tiQ0G+eSTT4Js8cUXb1/DamOkOnHtxgYPxmbgO/vsszM9rtZiA+i22WabIHvjjTcyrS82+6Yk7bvvvu1rWPONVCeu2/aIDZqNDbrdddddg+wrX/lKXdpUZm2e2XX38ZLeaUBbgJqidlFU1C6KiLpFXnXkmt0jzeyp9G2L8F4qKTM73MwmmtnE2Pz0QBNQuyiqNmuXukUOccxFU1Xb2f2DpLUkbSRphqQLWnuguw9390HuPqhfv35Vbg6oGWoXRZWpdqlb5AzHXDRdVZ1dd5/p7vOWh1sAAAzxSURBVJ+5+wJJV0oKL0IFcojaRVFRuygi6hZ5UNX0LGa2krvPSL/dQ1I4TQw6bODAgUE2YcKEILv99tuD7Kijjgqyn/zkJ0HWvXv3INt///2j7WnP4J+8onbjll9++SDr0qVLE1qS3R133NGh5d97770g++yzzzq0znrqTLV74oknNrsJ//Xwww8H2WOPPRZkscFosRn9YjNyDhgwoLrGFUBnqtv2uO2224LsnXfCy53PPz8c11eG/8WN1mZn18xukDREUl8ze13SaZKGmNlGklzSdEk/qmMbgapQuygqahdFRN0ir9rs7Lp77N4nV9ehLUBNUbsoKmoXRUTdIq+YQQ0AAAClRWcXAAAApVXVADU0zxJLLBFksVl2YrO0rbPOOkF28MEHB9n3vve96LYbMTMRkFVsBrT2WH/99YMs9veFxnvxxReDbPXVVw+yrl1r+y9s7NixQXbqqacG2dSpUzOtL3bM3XHHHYNs8803z7Q+FFNsZsbTTjstyGIzXsbuN7zUUkvVpmGdCGd2AQAAUFp0dgEAAFBadHYBAABQWnR2AQAAUFqFHqC2YMGCIJs/f36QxWYJK7vYjDzrrbdekMUGWtx1113RdbY2cA2ot4ceeijIrrnmmiBrz8xCP/jBDzrUJtRPbAa10aNHB1nWAWpz584Nsq9+9atB9sILL2RaX8xqq60WZBMnTgyyjz/+uOptoJiuvPLKIIsNsI0Nmt1vv/2C7PHHH69NwzoRzuwCAACgtOjsAgAAoLTo7AIAAKC06OwCAACgtAo9QG3cuHFBFhvYcMEFFwTZ1ltvXY8m5UaXLl2CLOtgjo8++qjWzUGTnHzyyUG25557Btnll18eZMcdd1yQrbHGGrVpWCo2yHTatGlBFhscGVs2NgORJK244opB9q1vfStLE1Ejsd+XJF177bVBFhto3K1bt0yPiw1mPProo4Ms62C02MyRw4YNC7LYjFixwdG9e/fOtF2Ux/333x9k55xzTpB95StfCbLtttsuyGIDLqmrRePMLgAAAEqLzi4AAABKi84uAAAASovOLgAAAEqrzRFLZraqpGslrShpgaTh7n6xmfWRdKOkAZKmS9rb3f9Tv6aGtthiiyBz9yAbMmRIkB144IHRdZ566qlBtvbaa7e/cU329ttvB9krr7wSZLHnq3///nVpU6PluXYbZa211gqy2CxjsSw2495ll10WZOuuu26Qxerq2WefDbLbbrstyP785z8HWUxsMFprM6ide+65mdaZB2Wo29gg19Z+BzfddFOQTZgwIcgee+yxIDv++OOD7IknnsjSxGh9//SnPw2yPfbYI8hiAx5RjtrtqDlz5gTZPffcE2QXXXRRkH3yySdBFut/xAZNYtGynNmdL+lYd/+ypK9JOsLMBko6SdL97r62pPvT74E8oXZRRNQtioraRS612dl19xnu/mT69VxJ0yT1l7SbpFHpw0ZJ2r1ejQSqQe2iiKhbFBW1i7xq1zW7ZjZA0saS/i5pBXefISUFLmn5VpY53MwmmtnEWbNmday1QJWoXRQRdYuionaRJ5k7u2bWS9Itkoa5e3hRSivcfbi7D3L3Qf369aumjUCHULsoIuoWRUXtIm8yTallZt2UFO517n5rGs80s5XcfYaZrSTprXo1sjU9evQIskcffTTIDj300CC78cYbo+t84IEHguyGG24IskGDBgXZ4osvHl1nvcVmETriiCOC7N133w2y2HO4+eab16ZhOZDX2m2U2Iw8u+8evoN4++23B9m8efOC7LDDDsu03dgAtdYGj1VrqaWWCrJLLrkk+tj999+/ptuut6LXbeyYeeaZZ0YfGxvguNVWWwXZLrvskmnba665ZpB94xvfCLLYrGobbLBBpm2gdUWv3Y4aP358kH388ceZlh05cmSQxY5zWWdDxefaPLNryX+oqyVNc/cLK340VtJB6dcHSQr/WwJNRO2iiKhbFBW1i7zK8vJgS0kHSHrazCan2SmSzpV0k5kdKulVSXvVp4lA1ahdFBF1i6KidpFLbXZ23f1hSa29/7hdbZsD1A61iyKiblFU1C7yihnUAAAAUFp0dgEAAFBapRvS17179yAbPXp0kJ122mnR5Xfeeecg23rrrYNslVVWCbLYiPaNNtooup0s3n///SCLTaN5+umnB9lDDz0UZLHR8KecckqQMRVhuV1//fVBdsghhwRZbBrXZjnuuOOC7IQTTgiyPn36NKI5aMPee+8dZNddd130sQ8++GCQ/exnPwuyX/3qV0F21llnBdl7770XZN26dQuyJZZYItoeoCO23377INt0002D7IILLgiyUaNGBdmWW25Zm4Z1cpzZBQAAQGnR2QUAAEBp0dkFAABAadHZBQAAQGmVboBaVl/60pei+ZQpU4Ls2WefDbLYlKuDBw8OsthAixVXXDHIJk+eHGR33313kMUGX8RsscUWQfaHP/whyAYOHJhpfSiP2LTW11xzTZDFpgYeNmxYkPXu3TvIHn/88SCL1doBBxwQZAceeGCQLbfcckHGlJn51atXryC7//77G7LtpZdeuiHbAWJ69OgRZH/961+DLDa1eWwg2+WXX16bhnVynNkFAABAadHZBQAAQGnR2QUAAEBp0dkFAABAaTHCo4XYDGwbbLBBkE2dOjXIZsyYkWkbv/71r4PshhtuyLRsbMDb0KFDgyw2GGjJJZfMtA10PrFBa9tuu22QPfXUU41oDgCURmwQ76mnnpopQ21wZhcAAAClRWcXAAAApUVnFwAAAKXVZmfXzFY1swfNbJqZTTWzo9P8dDP7t5lNTj92rn9zgeyoXRQRdYuionaRV1kGqM2XdKy7P2lmvSVNMrP70p9d5O7n1695+dWzZ88gW3PNNTMtO3LkyEwZOozaRRFRtygqahe51GZn191nSJqRfj3XzKZJ6l/vhgEdRe2iiKhbFBW1i7xq1zW7ZjZA0saS/p5GR5rZU2Y2wsyWbWWZw81soplNnDVrVocaC1SL2kURUbcoKmoXeZK5s2tmvSTdImmYu8+R9AdJa0naSMkruQtiy7n7cHcf5O6D+vXrV4MmA+1D7aKIqFsUFbWLvMnU2TWzbkoK9zp3v1WS3H2mu3/m7gskXSlpcP2aCVSH2kURUbcoKmoXeZTlbgwm6WpJ09z9wop8pYqH7SFpSu2bB1SP2kURUbcoKmoXeZXlbgxbSjpA0tNmNjnNTpG0r5ltJMklTZf0o7q0EKgetYsiom5RVNQucinL3RgelmSRH91d++YAtUPtooioWxQVtYu8YgY1AAAAlBadXQAAAJQWnV0AAACUFp1dAAAAlBadXQAAAJQWnV0AAACUFp1dAAAAlJa5e+M2ZjZL0iuS+kqa3bAN1xf7Un+ru3tTJ0qndnMtr/uRp7qV8vs8tVdZ9kPK775Qu/VRlv2Q8rsv0dptaGf3vxs1m+jugxq+4TpgXzqXMj1HZdmXsuxHvZXleSrLfkjl2pd6KsvzVJb9kIq3L1zGAAAAgNKiswsAAIDSalZnd3iTtlsP7EvnUqbnqCz7Upb9qLeyPE9l2Q+pXPtST2V5nsqyH1LB9qUp1+wCAAAAjcBlDAAAACgtOrsAAAAorYZ3ds1sJzN7zsxeMLOTGr39jjCzEWb2lplNqcj6mNl9ZvZ8+nnZZrYxCzNb1cweNLNpZjbVzI5O88LtSyNRu81H7bYfdZsP1G77UbvNV5a6bWhn18y6SPq9pG9JGihpXzMb2Mg2dNBISTu1yE6SdL+7ry3p/vT7vJsv6Vh3/7Kkr0k6Iv09FHFfGoLazQ1qtx2o21yhdtuB2s2NUtRto8/sDpb0gru/5O7zJI2RtFuD21A1dx8v6Z0W8W6SRqVfj5K0e0MbVQV3n+HuT6Zfz5U0TVJ/FXBfGojazQFqt92o25ygdtuN2s2BstRtozu7/SW9VvH962lWZCu4+wwpKQpJyze5Pe1iZgMkbSzp7yr4vtQZtZsz1G4m1G0OUbuZULs5U+S6bXRn1yIZ9z5rEjPrJekWScPcfU6z25Nz1G6OULuZUbc5Q+1mRu3mSNHrttGd3dclrVrx/SqS3mhwG2ptppmtJEnp57ea3J5MzKybksK9zt1vTeNC7kuDULs5Qe22C3WbI9Ruu1C7OVGGum10Z3eCpLXNbA0z6y5pH0ljG9yGWhsr6aD064Mk3d7EtmRiZibpaknT3P3Cih8Vbl8aiNrNAWq33ajbnKB2243azYHS1K27N/RD0s6S/iXpRUmnNnr7HWz7DZJmSPpUyavOQyUtp2Qk4vPp5z7NbmeG/dhKydtBT0manH7sXMR9afDzRu02fz+o3fY/Z9RtDj6o3aqeM2q3+ftRirplumAAAACUFjOoAQAAoLTo7AIAAKC06OwCAACgtOjsAgAAoLTo7AIAAKC06OwCAACgtOjsAgAAoLT+P58QtAf/PGrbAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 864x864 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 재연성을 위하여 랜덤시드 고정\n",
    "random.seed('intel') \n",
    "\n",
    "# TEST 데이터 예측하기\n",
    "predicted_result = model.predict(X_test) \n",
    "predicted_labels = np.argmax(predicted_result, axis=1)\n",
    "\n",
    "# TEST 데이터의 정답 가져오기\n",
    "test_labels = np.argmax(y_test, axis=1)\n",
    "\n",
    "# 잘못 예측한 데이터 찾기\n",
    "wrong_result = []\n",
    "for n in range(0, len(test_labels)):\n",
    "    if predicted_labels[n] != test_labels[n]:\n",
    "        wrong_result.append(n)\n",
    "\n",
    "# random 함수를 통해서 임의의 16개 데이터 가져오기\n",
    "# samples = random.choices(population=wrong_result, k=16)\n",
    "wrong_result = np.array(wrong_result)\n",
    "samples = wrong_result[np.random.randint(len(wrong_result),size=16)]\n",
    "\n",
    "# MNIST를 그릴 Figure 준비\n",
    "plt.figure(figsize=(12,12))\n",
    "\n",
    "# 16개의 이미지 시각화\n",
    "for count, n in enumerate(samples,start=1):\n",
    "    plt.subplot(4, 4, count)\n",
    "    plt.imshow(X_test[n].reshape(28, 28), cmap='Greys', interpolation='nearest')\n",
    "    tmp = \"Label:\" + str(test_labels[n]) + \", Prediction:\" + str(predicted_labels[n])\n",
    "    plt.title(tmp)\n",
    "\n",
    "plt.savefig('./img/result/mnist_wrong_result.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "놀랍게도 모델이 틀린 데이터는 언뜻보면 사람조차도 헷갈릴 수 있을만한 데이터라는 걸 확인할 수 있다.\n",
    "\n",
    "아래 그림은 CNN 모델일 학습이 진행될때마다 활성화되는 부분을 시각화한 자료이다.\n",
    "\n",
    "![](https://thumbs.gfycat.com/SmoggyLittleFlickertailsquirrel-size_restricted.gif)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 다음 예제에서는\n",
    "\n",
    "CNN을 통해서 MNIST와 같은 간단한 예제를 분류할 수 있었다. 다음 예제에서는 RGB 이미지를 한번 분류하는 예제를 더 깊은 CNN으로 수행해보도록 하자."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## 참고\n",
    "\n",
    "- Intel\n",
    "    - https://www.intel.co.kr/\n",
    "- Intel OpenVINO\n",
    "    - https://software.intel.com/en-us/openvino-toolkit\n",
    "- MNIST\n",
    "    - http://yann.lecun.com/exdb/mnist/\n",
    "    - https://pinkwink.kr/1121\n",
    "- CIFAR10\n",
    "    - https://www.cs.toronto.edu/~kriz/cifar.html\n",
    "- ImageNet\n",
    "    - http://www.image-net.org\n",
    "- Tensorflow\n",
    "    - https://www.tensorflow.org/?hl=ko\n",
    "- Keras\n",
    "    - https://keras.io/\n",
    "    - https://tensorflow.blog/2019/03/06/tensorflow-2-0-keras-api-overview/\n",
    "    - https://tykimos.github.io/2017/02/22/Integrating_Keras_and_TensorFlow/\n",
    "    - https://tykimos.github.io/2017/03/08/CNN_Getting_Started/\n",
    "    - https://raw.githubusercontent.com/keras-team/keras-docs-ko/master/sources/why-use-keras.md\n",
    "- Keras to Caffe\n",
    "     - https://github.com/uhfband/keras2caffe\n",
    "     - http://www.deepvisionconsulting.com/from-keras-to-caffe/\n",
    "- Fully Connected Layer\n",
    "    - https://sonofgodcom.wordpress.com/2018/12/31/cnn%EC%9D%84-%EC%9D%B4%ED%95%B4%ED%95%B4%EB%B3%B4%EC%9E%90-fully-connected-layer%EB%8A%94-%EB%AD%94%EA%B0%80/\n",
    "- Convultional Nueral Network\n",
    "    - http://aikorea.org/cs231n/convolutional-networks/\n",
    "    - http://cs231n.stanford.edu/\n",
    "- CNN Models\n",
    "    - https://ratsgo.github.io/deep%20learning/2017/10/09/CNNs/\n",
    "\n",
    "- VOC2012\n",
    "    - https://blog.godatadriven.com/rod-keras-multi-label\n",
    "    - https://gist.github.com/rragundez/ae3a17428bfec631d1b35dcdc6296a85#file-multi-label_classification_with_keras_imagedatagenerator-ipynbhttps://fairyonice.github.io/Part_5_Object_Detection_with_Yolo_using_VOC_2012_data_training.html\n",
    "    - http://research.sualab.com/introduction/2017/11/29/image-recognition-overview-1.html"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:work]",
   "language": "python",
   "name": "conda-env-work-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "강의목차",
   "title_sidebar": "강의목차",
   "toc_cell": true,
   "toc_position": {
    "height": "504px",
    "left": "27px",
    "top": "110px",
    "width": "264.297px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}