Lab08_Cifar10.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Cifar10.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jeonggunlee/DeepLearningBasics/blob/master/Lab08_Cifar10.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FpdBPshVqqJj",
        "colab_type": "text"
      },
      "source": [
        "# Cifar 10 분류!\n",
        "\n",
        "## 한림대학교 소프트웨어 중심대학\n",
        "### 소프트웨어 융합대학 이정근 교수 (jeonggun.lee@gmail.com)\n",
        "\n",
        "## 딥러닝 기초:오픈소스 케라스와 함께\n",
        "\n",
        "### Cifar 10 분류 예제\n",
        "\n",
        "머신러닝을 공부하거나, 어떤 머신러닝 알고리즘을 설계했을때 알고리즘을 검증하기 위해서 가장 먼저 적용해보는 데이터셋은 **MNIST** 데이터셋이며,\n",
        "MNIST 데이터셋 다음에는 많은 사람들이 보통 C**IFAR-10** 데이터셋을 활용하여 딥러닝을 학습한다. **CIFAR-10** 데이터셋은 아래와 같이 총 **10개의 레이블**로 이루어진 이미지 분류를 위한 데이셋이다.\n",
        "\n",
        "\n",
        "**airplane, automobile, bird, cat, deer, dog, frog, horse, ship, and truck**\n",
        "\n",
        "*  *  *\n",
        "![Cifar-10](https://cdn-images-1.medium.com/max/1200/1*SZnidBt7CQ4Xqcag6rd8Ew.png)\n",
        "*  *  *\n",
        "\n",
        "\n",
        "각각의 레이블마다 32×32 크기 이미지인 50,000개의 training 데이터셋, 10,000개의 test 데이터셋이 존재하고, 결과적으로 총 60,000개의 32×32 크기의 이미지로 데이터셋이 구성되어 있다.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f4HbPgO_glsG",
        "colab_type": "code",
        "outputId": "39a8c90d-c960-4155-87e0-b955491e4a52",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 125
        }
      },
      "source": [
        "   \n",
        "from __future__ import print_function\n",
        "import keras\n",
        "from keras.datasets import cifar10\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout, Activation, Flatten\n",
        "from keras.layers import Conv2D, MaxPooling2D\n",
        "import os\n",
        "\n",
        "batch_size = 32\n",
        "num_classes = 10\n",
        "epochs = 10\n",
        "data_augmentation = True\n",
        "num_predictions = 20\n",
        "save_dir = os.path.join(os.getcwd(), 'saved_models')\n",
        "model_name = 'keras_cifar10_trained_model.h5'\n",
        "\n",
        "# The data, split between train and test sets:\n",
        "(x_train, y_train), (x_test, y_test) = cifar10.load_data()\n",
        "print('x_train shape:', x_train.shape)\n",
        "print(x_train.shape[0], 'train samples')\n",
        "print(x_test.shape[0], 'test samples')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
            "170500096/170498071 [==============================] - 13s 0us/step\n",
            "x_train shape: (50000, 32, 32, 3)\n",
            "50000 train samples\n",
            "10000 test samples\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CkOAI2AED1OY",
        "colab_type": "code",
        "outputId": "afb46c6a-4a24-48ec-e695-079dc8b56e33",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        }
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "\n",
        "plt.imshow(x_train[0])      # 개구리\n",
        "print(y_train[0])           # 개구리"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[6]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r5a0iGlPJLbc",
        "colab_type": "code",
        "outputId": "395953df-2ec0-4798-f14d-44b5b5ee1340",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        }
      },
      "source": [
        "plt.imshow(x_train[1])      # 트럭\n",
        "print(y_train[1])           # 트럭"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[9]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKAp+IRKKgl+IhKLgFyKhKPiFSCj/D/EcoRMKOMsFAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BmAGt0DTJ6xL",
        "colab_type": "code",
        "outputId": "2b92dd7f-b4d8-48f5-c863-e1493a450925",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        }
      },
      "source": [
        "plt.imshow(x_train[2])      # 트럭\n",
        "print(y_train[2])           # 트럭"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[9]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAG2RJREFUeJztnXuMnFd5xp93Lnvz7tpe3+LYTjcJ\nKZdyCdFiKKE0BIFSlCqhrSKQilIJYVSBVCT6R5RKJZX6B1QFxB8VyJCIgCiXkqSkKAVCCrlwSbIJ\nie3ESezE68tmvd61vd7L7M717R8zQc7mPGfHe5m1Oc9Psjx73j3fd+bM98w3c55932PuDiFEemRW\newBCiNVB4hciUSR+IRJF4hciUSR+IRJF4hciUSR+IRJF4hciUSR+IRIlt5TOZnYdgK8AyAL4hrt/\nPvb7Gzdu9P7+/qWcUrSYWq1GY5VKhcZyuWyw3Wv8L0ozGX4vsozRGMBj7Gyxo13IDA0NYXx8vKmn\nt2jxm1kWwH8A+ACAYwAeN7N73f1Z1qe/vx+Dg4PBWOwiE8tA5K+4zfi1MjtToLGTp8ZprK9vfbC9\nWpqjfTq7umgs29ZOY278TaNGZB5+a7rw2blzZ9O/u5SP/TsBHHT3l9y9BOB7AG5YwvGEEC1kKeLf\nBuDoWT8fa7QJIS4AVnzBz8x2mdmgmQ2OjY2t9OmEEE2yFPEPA9hx1s/bG22vwt13u/uAuw9s2rRp\nCacTQiwnSxH/4wCuMLNLzawNwEcA3Ls8wxJCrDSLXu1394qZfRrAT1FfPL3D3Z9Z7PFiNo9YPYqF\nMzR26thLNHZ0f7jfmckZ2ufqa99PY72dHTQWu4cZWe3X1bZEn9/d7wNw3zKNRQjRQvQGKESiSPxC\nJIrEL0SiSPxCJIrEL0SiLGm1fznR/gErS2x+M8Zjx48eorE9v3mIxsqz4YSgfHc44QcAZie5rdjb\n10djLHkH4Ek/utp05xciWSR+IRJF4hciUSR+IRJF4hciUc6b1f5YKSmxdBy8TFq5yEt1vXz0MI31\ndnXSWNe6nmD7idNTtM/JkddkhP+eLTsuoTFkeFEuWsMvWhMwDXTnFyJRJH4hEkXiFyJRJH4hEkXi\nFyJRJH4hEuW8sfrE8sASeGLJO2OnTtLY0NARGitG+vV0tAXbC9OTtM9zT/+Oxi7qv5zG1l0U2S6C\nzEcsjywV21l3fiESReIXIlEkfiESReIXIlEkfiESReIXIlGWZPWZ2RCAKQBVABV3H1iOQYmlwKyt\nKu0xfOwYjR06wmNHD/Ltujb2dAfbt29cQ/uMHOEZhHsHH6exgWvW0VhX79pwIA03L8py+Pzvc/fx\nZTiOEKKF6GO/EImyVPE7gJ+Z2RNmtms5BiSEaA1L/dj/HncfNrPNAO43s+fc/VXF3BtvCrsA4JJL\nItVYhBAtZUl3fncfbvx/AsA9AHYGfme3uw+4+8CmTZuWcjohxDKyaPGb2Roz63nlMYAPAti3XAMT\nQqwsS/nYvwXAPY0MqByA/3T3nyz+cLzA5OJ8mRXwckgmmMc2f/LI84pkj9mi35fDx6zVKrRHuVKm\nsanCHI0dGz1FY6MkVq1upn22b+bP+bnHH6OxzRdtpbE/fsdrPow24Jd+xiOvS2yfr8hLFjkkLHaN\nrCCLFr+7vwTgbcs4FiFEC5HVJ0SiSPxCJIrEL0SiSPxCJIrEL0SinEcFPGMeymKOtkirLzYMWgyS\nd3Jwiy1q50VtwFjs3COX9PfTWFdPL41NzszSGCz83PYdPUG7dObaaSw3V6KxZ379II1t2LYl2L5+\n+2W0j1X462kRzy52zdUy/JiR0IqiO78QiSLxC5EoEr8QiSLxC5EoEr8QiXIerfYv7/tQNAEjQmzl\nHrVwrBapj1eu8FXqtrbwllYAYNEnEFtxZl2ytM/69Rtp7D3vvYbG9j71HI0NHQrX46tW+FwdzB6n\nsY7+i2ms+vwBGtv74K+C7e/8S55e3tkVrj8IANVYgk4sxkOoLMLpYo7PuRxJd34hEkXiFyJRJH4h\nEkXiFyJRJH4hEkXiFyJRzh+rL1rkbDHHiyXbRBI3IoeseDhJ58BBbjXNzs7Q2Bve+EYaa2/n1lwm\n5ikRas6PV4tcBu+++s9o7MihYRr7xte+EWyvzHLr88jYBI21d/Gknyv6+D3s+YcHg+2bIok9b7ia\n1f0DCpFErXyNj6Mt8pqdKpwJthdLRdqHWaalMu8zH935hUgUiV+IRJH4hUgUiV+IRJH4hUgUiV+I\nRFnQ6jOzOwBcD+CEu7+50dYH4PsA+gEMAbjJ3U8vZSC1iDXHEtyitfOqkdp5sbe8iCVzdPhIsP1/\n7vsx7TM5GbZxAODd47ye3fv+/Foaa2/nthebx9iGUJUqj3b39NDY9TdcT2MHn38h2P7z/72f9pks\n89fsuWGe8bfeOmmsYy78Yv/2Jz+jfXIbeFZfZss6GpuZ4K91vsazGUcmjwXbz0zx483NhbdRmy5M\n0j7zaebO/00A181ruwXAA+5+BYAHGj8LIS4gFhS/uz8EYP6uizcAuLPx+E4ANy7zuIQQK8xiv/Nv\ncfeRxuPjqO/YK4S4gFjygp/Xv3jzAjJmu8xs0MwGx8bGlno6IcQysVjxj5rZVgBo/E9Xrtx9t7sP\nuPvApk28dJIQorUsVvz3Ari58fhmAD9anuEIIVpFM1bfdwFcA2CjmR0D8DkAnwfwAzP7OIDDAG5a\n+lC4FcK8udOnT9IuZ07PX6M863BZbucdH+P2228GHwu2P/HM07TP5CmeqVYs8wy3P3nLm2ls8yZe\ncDObDb+kk1MF2mdigo+xf/t2Grt4+2Ya+7tP/G2w/ejwi7TPo0/vobHiDM9KPHCM24BdF4X7ndy3\nj/Yp3E1DuPzqq2js9PQUP2bEgitaeP5jGXo1Ukw2VjB2PguK390/SkLvb/osQojzDv2FnxCJIvEL\nkSgSvxCJIvELkSgSvxCJ0uICng4gbF/UIllPrKrmmclx2uXhXz9CY4dfDmdRAcD4JLe9Ts+ErZzM\nGr7nXkdxDY2dOBkb/8M01t+/g8ZYxt/wMf7XleUSt4dmC3w+pqd4LE+urDe+gxfOfOrgXhorTfEM\nzmMT3EbragvPx/a1HbTPocEnaSzbzu+XmYv7aOxMhVut1MR0fl0Vi2EdeSx9cx668wuRKBK/EIki\n8QuRKBK/EIki8QuRKBK/EInSUqtvdq6AZ/aHM+ByuTztx6yo05FstIlpXvzwyAjfY27t5g001rc2\nXChyw0Zep2DsxREa27+PW1v3/5wXulzbywtWZnNh46hY4lZZqRguBgkAP/kpj+Ujtw6W8de1kb/O\nb7vyDTT2u0eep7FCpDzpCydHg+2dVW7Brq/woqUHf/sEjU1s4vbhqQwfY74U7leJFDQtFMLW4dTk\nLO0zH935hUgUiV+IRJH4hUgUiV+IRJH4hUiUlq72z8xM49eP/ToYm52cof3WdIRXZq+//gbap+J8\nS6sn9j5HY2t71tPYbC288n3xZr5tQXmUr76emeHJHoUDfHV7fSS5ZM3a8Fx1r+eORMcavhK9dh2v\nnbe2t5fGenvDW151dnfRPtdc+04aOzPO3Zt9+16isWo5nBV2ZCLiYuS5I5E7zlfgp07zWKWHOzSZ\nznBNxuGj3CmaJHopzTVfw093fiESReIXIlEkfiESReIXIlEkfiESReIXIlGa2a7rDgDXAzjh7m9u\ntN0G4BMAXikMd6u737fQsYrFEl4aCtsyZ06cpv2uuPSKYHtnJ0/OePllvu3W4UNHaKx7DbdkiuWw\nNWeRZIrZCW7/IMO3DXvd5bzW3eWb1tJYz/qw/XbiBLfK1vfxe8DWHXyOpya5VdlG3MOOGrcOeyPP\n6wPXvY/GTp3mNfxGj4Wvg/Eitze7zvDjbY7YmznjyVPbenh9vzVbLgq2Dw8N0T6lQriepMdqYc6j\nmTv/NwFcF2j/srtf2fi3oPCFEOcXC4rf3R8CwHe9FEJckCzlO/+nzWyPmd1hZvzP4oQQ5yWLFf9X\nAVwO4EoAIwC+yH7RzHaZ2aCZDRYKzRcaEEKsLIsSv7uPunvV3WsAvg5gZ+R3d7v7gLsPdHXxxTQh\nRGtZlPjNbOtZP34YwL7lGY4QolU0Y/V9F8A1ADaa2TEAnwNwjZldifr+W0MAPtnMyWrVKmbOhC2n\nwhz/StDeFa5xdmaK21eHjw7R2Lq13K6pzvBsL5sLb5E0cvwg7TPyMt+SyzLh4wHATX/9VzRWm+br\nr//3yC+D7Yf38LqFG9bybaGOH+B25LaLL6GxM+Vw7TzkuQXbt4FnR77l9W+msdKN/DK+4/ZvB9tn\np/jr/PLENI0hF9lCq8Ttw+nxkzR2Mbke2zp5duHGzeuC7eMnyLwHWFD87v7RQPPtTZ9BCHFeor/w\nEyJRJH4hEkXiFyJRJH4hEkXiFyJRWlrAs+Y1lIphS69Q5AU8Dx4KW2n3/PddtM8jDz5IY+bcvhqd\n5DbP2OGjwfY8d3hQjmRZtV3Es9h+9dDDNFac5PbhswdeCLbPjPLswokxPsZ1G/gWVGORYpaTZ8Kv\n5/p1/A+9StXw2AHgl798ksY6e/kWa+s3hrcNGy9z661Q5M9rOGIReju/rrrIfABAdixsf67bwK+P\nbDYs3RcP8GKm89GdX4hEkfiFSBSJX4hEkfiFSBSJX4hEkfiFSJSWWn3ZXBZr+8L2RTnyNjQ5HS6o\n+OxTT9E+o4cO0Vgm8rS7cjyTqi0TzujyEt8fLQNu/2zfuo3G+iJ7Bp6OFEW5rP/1wfbDVV4gdeIU\nt72q7eHsMQAYjWRAFgph+3DiFM86sywv7jlnkfEXXqSxTFvYWqxleXaet/FxFMB93WqFx9aQcQBA\n99rwa53NclHUPDy/2cgczkd3fiESReIXIlEkfiESReIXIlEkfiESpbWr/dksuslqf66HbwtVOhlO\nihh/IZxoAwA7unlShJFVewCYmuUr2HOZcMKHdfLkl3bjq69jo7wW3xOPPk1jW3p6aOzk6Ylg+5lZ\n7hBMRxKTZsf51lWIOBk5spremedbWs1FXJOxifDzAoBqhs9xVy68ym4Zft/LdMRWzCOT5WUampnh\n8z9Jtntbv4E7LaixueevyXx05xciUSR+IRJF4hciUSR+IRJF4hciUSR+IRKlme26dgD4FoAtqG/P\ntdvdv2JmfQC+D6Af9S27bnJ3nn0BwA2otYXfb7zKLYo2kuCQL/Pac5f09tFYJWINTUUssWxvd7A9\n08atvtlRvqVYcaLAx3FyisbGa/w9e6IYPmb/VW+lfY6P8cSeidN8/N3d3J6dK4Tt2XKez9VcpHbe\nbJlbbJkMv3Y6yGvjxm25asTOy+a4ZDIVbmPWavyYJ8bCNmaFX97ItYWfc6UasSLn0cydvwLgs+7+\nJgDvAvApM3sTgFsAPODuVwB4oPGzEOICYUHxu/uIuz/ZeDwFYD+AbQBuAHBn49fuBHDjSg1SCLH8\nnNN3fjPrB/B2AI8C2OLuI43QcdS/FgghLhCaFr+ZdQO4C8Bn3P1Vf/Pp7o76ekCo3y4zGzSzwcI0\n/z4thGgtTYnfzPKoC/877n53o3nUzLY24lsBBHcecPfd7j7g7gNd3byaiRCitSwofjMzALcD2O/u\nXzordC+AmxuPbwbwo+UfnhBipWgmq+9qAB8DsNfMXimadyuAzwP4gZl9HMBhADctdKBqtYaJibCF\nVSzwjK41pbA1t+mii2mfk4fDWyABwMGhwzQ2VuZZfX19Yfsw08E/0czUuPtZLXOLqlIo0thckXtA\nFQvbTWPH+RZfM9PccvQyt6+62rtorESyI629nfapzPHn3LaG24oesbfmiuHrqpbhz6tU4ddie55n\nhLZ18OfW3RW2iQGgk8TKkbnPsKxE3uU1LCh+d38EPE/w/c2fSghxPqG/8BMiUSR+IRJF4hciUSR+\nIRJF4hciUVpawBM1A2bJdljc5UHFwvbKTKTO4kikcOZIZFul6VIkK+pkOMMtm+dWWSGSzeW0CCMw\nW+EZbk62agKANmJFDY9xqy+WCWaRgpBjpyNJnBbu51U+9nwnt0x727jFVo2kv9X/+PS1ZHP8vtcJ\nvmVbJrKFVj5iA1pk/E6uEYucK2NEumTeg8do+jeFEH9QSPxCJIrEL0SiSPxCJIrEL0SiSPxCJEpL\nrT4zQ87CNkqZWDIAMD0b9gFPTfJ95E6VuHdYyfOn7RVuEc6xTDWSOQYAZY8VnuTnWrO2l8ayWd6P\nFZj0yNs8s8MWPFckxopqRrbIQy22f170OfM5rtbCNqBHin7GzkWz6VC/vnmQ96uRMUbcXlRYMPJa\nzkd3fiESReIXIlEkfiESReIXIlEkfiESpaWr/bVqFdNT08HY5GR4eycAmCElv2dmeL292MJr7zq+\nkt7eyeuw0XNFVoA7czyhI9/GzxVbSc9H3Aq22l+NJRhFV4h5LNYty+aE1BgEgGok6YeubiM+/jLp\nV408r2yOz30usl1XbBwdHXybsnbyejpxAQCgndRCjDoO89CdX4hEkfiFSBSJX4hEkfiFSBSJX4hE\nkfiFSJQFrT4z2wHgW6hvwe0Adrv7V8zsNgCfADDW+NVb3f2+2LEqlQrGT54MxsolbmvMzYUTZ0ol\nnlCT7+B12PId3H6bneU7CbP6bbEEHURi7pHtuqrc2srE6s91EQsollETsahiFmEMZjnFagLGKBR4\nncSYRZhjNloksSc2VzErLW6ZRp436dYR2QaOWX2xxKP5NOPzVwB81t2fNLMeAE+Y2f2N2Jfd/d+b\nPpsQ4ryhmb36RgCMNB5Pmdl+ANtWemBCiJXlnL7zm1k/gLcDeLTR9Gkz22Nmd5jZ+mUemxBiBWla\n/GbWDeAuAJ9x90kAXwVwOYArUf9k8EXSb5eZDZrZYLEYKc4vhGgpTYnfzPKoC/877n43ALj7qLtX\n3b0G4OsAdob6uvtudx9w9wG2SCGEaD0Lit/qy5u3A9jv7l86q33rWb/2YQD7ln94QoiVopnV/qsB\nfAzAXjN7qtF2K4CPmtmVqBsVQwA+udCBau4ol4k9Fykyl8uFbbvYB4n2yNZPMdeF7YIE8Ey7WsTh\nqUbsvJhFlY1YhNm2SI25fHge28gcAnGLKjbGuLUVJpKoFrWp1q1bR2PlcpnGisQOrkayCxdr58Uy\nDysVPkZUWezcX5dqZOu1+TSz2v8IwnKJevpCiPMb/YWfEIki8QuRKBK/EIki8QuRKBK/EInS0gKe\nuVwOGzZsCMYy4FZUtRq2PMqVyDZNEStnbo5n7lk2ku1FtlyqRTLfShHrJVuLZANGiBX3rHnYAorN\n1WIz7WK1ImvE/6xUuNdXI68zEC+qGbPYWAHPci2SNRmZ38XagNGtzYilF7NZ2TXnke3hXnteIUSS\nSPxCJIrEL0SiSPxCJIrEL0SiSPxCJEpLrb5sNove3vA+ebVqrMBh+D2qWOKZUpOF8J6AAJDLRzLm\nIjFqvUQy1fKRTLVKxCKsxWweYucBAIgdaZHswmhaYoRaxNqqEYvTI/ebWsSmKs3yYq2xrL4ay4yL\nFPCMzUbM1vVIz67IXn1txMbMRGxFtmfguRTw1J1fiESR+IVIFIlfiESR+IVIFIlfiESR+IVIlJZa\nfQBg5P3GIll4pXK43v9ckWfn0UKhiGdt5SJWiRP7qhTJKitGsthskfvFxSwgZvXUKnx+F7nDHGL5\nY07GGNv7z43HMjk+knyWZ4Tyc0Vi0YKmEXszNpERGzND7NlYn0o5fF0pq08IsSASvxCJIvELkSgS\nvxCJIvELkSgLrvabWQeAhwC0N37/h+7+OTO7FMD3AGwA8ASAj7k7X2IHAOeJEcViLHEjHCuV5mif\nUuR4pTJfnY8ll7Bad7H6bB2RPcUykbp01YiDEFuNZvNrke2/YjX8YokibZHnzZib469ZrBZfNjKO\n2PyzuYrtGF0oRGo8RpyWjkjyTmz8lVJ4LNQFANDREb6uYuN7zfGb+J0igGvd/W2ob8d9nZm9C8AX\nAHzZ3V8H4DSAjzd9ViHEqrOg+L3OK/mx+cY/B3AtgB822u8EcOOKjFAIsSI09Z3fzLKNHXpPALgf\nwIsAJtz9lc9pxwBsW5khCiFWgqbE7+5Vd78SwHYAOwG8odkTmNkuMxs0s8HZWf5dSgjRWs5ptd/d\nJwD8AsCfAlhn9vvd7LcDGCZ9drv7gLsPdHZ2LmmwQojlY0Hxm9kmM1vXeNwJ4AMA9qP+JvA3jV+7\nGcCPVmqQQojlp5nEnq0A7jSzLOpvFj9w9x+b2bMAvmdm/wrgdwBuX+hA7k7rrcUScagFFLG8WI0z\nAEDU9uIwSylmh3kkeYdtJQXExx/bxslImk42kvySic3HIrencmI5trW1RcbB53GxFmE+H37e0e2z\nIuOIzX1sHG3EmgOArvauYHvsWmSvy7lsvbag+N19D4C3B9pfQv37vxDiAkR/4SdEokj8QiSKxC9E\nokj8QiSKxC9EoljMrln2k5mNATjc+HEjgPGWnZyjcbwajePVXGjj+CN339TMAVsq/led2GzQ3QdW\n5eQah8ahcehjvxCpIvELkSirKf7dq3jus9E4Xo3G8Wr+YMexat/5hRCriz72C5EoqyJ+M7vOzJ43\ns4NmdstqjKExjiEz22tmT5nZYAvPe4eZnTCzfWe19ZnZ/WZ2oPH/+lUax21mNtyYk6fM7EMtGMcO\nM/uFmT1rZs+Y2T802ls6J5FxtHROzKzDzB4zs6cb4/iXRvulZvZoQzffNzOeItkM7t7SfwCyqJcB\nuwxAG4CnAbyp1eNojGUIwMZVOO97AVwFYN9Zbf8G4JbG41sAfGGVxnEbgH9s8XxsBXBV43EPgBcA\nvKnVcxIZR0vnBPVs3u7G4zyARwG8C8APAHyk0f41AH+/lPOsxp1/J4CD7v6S10t9fw/ADaswjlXD\n3R8CcGpe8w2oF0IFWlQQlYyj5bj7iLs/2Xg8hXqxmG1o8ZxExtFSvM6KF81dDfFvA3D0rJ9Xs/in\nA/iZmT1hZrtWaQyvsMXdRxqPjwPYsopj+bSZ7Wl8LVjxrx9nY2b9qNePeBSrOCfzxgG0eE5aUTQ3\n9QW/97j7VQD+AsCnzOy9qz0goP7Oj/jO2SvJVwFcjvoeDSMAvtiqE5tZN4C7AHzG3SfPjrVyTgLj\naPmc+BKK5jbLaoh/GMCOs36mxT9XGncfbvx/AsA9WN3KRKNmthUAGv+fWI1BuPto48KrAfg6WjQn\nZpZHXXDfcfe7G80tn5PQOFZrThrnPueiuc2yGuJ/HMAVjZXLNgAfAXBvqwdhZmvMrOeVxwA+CGBf\nvNeKci/qhVCBVSyI+orYGnwYLZgTqxekux3Afnf/0lmhls4JG0er56RlRXNbtYI5bzXzQ6ivpL4I\n4J9WaQyXoe40PA3gmVaOA8B3Uf/4WEb9u9vHUd/z8AEABwD8HEDfKo3j2wD2AtiDuvi2tmAc70H9\nI/0eAE81/n2o1XMSGUdL5wTAW1EvirsH9Teafz7rmn0MwEEA/wWgfSnn0V/4CZEoqS/4CZEsEr8Q\niSLxC5EoEr8QiSLxC5EoEr8QiSLxC5EoEr8QifL/3fecMPmUbVoAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sRnx9TtKEsTW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Convert class vectors to binary class matrices.\n",
        "# One-hot 인코딩\n",
        "y_train = keras.utils.to_categorical(y_train, num_classes)  # num_classes = 10\n",
        "y_test = keras.utils.to_categorical(y_test, num_classes)    # num_classes = 10"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0iiQMovZKJTL",
        "colab_type": "text"
      },
      "source": [
        "Keras를 이용하여 모델 구성하기!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "X92ke6ChFFa8",
        "colab_type": "code",
        "outputId": "6d0dd46c-a85a-42a2-a405-806e1762183f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 145
        }
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Conv2D(32, (3, 3), padding='same', input_shape=x_train.shape[1:]))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Conv2D(32, (3, 3)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Dropout(0.25))\n",
        "\n",
        "model.add(Conv2D(64, (3, 3), padding='same'))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Conv2D(64, (3, 3)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Dropout(0.25))\n",
        "\n",
        "model.add(Flatten())\n",
        "model.add(Dense(512))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.5))\n",
        "model.add(Dense(num_classes))\n",
        "model.add(Activation('softmax'))\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Colocations handled automatically by placer.\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uj4kWTU2FH0V",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# initiate RMSprop optimizer\n",
        "opt = keras.optimizers.rmsprop(lr=0.0001, decay=1e-6)\n",
        "\n",
        "# Let's train the model using RMSprop\n",
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer=opt,\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "x_train = x_train.astype('float32')\n",
        "x_test = x_test.astype('float32')\n",
        "x_train /= 255\n",
        "x_test /= 255\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sYD-7MYoFKhd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "if not data_augmentation:\n",
        "    print('Not using data augmentation.')\n",
        "    model.fit(x_train, y_train,\n",
        "              batch_size=batch_size,\n",
        "              epochs=epochs,\n",
        "              validation_data=(x_test, y_test),\n",
        "              shuffle=True)\n",
        "else:\n",
        "    print('Using real-time data augmentation.')\n",
        "    # This will do preprocessing and realtime data augmentation:\n",
        "    datagen = ImageDataGenerator(\n",
        "        featurewise_center=False,  # set input mean to 0 over the dataset\n",
        "        samplewise_center=False,  # set each sample mean to 0\n",
        "        featurewise_std_normalization=False,  # divide inputs by std of the dataset\n",
        "        samplewise_std_normalization=False,  # divide each input by its std\n",
        "        zca_whitening=False,  # apply ZCA whitening\n",
        "        zca_epsilon=1e-06,  # epsilon for ZCA whitening\n",
        "        rotation_range=0,  # randomly rotate images in the range (degrees, 0 to 180)\n",
        "        # randomly shift images horizontally (fraction of total width)\n",
        "        width_shift_range=0.1,\n",
        "        # randomly shift images vertically (fraction of total height)\n",
        "        height_shift_range=0.1,\n",
        "        shear_range=0.,  # set range for random shear\n",
        "        zoom_range=0.,  # set range for random zoom\n",
        "        channel_shift_range=0.,  # set range for random channel shifts\n",
        "        # set mode for filling points outside the input boundaries\n",
        "        fill_mode='nearest',\n",
        "        cval=0.,  # value used for fill_mode = \"constant\"\n",
        "        horizontal_flip=True,  # randomly flip images\n",
        "        vertical_flip=False,  # randomly flip images\n",
        "        # set rescaling factor (applied before any other transformation)\n",
        "        rescale=None,\n",
        "        # set function that will be applied on each input\n",
        "        preprocessing_function=None,\n",
        "        # image data format, either \"channels_first\" or \"channels_last\"\n",
        "        data_format=None,\n",
        "        # fraction of images reserved for validation (strictly between 0 and 1)\n",
        "        validation_split=0.0)\n",
        "\n",
        "    # Compute quantities required for feature-wise normalization\n",
        "    # (std, mean, and principal components if ZCA whitening is applied).\n",
        "    datagen.fit(x_train)\n",
        "\n",
        "    # Fit the model on the batches generated by datagen.flow().\n",
        "    model.fit_generator(datagen.flow(x_train, y_train,\n",
        "                                     batch_size=batch_size),\n",
        "                        epochs=epochs,\n",
        "                        steps_per_epoch = x_train.shape[0] // batch_size,\n",
        "                        validation_data=(x_test, y_test),\n",
        "                        workers=4)\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LKyT2-3lFMb2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Save model and weights\n",
        "if not os.path.isdir(save_dir):\n",
        "    os.makedirs(save_dir)\n",
        "model_path = os.path.join(save_dir, model_name)\n",
        "model.save(model_path)\n",
        "print('Saved trained model at %s ' % model_path)\n",
        "\n",
        "# Score trained model.\n",
        "scores = model.evaluate(x_test, y_test, verbose=1)\n",
        "print('Test loss:', scores[0])\n",
        "print('Test accuracy:', scores[1])"
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}