{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MNIST.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "metadata": {
        "id": "SUBpfO01XQ6d",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 2484
        },
        "outputId": "86bae33f-7490-4610-8ea1-ea145ab484ff"
      },
      "cell_type": "code",
      "source": [
        "from __future__ import print_function\n",
        "\n",
        "import keras\n",
        "from keras.datasets import mnist\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout, Activation, Flatten\n",
        "from keras.layers import Conv2D, MaxPooling2D\n",
        "from keras.wrappers.scikit_learn import KerasClassifier\n",
        "\n",
        "num_classes = 10\n",
        "\n",
        "# input image dimensions\n",
        "img_rows, img_cols = 28, 28\n",
        "\n",
        "# load training data and do basic data normalization\n",
        "(x_train, y_train), (x_test, y_test) = mnist.load_data()\n",
        "\n",
        "if K.image_data_format() == 'channels_first':\n",
        "    x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)\n",
        "    x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)\n",
        "    input_shape = (1, img_rows, img_cols)\n",
        "else:\n",
        "    x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)\n",
        "    x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)\n",
        "    input_shape = (img_rows, img_cols, 1)\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",
        "# convert class vectors to binary class matrices\n",
        "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
        "y_test = keras.utils.to_categorical(y_test, num_classes)\n",
        "\n",
        "filters = 8\n",
        "kernel_size = 3\n",
        "pool_size = 2\n",
        "dense_layer_sizes = [32]\n",
        "model = Sequential()\n",
        "model.add(Conv2D(filters, kernel_size,\n",
        "                 padding='valid',\n",
        "                 input_shape=input_shape))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Conv2D(filters, kernel_size))\n",
        "model.add(Activation('relu'))\n",
        "model.add(MaxPooling2D(pool_size=pool_size))\n",
        "model.add(Dropout(0.25))\n",
        "model.add(Flatten())\n",
        "for layer_size in dense_layer_sizes:\n",
        "    model.add(Dense(layer_size))\n",
        "    model.add(Activation('relu'))\n",
        "model.add(Dropout(0.5))\n",
        "model.add(Dense(num_classes))\n",
        "model.add(Activation('softmax'))\n",
        "\n",
        "model.compile(loss='categorical_crossentropy',\n",
        "                  optimizer='adadelta',\n",
        "                  metrics=['accuracy'])\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "x_train_cut = x_train[:10000]\n",
        "y_train_cut = y_train[:10000]\n",
        "history = model.fit(x_train_cut, y_train_cut, validation_data = (x_test, y_test), epochs=15, batch_size=16, verbose=1)\n",
        "#plt.plot([1 - i for i in history.history['acc']])\n",
        "plt.plot([1 - i for i in history.history['val_acc']], 'c')\n",
        "\n",
        "x_train_cut = x_train[:30000]\n",
        "y_train_cut = y_train[:30000]\n",
        "history = model.fit(x_train_cut, y_train_cut, validation_data = (x_test, y_test), epochs=15, batch_size=16, verbose=1)\n",
        "#plt.plot([1 - i for i in history.history['acc']])\n",
        "plt.plot([1 - i for i in history.history['val_acc']], 'r')\n",
        "\n",
        "x_train_cut = x_train[:45000]\n",
        "y_train_cut = y_train[:45000]\n",
        "history = model.fit(x_train_cut, y_train_cut, validation_data = (x_test, y_test), epochs=15, batch_size=16, verbose=1)\n",
        "#plt.plot([1 - i for i in history.history['acc']])\n",
        "plt.plot([1 - i for i in history.history['val_acc']], 'g')\n",
        "\n",
        "history = model.fit(x_train, y_train, validation_data = (x_test, y_test), epochs=15, batch_size=16, verbose=1)\n",
        "#plt.plot([1 - i for i in history.history['acc']])\n",
        "plt.plot([1 - i for i in history.history['val_acc']], 'b')\n",
        "\n",
        "plt.title('Error')\n",
        "plt.ylabel('Error')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['10000', '30000', '45000', '60000'], loc='upper right')\n",
        "plt.show()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 10000 samples, validate on 10000 samples\n",
            "Epoch 1/15\n",
            "10000/10000 [==============================] - 11s 1ms/step - loss: 0.9608 - acc: 0.6794 - val_loss: 0.2564 - val_acc: 0.9292\n",
            "Epoch 2/15\n",
            "10000/10000 [==============================] - 9s 939us/step - loss: 0.5767 - acc: 0.8209 - val_loss: 0.2122 - val_acc: 0.9430\n",
            "Epoch 3/15\n",
            "10000/10000 [==============================] - 9s 931us/step - loss: 0.4851 - acc: 0.8497 - val_loss: 0.1764 - val_acc: 0.9489\n",
            "Epoch 4/15\n",
            "10000/10000 [==============================] - 9s 935us/step - loss: 0.4274 - acc: 0.8680 - val_loss: 0.1581 - val_acc: 0.9539\n",
            "Epoch 5/15\n",
            "10000/10000 [==============================] - 9s 918us/step - loss: 0.4014 - acc: 0.8698 - val_loss: 0.1464 - val_acc: 0.9572\n",
            "Epoch 6/15\n",
            "10000/10000 [==============================] - 9s 929us/step - loss: 0.3853 - acc: 0.8771 - val_loss: 0.1369 - val_acc: 0.9587\n",
            "Epoch 7/15\n",
            "10000/10000 [==============================] - 10s 951us/step - loss: 0.3565 - acc: 0.8881 - val_loss: 0.1412 - val_acc: 0.9570\n",
            "Epoch 8/15\n",
            "10000/10000 [==============================] - 10s 956us/step - loss: 0.3517 - acc: 0.8929 - val_loss: 0.1215 - val_acc: 0.9629\n",
            "Epoch 9/15\n",
            "10000/10000 [==============================] - 10s 964us/step - loss: 0.3143 - acc: 0.8995 - val_loss: 0.1207 - val_acc: 0.9636\n",
            "Epoch 10/15\n",
            "10000/10000 [==============================] - 9s 949us/step - loss: 0.3184 - acc: 0.8996 - val_loss: 0.1185 - val_acc: 0.9644\n",
            "Epoch 11/15\n",
            "10000/10000 [==============================] - 10s 956us/step - loss: 0.3089 - acc: 0.9078 - val_loss: 0.1164 - val_acc: 0.9645\n",
            "Epoch 12/15\n",
            "10000/10000 [==============================] - 9s 913us/step - loss: 0.3066 - acc: 0.9049 - val_loss: 0.1132 - val_acc: 0.9660\n",
            "Epoch 13/15\n",
            "10000/10000 [==============================] - 9s 865us/step - loss: 0.2942 - acc: 0.9076 - val_loss: 0.1152 - val_acc: 0.9637\n",
            "Epoch 14/15\n",
            "10000/10000 [==============================] - 8s 843us/step - loss: 0.2878 - acc: 0.9104 - val_loss: 0.1059 - val_acc: 0.9666\n",
            "Epoch 15/15\n",
            "10000/10000 [==============================] - 8s 847us/step - loss: 0.2884 - acc: 0.9109 - val_loss: 0.1021 - val_acc: 0.9683\n",
            "Train on 30000 samples, validate on 10000 samples\n",
            "Epoch 1/15\n",
            "30000/30000 [==============================] - 24s 816us/step - loss: 0.3012 - acc: 0.9058 - val_loss: 0.0921 - val_acc: 0.9719\n",
            "Epoch 2/15\n",
            "30000/30000 [==============================] - 24s 816us/step - loss: 0.2951 - acc: 0.9104 - val_loss: 0.0892 - val_acc: 0.9732\n",
            "Epoch 3/15\n",
            "30000/30000 [==============================] - 25s 825us/step - loss: 0.2871 - acc: 0.9135 - val_loss: 0.0876 - val_acc: 0.9749\n",
            "Epoch 4/15\n",
            "30000/30000 [==============================] - 25s 832us/step - loss: 0.2798 - acc: 0.9137 - val_loss: 0.0829 - val_acc: 0.9756\n",
            "Epoch 5/15\n",
            "30000/30000 [==============================] - 25s 820us/step - loss: 0.2770 - acc: 0.9152 - val_loss: 0.0788 - val_acc: 0.9766\n",
            "Epoch 6/15\n",
            "30000/30000 [==============================] - 22s 742us/step - loss: 0.2696 - acc: 0.9193 - val_loss: 0.0802 - val_acc: 0.9752\n",
            "Epoch 7/15\n",
            "30000/30000 [==============================] - 23s 767us/step - loss: 0.2676 - acc: 0.9201 - val_loss: 0.0816 - val_acc: 0.9764\n",
            "Epoch 8/15\n",
            "30000/30000 [==============================] - 25s 839us/step - loss: 0.2557 - acc: 0.9232 - val_loss: 0.0814 - val_acc: 0.9769\n",
            "Epoch 9/15\n",
            "30000/30000 [==============================] - 25s 818us/step - loss: 0.2586 - acc: 0.9238 - val_loss: 0.0815 - val_acc: 0.9764\n",
            "Epoch 10/15\n",
            "30000/30000 [==============================] - 25s 831us/step - loss: 0.2562 - acc: 0.9233 - val_loss: 0.0726 - val_acc: 0.9792\n",
            "Epoch 11/15\n",
            "30000/30000 [==============================] - 25s 827us/step - loss: 0.2551 - acc: 0.9239 - val_loss: 0.0745 - val_acc: 0.9773\n",
            "Epoch 12/15\n",
            "30000/30000 [==============================] - 23s 757us/step - loss: 0.2550 - acc: 0.9257 - val_loss: 0.0710 - val_acc: 0.9802\n",
            "Epoch 13/15\n",
            "30000/30000 [==============================] - 22s 747us/step - loss: 0.2521 - acc: 0.9264 - val_loss: 0.0711 - val_acc: 0.9789\n",
            "Epoch 14/15\n",
            "30000/30000 [==============================] - 26s 856us/step - loss: 0.2478 - acc: 0.9261 - val_loss: 0.0967 - val_acc: 0.9755\n",
            "Epoch 15/15\n",
            "30000/30000 [==============================] - 24s 806us/step - loss: 0.2561 - acc: 0.9262 - val_loss: 0.0762 - val_acc: 0.9786\n",
            "Train on 45000 samples, validate on 10000 samples\n",
            "Epoch 1/15\n",
            "45000/45000 [==============================] - 36s 791us/step - loss: 0.2576 - acc: 0.9228 - val_loss: 0.0740 - val_acc: 0.9773\n",
            "Epoch 2/15\n",
            "45000/45000 [==============================] - 35s 770us/step - loss: 0.2604 - acc: 0.9261 - val_loss: 0.0759 - val_acc: 0.9782\n",
            "Epoch 3/15\n",
            "45000/45000 [==============================] - 32s 715us/step - loss: 0.2559 - acc: 0.9254 - val_loss: 0.0793 - val_acc: 0.9775\n",
            "Epoch 4/15\n",
            "45000/45000 [==============================] - 36s 807us/step - loss: 0.2555 - acc: 0.9272 - val_loss: 0.0765 - val_acc: 0.9775\n",
            "Epoch 5/15\n",
            "45000/45000 [==============================] - 36s 795us/step - loss: 0.2573 - acc: 0.9254 - val_loss: 0.0743 - val_acc: 0.9782\n",
            "Epoch 6/15\n",
            "45000/45000 [==============================] - 35s 778us/step - loss: 0.2573 - acc: 0.9250 - val_loss: 0.0810 - val_acc: 0.9776\n",
            "Epoch 7/15\n",
            "45000/45000 [==============================] - 32s 707us/step - loss: 0.2542 - acc: 0.9275 - val_loss: 0.0726 - val_acc: 0.9783\n",
            "Epoch 8/15\n",
            "45000/45000 [==============================] - 36s 805us/step - loss: 0.2541 - acc: 0.9277 - val_loss: 0.0733 - val_acc: 0.9788\n",
            "Epoch 9/15\n",
            "45000/45000 [==============================] - 36s 806us/step - loss: 0.2595 - acc: 0.9256 - val_loss: 0.0739 - val_acc: 0.9789\n",
            "Epoch 10/15\n",
            "45000/45000 [==============================] - 35s 787us/step - loss: 0.2560 - acc: 0.9273 - val_loss: 0.0900 - val_acc: 0.9768\n",
            "Epoch 11/15\n",
            "45000/45000 [==============================] - 33s 728us/step - loss: 0.2595 - acc: 0.9241 - val_loss: 0.0725 - val_acc: 0.9784\n",
            "Epoch 12/15\n",
            "45000/45000 [==============================] - 34s 754us/step - loss: 0.2564 - acc: 0.9269 - val_loss: 0.0735 - val_acc: 0.9789\n",
            "Epoch 13/15\n",
            "45000/45000 [==============================] - 36s 806us/step - loss: 0.2573 - acc: 0.9261 - val_loss: 0.0816 - val_acc: 0.9775\n",
            "Epoch 14/15\n",
            "45000/45000 [==============================] - 36s 795us/step - loss: 0.2617 - acc: 0.9253 - val_loss: 0.0758 - val_acc: 0.9788\n",
            "Epoch 15/15\n",
            "45000/45000 [==============================] - 34s 759us/step - loss: 0.2588 - acc: 0.9259 - val_loss: 0.0757 - val_acc: 0.9792\n",
            "Train on 60000 samples, validate on 10000 samples\n",
            "Epoch 1/15\n",
            "60000/60000 [==============================] - 44s 739us/step - loss: 0.2582 - acc: 0.9255 - val_loss: 0.0903 - val_acc: 0.9740\n",
            "Epoch 2/15\n",
            "60000/60000 [==============================] - 49s 825us/step - loss: 0.2614 - acc: 0.9259 - val_loss: 0.0737 - val_acc: 0.9785\n",
            "Epoch 3/15\n",
            "60000/60000 [==============================] - 46s 773us/step - loss: 0.2619 - acc: 0.9262 - val_loss: 0.0843 - val_acc: 0.9783\n",
            "Epoch 4/15\n",
            "60000/60000 [==============================] - 43s 725us/step - loss: 0.2590 - acc: 0.9260 - val_loss: 0.0821 - val_acc: 0.9759\n",
            "Epoch 5/15\n",
            "60000/60000 [==============================] - 49s 810us/step - loss: 0.2621 - acc: 0.9254 - val_loss: 0.0771 - val_acc: 0.9777\n",
            "Epoch 6/15\n",
            "60000/60000 [==============================] - 47s 790us/step - loss: 0.2602 - acc: 0.9255 - val_loss: 0.0763 - val_acc: 0.9779\n",
            "Epoch 7/15\n",
            "60000/60000 [==============================] - 43s 714us/step - loss: 0.2613 - acc: 0.9269 - val_loss: 0.0982 - val_acc: 0.9742\n",
            "Epoch 8/15\n",
            "60000/60000 [==============================] - 49s 823us/step - loss: 0.2662 - acc: 0.9255 - val_loss: 0.0786 - val_acc: 0.9773\n",
            "Epoch 9/15\n",
            "60000/60000 [==============================] - 48s 795us/step - loss: 0.2663 - acc: 0.9239 - val_loss: 0.0832 - val_acc: 0.9772\n",
            "Epoch 10/15\n",
            "60000/60000 [==============================] - 44s 726us/step - loss: 0.2675 - acc: 0.9238 - val_loss: 0.0879 - val_acc: 0.9770\n",
            "Epoch 11/15\n",
            "60000/60000 [==============================] - 46s 772us/step - loss: 0.2639 - acc: 0.9248 - val_loss: 0.0963 - val_acc: 0.9756\n",
            "Epoch 12/15\n",
            "60000/60000 [==============================] - 48s 808us/step - loss: 0.2626 - acc: 0.9270 - val_loss: 0.0802 - val_acc: 0.9777\n",
            "Epoch 13/15\n",
            "60000/60000 [==============================] - 45s 750us/step - loss: 0.2674 - acc: 0.9241 - val_loss: 0.0742 - val_acc: 0.9785\n",
            "Epoch 14/15\n",
            "60000/60000 [==============================] - 45s 746us/step - loss: 0.2654 - acc: 0.9252 - val_loss: 0.0756 - val_acc: 0.9782\n",
            "Epoch 15/15\n",
            "60000/60000 [==============================] - 49s 824us/step - loss: 0.2705 - acc: 0.9242 - val_loss: 0.1262 - val_acc: 0.9733\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFnCAYAAAC/5tBZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nR/+JIfcW0wAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fbb841f8da0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}