MNIST_handwritten_digit_recognition.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MNIST handwritten digit recognition.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyOq5xTBpvheHq+m9bCtocjM",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ShivamSinghal1/My-ML-Projects/blob/master/MNIST_handwritten_digit_recognition.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zL_H394Vrspd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import keras\n",
        "from keras import utils\n",
        "from keras.datasets import mnist\n",
        "from keras.models import Sequential, load_model\n",
        "from keras.layers.core import Dense, Dropout, Activation, Flatten \n",
        "from keras.utils import np_utils\n",
        "from keras.layers import Conv2D, MaxPooling2D"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4jhSoKi1vLRu",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "(X_train, Y_train), (X_test, Y_test) = mnist.load_data()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7I4lqgwRre8Z",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wRBaxl_1vtvt",
        "colab_type": "code",
        "outputId": "15adc5ca-3369-4523-ed7e-49238283fa40",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 334
        }
      },
      "source": [
        "fig = plt.figure()\n",
        "for i in range(9):\n",
        "  plt.subplot(3,3,i+1)\n",
        "  plt.tight_layout()\n",
        "  plt.imshow(X_train[i], cmap='gray', interpolation='none')\n",
        "  plt.title(\"Digit: {}\".format(y_train[i]))\n",
        "  plt.xticks([])\n",
        "  plt.yticks([])"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-70-34d747b4ed02>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m   \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtight_layout\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m   \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mimshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcmap\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'gray'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minterpolation\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'none'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m   \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtitle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Digit: {}\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_train\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m   \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxticks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m   \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0myticks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'y_train' is not defined"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAGYAAABnCAYAAADygzfHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v2u8l5X3v4c9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u3RVCbwCwGSE",
        "colab_type": "code",
        "outputId": "c68d5a6f-78ff-44ce-b81f-eab966fd47db",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        }
      },
      "source": [
        "print(\"X_train shape\", X_train.shape)\n",
        "print(\"y_train shape\", Y_train.shape)\n",
        "print(\"X_test shape\", X_test.shape)\n",
        "print(\"y_test shape\", Y_test.shape)\n",
        "\n",
        "# building the input vector from the 28x28 pixels\n",
        "X_train = X_train.reshape(60000, 784)\n",
        "X_test = X_test.reshape(10000, 784)\n",
        "X_train = X_train.astype('float32')\n",
        "X_test = X_test.astype('float32')\n",
        "\n",
        "# normalizing the data to help with the training\n",
        "X_train /= 255\n",
        "X_test /= 255\n",
        "\n",
        "# print the final input shape ready for training\n",
        "print(\"Train matrix shape\", X_train.shape)\n",
        "print(\"Test matrix shape\", X_test.shape)"
      ],
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "X_train shape (60000, 28, 28)\n",
            "y_train shape (60000,)\n",
            "X_test shape (10000, 28, 28)\n",
            "y_test shape (10000,)\n",
            "Train matrix shape (60000, 784)\n",
            "Test matrix shape (10000, 784)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xz8yVwQYwVZI",
        "colab_type": "code",
        "outputId": "aadb3e10-c241-404d-b32f-48b9a4204f7e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "print(np.unique(Y_train, return_counts=True))"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=uint8), array([5923, 6742, 5958, 6131, 5842, 5421, 5918, 6265, 5851, 5949]))\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-eRVqSnywZLA",
        "colab_type": "code",
        "outputId": "89c63400-413e-4b2d-c7dc-0073d1286c47",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "n_classes = 10\n",
        "print(\"Shape before one-hot encoding: \", Y_train.shape)\n",
        "Y_train = np_utils.to_categorical(Y_train, n_classes)\n",
        "Y_test = np_utils.to_categorical(Y_test, n_classes)\n",
        "print(\"Shape after one-hot encoding: \", Y_train.shape)"
      ],
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Shape before one-hot encoding:  (60000,)\n",
            "Shape after one-hot encoding:  (60000, 10)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vDorszmlwhgz",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        },
        "outputId": "a12d2c9b-bf6d-4b66-f973-4bd0a13ced19"
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Dense(80, input_shape=(784,)))\n",
        "model.add(Activation('relu'))                            \n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Dense(10))\n",
        "model.add(Activation('softmax'))\n",
        "model.summary()"
      ],
      "execution_count": 90,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_18\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_40 (Dense)             (None, 80)                62800     \n",
            "_________________________________________________________________\n",
            "activation_24 (Activation)   (None, 80)                0         \n",
            "_________________________________________________________________\n",
            "dropout_22 (Dropout)         (None, 80)                0         \n",
            "_________________________________________________________________\n",
            "dense_41 (Dense)             (None, 10)                810       \n",
            "_________________________________________________________________\n",
            "activation_25 (Activation)   (None, 10)                0         \n",
            "=================================================================\n",
            "Total params: 63,610\n",
            "Trainable params: 63,610\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PxOnP1JUw4fp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.compile(loss='categorical_crossentropy', metrics=['accuracy'], optimizer='adam')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "58AnpnTqxEX_",
        "colab_type": "code",
        "outputId": "17ed015a-a43d-4ef5-d7cd-2741a70c013d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 714
        }
      },
      "source": [
        "history = model.fit(X_train, Y_train,\n",
        "          batch_size=128, epochs=20, shuffle = True,\n",
        "          validation_split = 0.2)"
      ],
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 48000 samples, validate on 12000 samples\n",
            "Epoch 1/20\n",
            "48000/48000 [==============================] - 2s 49us/step - loss: 0.5143 - acc: 0.8548 - val_loss: 0.2414 - val_acc: 0.9330\n",
            "Epoch 2/20\n",
            "48000/48000 [==============================] - 2s 34us/step - loss: 0.2593 - acc: 0.9258 - val_loss: 0.1820 - val_acc: 0.9490\n",
            "Epoch 3/20\n",
            "48000/48000 [==============================] - 2s 36us/step - loss: 0.2053 - acc: 0.9399 - val_loss: 0.1566 - val_acc: 0.9548\n",
            "Epoch 4/20\n",
            "48000/48000 [==============================] - 2s 37us/step - loss: 0.1746 - acc: 0.9490 - val_loss: 0.1339 - val_acc: 0.9605\n",
            "Epoch 5/20\n",
            "48000/48000 [==============================] - 2s 36us/step - loss: 0.1521 - acc: 0.9558 - val_loss: 0.1231 - val_acc: 0.9639\n",
            "Epoch 6/20\n",
            "48000/48000 [==============================] - 2s 35us/step - loss: 0.1351 - acc: 0.9600 - val_loss: 0.1131 - val_acc: 0.9657\n",
            "Epoch 7/20\n",
            "48000/48000 [==============================] - 2s 35us/step - loss: 0.1201 - acc: 0.9642 - val_loss: 0.1068 - val_acc: 0.9687\n",
            "Epoch 8/20\n",
            "48000/48000 [==============================] - 2s 35us/step - loss: 0.1103 - acc: 0.9671 - val_loss: 0.1003 - val_acc: 0.9702\n",
            "Epoch 9/20\n",
            "48000/48000 [==============================] - 2s 36us/step - loss: 0.1021 - acc: 0.9694 - val_loss: 0.0977 - val_acc: 0.9698\n",
            "Epoch 10/20\n",
            "48000/48000 [==============================] - 2s 36us/step - loss: 0.0942 - acc: 0.9722 - val_loss: 0.0940 - val_acc: 0.9720\n",
            "Epoch 11/20\n",
            "48000/48000 [==============================] - 2s 35us/step - loss: 0.0867 - acc: 0.9735 - val_loss: 0.0935 - val_acc: 0.9717\n",
            "Epoch 12/20\n",
            "48000/48000 [==============================] - 2s 34us/step - loss: 0.0814 - acc: 0.9746 - val_loss: 0.0901 - val_acc: 0.9730\n",
            "Epoch 13/20\n",
            "48000/48000 [==============================] - 2s 35us/step - loss: 0.0761 - acc: 0.9761 - val_loss: 0.0867 - val_acc: 0.9748\n",
            "Epoch 14/20\n",
            "48000/48000 [==============================] - 2s 34us/step - loss: 0.0713 - acc: 0.9774 - val_loss: 0.0895 - val_acc: 0.9739\n",
            "Epoch 15/20\n",
            "48000/48000 [==============================] - 2s 34us/step - loss: 0.0679 - acc: 0.9788 - val_loss: 0.0879 - val_acc: 0.9743\n",
            "Epoch 16/20\n",
            "48000/48000 [==============================] - 2s 35us/step - loss: 0.0642 - acc: 0.9799 - val_loss: 0.0901 - val_acc: 0.9745\n",
            "Epoch 17/20\n",
            "48000/48000 [==============================] - 2s 34us/step - loss: 0.0617 - acc: 0.9805 - val_loss: 0.0845 - val_acc: 0.9742\n",
            "Epoch 18/20\n",
            "48000/48000 [==============================] - 2s 34us/step - loss: 0.0596 - acc: 0.9804 - val_loss: 0.0846 - val_acc: 0.9765\n",
            "Epoch 19/20\n",
            "48000/48000 [==============================] - 2s 34us/step - loss: 0.0545 - acc: 0.9823 - val_loss: 0.0858 - val_acc: 0.9766\n",
            "Epoch 20/20\n",
            "48000/48000 [==============================] - 2s 34us/step - loss: 0.0525 - acc: 0.9831 - val_loss: 0.0869 - val_acc: 0.9761\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wAjYNJGMxitf",
        "colab_type": "code",
        "outputId": "bd334efc-7c15-490c-da58-4d353d1cff23",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        }
      },
      "source": [
        "fig = plt.figure()\n",
        "plt.subplot(2,1,1)\n",
        "plt.plot(history.history['acc'])\n",
        "plt.plot(history.history['val_acc'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'test'], loc='lower right')\n",
        "\n",
        "plt.subplot(2,1,2)\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'test'], loc='upper right')\n",
        "\n",
        "plt.tight_layout()\n"
      ],
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "02S97OBcyxoT",
        "colab_type": "code",
        "outputId": "e6f8cfe9-51f4-476b-cbef-030454c97983",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "print(history.params)"
      ],
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'batch_size': 128, 'epochs': 20, 'steps': None, 'samples': 48000, 'verbose': 1, 'do_validation': True, 'metrics': ['loss', 'acc', 'val_loss', 'val_acc']}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "86ijlzIMy0i5",
        "colab_type": "code",
        "outputId": "742edeb0-743a-441b-aef5-528d506971ed",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "source": [
        "loss_and_metrics = model.evaluate(X_test, Y_test)\n",
        "\n",
        "print(\"Test Loss\", loss_and_metrics[0])\n",
        "print(\"Test Accuracy\", loss_and_metrics[1])"
      ],
      "execution_count": 95,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "10000/10000 [==============================] - 0s 24us/step\n",
            "Test Loss 0.08155111578051001\n",
            "Test Accuracy 0.9774\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6Q14OpuhzEmD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nSpSsV9Zr4Nd",
        "colab_type": "text"
      },
      "source": [
        "CNN Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CrM01aOhr1Ng",
        "colab_type": "code",
        "outputId": "2b678f00-5725-42c5-a3ff-65db57cd05a0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        }
      },
      "source": [
        "(X_train, Y_train), (X_test, Y_test) = mnist.load_data()\n",
        "X_train = X_train.reshape(60000,28,28,1)\n",
        "X_test = X_test.reshape(10000,28,28,1)\n",
        "\n",
        "Y_train = utils.to_categorical(Y_train, 10)\n",
        "Y_test = utils.to_categorical(Y_test, 10)\n",
        "\n",
        "print(X_train.shape)\n",
        "print(X_test.shape)\n",
        "print(Y_train.shape)\n",
        "print(Y_test.shape)"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(60000, 28, 28, 1)\n",
            "(10000, 28, 28, 1)\n",
            "(60000, 10)\n",
            "(10000, 10)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lcPZtIYfsKkt",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 425
        },
        "outputId": "a1019ea3-93fa-423c-f6bd-22bad7730451"
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(28,28,1)))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Conv2D(64, (3, 3), activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(128, activation='relu'))\n",
        "model.add(Dropout(0.5))\n",
        "model.add(Dense(10, activation='softmax'))\n",
        "model.summary()"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_9\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_15 (Conv2D)           (None, 26, 26, 32)        320       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_15 (MaxPooling (None, 13, 13, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_16 (Conv2D)           (None, 11, 11, 64)        18496     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_16 (MaxPooling (None, 5, 5, 64)          0         \n",
            "_________________________________________________________________\n",
            "flatten_8 (Flatten)          (None, 1600)              0         \n",
            "_________________________________________________________________\n",
            "dense_18 (Dense)             (None, 128)               204928    \n",
            "_________________________________________________________________\n",
            "dropout_9 (Dropout)          (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "dense_19 (Dense)             (None, 10)                1290      \n",
            "=================================================================\n",
            "Total params: 225,034\n",
            "Trainable params: 225,034\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xi-4SPX0VmMh",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 714
        },
        "outputId": "53066420-d134-4d0c-c771-3d16454e7253"
      },
      "source": [
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer= \"adam\",\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "history = model.fit(X_train, Y_train,\n",
        "                    batch_size= 128,\n",
        "                    epochs= 20,\n",
        "                    validation_split = 0.2)\n"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 48000 samples, validate on 12000 samples\n",
            "Epoch 1/20\n",
            "48000/48000 [==============================] - 36s 753us/step - loss: 1.3475 - acc: 0.8419 - val_loss: 0.0817 - val_acc: 0.9756\n",
            "Epoch 2/20\n",
            "48000/48000 [==============================] - 35s 739us/step - loss: 0.1326 - acc: 0.9625 - val_loss: 0.0614 - val_acc: 0.9827\n",
            "Epoch 3/20\n",
            "48000/48000 [==============================] - 35s 738us/step - loss: 0.0983 - acc: 0.9719 - val_loss: 0.0583 - val_acc: 0.9843\n",
            "Epoch 4/20\n",
            "48000/48000 [==============================] - 36s 740us/step - loss: 0.0793 - acc: 0.9764 - val_loss: 0.0639 - val_acc: 0.9822\n",
            "Epoch 5/20\n",
            "48000/48000 [==============================] - 35s 733us/step - loss: 0.0645 - acc: 0.9806 - val_loss: 0.0522 - val_acc: 0.9863\n",
            "Epoch 6/20\n",
            "48000/48000 [==============================] - 35s 736us/step - loss: 0.0516 - acc: 0.9846 - val_loss: 0.0479 - val_acc: 0.9882\n",
            "Epoch 7/20\n",
            "48000/48000 [==============================] - 35s 738us/step - loss: 0.0491 - acc: 0.9853 - val_loss: 0.0540 - val_acc: 0.9868\n",
            "Epoch 8/20\n",
            "48000/48000 [==============================] - 36s 740us/step - loss: 0.0428 - acc: 0.9864 - val_loss: 0.0429 - val_acc: 0.9890\n",
            "Epoch 9/20\n",
            "48000/48000 [==============================] - 35s 739us/step - loss: 0.0415 - acc: 0.9876 - val_loss: 0.0533 - val_acc: 0.9877\n",
            "Epoch 10/20\n",
            "48000/48000 [==============================] - 36s 742us/step - loss: 0.0393 - acc: 0.9879 - val_loss: 0.0539 - val_acc: 0.9874\n",
            "Epoch 11/20\n",
            "48000/48000 [==============================] - 36s 743us/step - loss: 0.0386 - acc: 0.9878 - val_loss: 0.0528 - val_acc: 0.9884\n",
            "Epoch 12/20\n",
            "48000/48000 [==============================] - 35s 739us/step - loss: 0.0371 - acc: 0.9885 - val_loss: 0.0457 - val_acc: 0.9892\n",
            "Epoch 13/20\n",
            "48000/48000 [==============================] - 36s 740us/step - loss: 0.0307 - acc: 0.9903 - val_loss: 0.0502 - val_acc: 0.9895\n",
            "Epoch 14/20\n",
            "48000/48000 [==============================] - 36s 748us/step - loss: 0.0325 - acc: 0.9896 - val_loss: 0.0558 - val_acc: 0.9887\n",
            "Epoch 15/20\n",
            "48000/48000 [==============================] - 36s 753us/step - loss: 0.0298 - acc: 0.9904 - val_loss: 0.0540 - val_acc: 0.9888\n",
            "Epoch 16/20\n",
            "48000/48000 [==============================] - 36s 754us/step - loss: 0.0322 - acc: 0.9904 - val_loss: 0.0601 - val_acc: 0.9878\n",
            "Epoch 17/20\n",
            "48000/48000 [==============================] - 36s 748us/step - loss: 0.0273 - acc: 0.9912 - val_loss: 0.0497 - val_acc: 0.9888\n",
            "Epoch 18/20\n",
            "48000/48000 [==============================] - 36s 743us/step - loss: 0.0260 - acc: 0.9918 - val_loss: 0.0531 - val_acc: 0.9894\n",
            "Epoch 19/20\n",
            "48000/48000 [==============================] - 36s 746us/step - loss: 0.0277 - acc: 0.9913 - val_loss: 0.0522 - val_acc: 0.9895\n",
            "Epoch 20/20\n",
            "48000/48000 [==============================] - 36s 740us/step - loss: 0.0256 - acc: 0.9925 - val_loss: 0.0612 - val_acc: 0.9891\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kcbn5wjoW1jv",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "b6272f93-a3f8-4aa9-b7a7-d9acceb6d30c"
      },
      "source": [
        "loss_and_metrics = model.evaluate(X_test, Y_test)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "10000/10000 [==============================] - 3s 280us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "73pqSxeka1q4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "outputId": "be9ab6c2-df95-43bf-e289-67a386219f4a"
      },
      "source": [
        "fig = plt.figure()\n",
        "plt.subplot(2,1,1)\n",
        "plt.plot(history.history['acc'])\n",
        "plt.plot(history.history['val_acc'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'test'], loc='lower right')\n",
        "\n",
        "plt.subplot(2,1,2)\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'test'], loc='upper right')\n",
        "\n",
        "plt.tight_layout()"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkkbJBiVJGiUblCRplGxQkqRR+gbINxpsZn3juAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NK6aEdLwf3oz",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "168bdd67-6ad3-469d-dc29-37cf2238f9c8"
      },
      "source": [
        "print(\"Test Loss\", loss_and_metrics[0])\n",
        "print(\"Test Accuracy\", loss_and_metrics[1])"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Test Loss 0.04276533186764595\n",
            "Test Accuracy 0.989\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "obOq-P5NiMxf",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}