{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The MNIST database has a training set of 60000 examples.\n",
      "The MNIST database has a test set of 10000 examples.\n"
     ]
    }
   ],
   "source": [
    "from keras.datasets import mnist\n",
    "\n",
    "# use Keras to import pre-shuffled MNIST database\n",
    "(X_train, y_train), (X_test, y_test) = mnist.load_data()\n",
    "\n",
    "print(\"The MNIST database has a training set of %d examples.\" % len(X_train))\n",
    "print(\"The MNIST database has a test set of %d examples.\" % len(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHEAAADBCAYAAABIbSwnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4gAAAADUwP8FPgp4o18akEgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x23cabbe5da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import matplotlib.cm as cm\n",
    "import numpy as np\n",
    "\n",
    "# plot first six training images\n",
    "fig = plt.figure(figsize=(20,20))\n",
    "for i in range(6):\n",
    "    ax = fig.add_subplot(1, 6, i+1, xticks=[], yticks=[])\n",
    "    ax.imshow(X_train[i], cmap='gray')\n",
    "    ax.set_title(str(y_train[i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArEAAAKvCAYAAAB9BpfGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQywAAADM+X/gOTI21RArLQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x23ca92a8a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def visualize_input(img, ax):\n",
    "    ax.imshow(img, cmap='gray')\n",
    "    width, height = img.shape\n",
    "    thresh = img.max()/2.5\n",
    "    for x in range(width):\n",
    "        for y in range(height):\n",
    "            ax.annotate(str(round(img[x][y],2)), xy=(y,x),\n",
    "                        horizontalalignment='center',\n",
    "                        verticalalignment='center',\n",
    "                        color='white' if img[x][y]<thresh else 'black')\n",
    "\n",
    "fig = plt.figure(figsize = (12,12)) \n",
    "ax = fig.add_subplot(111)\n",
    "visualize_input(X_train[0], ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rescale [0,255] --> [0,1]\n",
    "X_train = X_train.astype('float32')/255\n",
    "X_test = X_test.astype('float32')/255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Integer-valued labels:\n",
      "[5 0 4 1 9 2 1 3 1 4]\n",
      "One-hot labels:\n",
      "[[ 0.  0.  0.  0.  0.  1.  0.  0.  0.  0.]\n",
      " [ 1.  0.  0.  0.  0.  0.  0.  0.  0.  0.]\n",
      " [ 0.  0.  0.  0.  1.  0.  0.  0.  0.  0.]\n",
      " [ 0.  1.  0.  0.  0.  0.  0.  0.  0.  0.]\n",
      " [ 0.  0.  0.  0.  0.  0.  0.  0.  0.  1.]\n",
      " [ 0.  0.  1.  0.  0.  0.  0.  0.  0.  0.]\n",
      " [ 0.  1.  0.  0.  0.  0.  0.  0.  0.  0.]\n",
      " [ 0.  0.  0.  1.  0.  0.  0.  0.  0.  0.]\n",
      " [ 0.  1.  0.  0.  0.  0.  0.  0.  0.  0.]\n",
      " [ 0.  0.  0.  0.  1.  0.  0.  0.  0.  0.]]\n"
     ]
    }
   ],
   "source": [
    "from keras.utils import np_utils\n",
    "\n",
    "# print first ten (integer-valued) training labels\n",
    "print('Integer-valued labels:')\n",
    "print(y_train[:10])\n",
    "\n",
    "# one-hot encode the labels\n",
    "y_train = np_utils.to_categorical(y_train, 10)\n",
    "y_test = np_utils.to_categorical(y_test, 10)\n",
    "\n",
    "# print first ten (one-hot) training labels\n",
    "print('One-hot labels:')\n",
    "print(y_train[:10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "flatten_2 (Flatten)          (None, 784)               0         \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 512)               401920    \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 512)               262656    \n",
      "_________________________________________________________________\n",
      "dropout_4 (Dropout)          (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_6 (Dense)              (None, 10)                5130      \n",
      "=================================================================\n",
      "Total params: 669,706\n",
      "Trainable params: 669,706\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten\n",
    "\n",
    "# define the model\n",
    "model = Sequential()\n",
    "model.add(Flatten(input_shape=X_train.shape[1:]))\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(10, activation='softmax'))\n",
    "\n",
    "# summarize the model\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# compile the model\n",
    "model.compile(loss='categorical_crossentropy', optimizer='rmsprop', \n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test accuracy: 11.4900%\n"
     ]
    }
   ],
   "source": [
    "# evaluate test accuracy\n",
    "score = model.evaluate(X_test, y_test, verbose=0)\n",
    "accuracy = 100*score[1]\n",
    "\n",
    "# print test accuracy\n",
    "print('Test accuracy: %.4f%%' % accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 48000 samples, validate on 12000 samples\n",
      "Epoch 1/10\n",
      "47744/48000 [============================>.] - ETA: 0s - loss: 0.2740 - acc: 0.9170Epoch 00000: val_loss improved from inf to 0.15338, saving model to mnist.model.best.hdf5\n",
      "48000/48000 [==============================] - 15s - loss: 0.2735 - acc: 0.9171 - val_loss: 0.1534 - val_acc: 0.9532\n",
      "Epoch 2/10\n",
      "47872/48000 [============================>.] - ETA: 0s - loss: 0.1120 - acc: 0.9660Epoch 00001: val_loss improved from 0.15338 to 0.10635, saving model to mnist.model.best.hdf5\n",
      "48000/48000 [==============================] - 15s - loss: 0.1120 - acc: 0.9660 - val_loss: 0.1064 - val_acc: 0.9688\n",
      "Epoch 3/10\n",
      "47744/48000 [============================>.] - ETA: 0s - loss: 0.0790 - acc: 0.9753Epoch 00002: val_loss improved from 0.10635 to 0.09692, saving model to mnist.model.best.hdf5\n",
      "48000/48000 [==============================] - 15s - loss: 0.0788 - acc: 0.9754 - val_loss: 0.0969 - val_acc: 0.9729\n",
      "Epoch 4/10\n",
      "47872/48000 [============================>.] - ETA: 0s - loss: 0.0614 - acc: 0.9807Epoch 00003: val_loss did not improve\n",
      "48000/48000 [==============================] - 15s - loss: 0.0615 - acc: 0.9806 - val_loss: 0.1038 - val_acc: 0.9751\n",
      "Epoch 5/10\n",
      "47744/48000 [============================>.] - ETA: 0s - loss: 0.0506 - acc: 0.9844Epoch 00004: val_loss improved from 0.09692 to 0.09438, saving model to mnist.model.best.hdf5\n",
      "48000/48000 [==============================] - 14s - loss: 0.0505 - acc: 0.9844 - val_loss: 0.0944 - val_acc: 0.9762\n",
      "Epoch 6/10\n",
      "47744/48000 [============================>.] - ETA: 0s - loss: 0.0437 - acc: 0.9865Epoch 00005: val_loss improved from 0.09438 to 0.09354, saving model to mnist.model.best.hdf5\n",
      "48000/48000 [==============================] - 17s - loss: 0.0436 - acc: 0.9865 - val_loss: 0.0935 - val_acc: 0.9783\n",
      "Epoch 7/10\n",
      "47872/48000 [============================>.] - ETA: 0s - loss: 0.0358 - acc: 0.9890Epoch 00006: val_loss did not improve\n",
      "48000/48000 [==============================] - 17s - loss: 0.0358 - acc: 0.9890 - val_loss: 0.0978 - val_acc: 0.9783\n",
      "Epoch 8/10\n",
      "47744/48000 [============================>.] - ETA: 0s - loss: 0.0360 - acc: 0.9891Epoch 00007: val_loss improved from 0.09354 to 0.09142, saving model to mnist.model.best.hdf5\n",
      "48000/48000 [==============================] - 14s - loss: 0.0359 - acc: 0.9891 - val_loss: 0.0914 - val_acc: 0.9802\n",
      "Epoch 9/10\n",
      "47872/48000 [============================>.] - ETA: 0s - loss: 0.0308 - acc: 0.9907Epoch 00008: val_loss did not improve\n",
      "48000/48000 [==============================] - 14s - loss: 0.0309 - acc: 0.9906 - val_loss: 0.1176 - val_acc: 0.9778\n",
      "Epoch 10/10\n",
      "47744/48000 [============================>.] - ETA: 0s - loss: 0.0266 - acc: 0.9920Epoch 00009: val_loss did not improve\n",
      "48000/48000 [==============================] - 12s - loss: 0.0268 - acc: 0.9919 - val_loss: 0.1140 - val_acc: 0.9775\n"
     ]
    }
   ],
   "source": [
    "from keras.callbacks import ModelCheckpoint   \n",
    "\n",
    "# train the model\n",
    "checkpointer = ModelCheckpoint(filepath='mnist.model.best.hdf5', \n",
    "                               verbose=1, save_best_only=True)\n",
    "hist = model.fit(X_train, y_train, batch_size=128, epochs=10,\n",
    "          validation_split=0.2, callbacks=[checkpointer],\n",
    "          verbose=1, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the weights that yielded the best validation accuracy\n",
    "model.load_weights('mnist.model.best.hdf5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test accuracy: 98.3300%\n"
     ]
    }
   ],
   "source": [
    "# evaluate test accuracy\n",
    "score = model.evaluate(X_test, y_test, verbose=0)\n",
    "accuracy = 100*score[1]\n",
    "\n",
    "# print test accuracy\n",
    "print('Test accuracy: %.4f%%' % accuracy)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}