mnist_dense.py.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "'''Fully connected neural networks with two Dense layers.\n",
    "'''\n",
    "# step 1: Importing packages\n",
    "import keras\n",
    "from keras.datasets import mnist\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout\n",
    "from keras.optimizers import Adam\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.gridspec as gridspec\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# step 2: Loading data and visualizing some examples\n",
    "(x_train, y_train), (x_test, y_test) = mnist.load_data()\n",
    "\n",
    "# training parameters\n",
    "batch_size = 128\n",
    "num_classes = 10\n",
    "epochs = 5\n",
    "\n",
    "# visualization of some examples\n",
    "def show_images(images, rows = 6):\n",
    "    \"\"\"Display a list of images in a single figure with matplotlib.\n",
    "    \n",
    "    Args:\n",
    "        images: List of np.arrays compatible with plt.imshow.\n",
    "        rows (Default = 1): Number of columns in figure (number of rows is \n",
    "                        set to np.ceil(n_images/float(rows))).\n",
    "    \"\"\"\n",
    "    n_images = len(images)\n",
    "    cols = int(n_images/rows)\n",
    "    plt.figure(figsize = (rows, cols))\n",
    "    gs1 = gridspec.GridSpec(rows, cols)\n",
    "    gs1.update(wspace=0.025, hspace=0.025) # set the spacing between axes. \n",
    "\n",
    "    for i in range(n_images):\n",
    "        ax1 = plt.subplot(gs1[i])\n",
    "        plt.axis('on')\n",
    "        ax1.set_xticklabels([])\n",
    "        ax1.set_yticklabels([])\n",
    "        ax1.set_aspect('equal')\n",
    "        plt.gray()\n",
    "        plt.imshow(images[i])\n",
    "\n",
    "    plt.show()\n",
    "\n",
    "show_images(x_train[:48])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# step 3: Reshaping to 2D arrays and normalizing data\n",
    "x_train = x_train.reshape(-1, 784).astype('float32')\n",
    "x_test = x_test.reshape(-1, 784).astype('float32')\n",
    "\n",
    "x_train /= 255\n",
    "x_test /= 255\n",
    "print('number of training examples: ',x_train.shape[0])\n",
    "print('number of testing examples: ',x_test.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# step 4: Making categorical labels (one-hot labels)\n",
    "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
    "y_test = keras.utils.to_categorical(y_test, num_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# step 5: Building model\n",
    "model = Sequential()\n",
    "model.add(Dense(512, activation='relu', input_shape=(784,)))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(num_classes, activation='softmax'))\n",
    "\n",
    "# summary of model's parameters\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# step 6: Compiling model\n",
    "model.compile(loss='categorical_crossentropy', optimizer=Adam(), metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# step 7: Training\n",
    "model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, verbose=1, validation_data=(x_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# step 8: Evaluating\n",
    "score = model.evaluate(x_test, y_test, verbose=0)\n",
    "print('test_loss={:.4f} test_accuracy={:.4f}'.format(score[0], score[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}