deep-digits-local.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (7,7) # Make the figures a bit bigger\n",
    "\n",
    "from keras.datasets import mnist\n",
    "from keras.models import Sequential\n",
    "from keras.layers.core import Dense, Dropout, Activation\n",
    "from keras.layers import Conv2D, MaxPooling2D, Flatten\n",
    "from keras.utils import np_utils\n",
    "from keras.utils import plot_model\n",
    "import matplotlib.image as mpimg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "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"
     ]
    }
   ],
   "source": [
    "(X_train, y_train), (X_test, y_test) = mnist.load_data()\n",
    "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"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2 µs, sys: 1e+03 ns, total: 3 µs\n",
      "Wall time: 6.2 µs\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaoAAAGfCAYAAAAOOJboAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ntP8H+PJ9Fq8XCg0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x132106950>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%time\n",
    "plt.figure()\n",
    "for i in range(25):\n",
    "    plt.subplot(5,5,i+1)\n",
    "    plt.imshow(X_train[i], cmap='gray_r', interpolation='none')\n",
    "    #plt.title(\"Class {}\".format(y_train[i]))\n",
    "    plt.axis(\"off\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set shape (60000, 28, 28, 1)\n",
      "Testing set shape (10000, 28, 28, 1)\n"
     ]
    }
   ],
   "source": [
    "# reshape\n",
    "X_train = X_train.reshape( X_train.shape[0], X_train.shape[1], X_train.shape[2], 1 )\n",
    "X_test = X_test.reshape( X_test.shape[0], X_test.shape[1], X_test.shape[2], 1 )\n",
    "#X_train = X_train.reshape(60000, 784)\n",
    "#X_test = X_test.reshape(10000, 784)\n",
    "\n",
    "# normalize\n",
    "X_train = X_train.astype('float32') / 255.0\n",
    "X_test = X_test.astype('float32') / 255.0\n",
    "\n",
    "print \"Training set shape\", X_train.shape\n",
    "print \"Testing set shape\", X_test.shape\n",
    "\n",
    "# convert labels to categories\n",
    "Y_train = np_utils.to_categorical(y_train, 10)\n",
    "Y_test = np_utils.to_categorical(y_test, 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# build the model\n",
    "model = Sequential()\n",
    "model.add(Conv2D(32,(3,3), activation='relu', padding='same', input_shape=(28,28,1,)))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Conv2D(32,(3,3), activation='relu', padding='same') )\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Flatten())\n",
    "model.add(Dense(128, activation='relu'))\n",
    "model.add(Dense(10, activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_9 (Conv2D)            (None, 28, 28, 32)        320       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_6 (MaxPooling2 (None, 14, 14, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_10 (Conv2D)           (None, 14, 14, 32)        9248      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_7 (MaxPooling2 (None, 7, 7, 32)          0         \n",
      "_________________________________________________________________\n",
      "flatten_5 (Flatten)          (None, 1568)              0         \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 128)               200832    \n",
      "_________________________________________________________________\n",
      "dense_10 (Dense)             (None, 10)                1290      \n",
      "=================================================================\n",
      "Total params: 211,690\n",
      "Trainable params: 211,690\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()\n",
    "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1\n",
      "60000/60000 [==============================] - 42s - loss: 0.0558 - acc: 0.9828    \n",
      "CPU times: user 1min 46s, sys: 18.4 s, total: 2min 4s\n",
      "Wall time: 42.9 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "history = model.fit([X_train,], Y_train, batch_size=100, nb_epoch=1, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss and accuracy: [0.03794078426444903, 0.98670000000000002]\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(X_test, Y_test, verbose=0)\n",
    "print 'Test loss and accuracy:', score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-0.5, 292.5, 838.5, -0.5)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQ8AAAKvCAYAAAB020pNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhgcRKWF4EJEShgcRKWF4EJGS/wOPca28eGR+0QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1308e44d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_model(model, to_file = \"model.png\")\n",
    "modelplot = mpimg.imread(\"model.png\")\n",
    "plt.figure(figsize=(12,12))\n",
    "plt.imshow(modelplot)\n",
    "plt.axis(\"off\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}