notebook/12_keras_visualize_model_training_history.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using matplotlib backend: MacOSX\n",
      "Populating the interactive namespace from numpy and matplotlib\n",
      "CPython 2.7.12\n",
      "IPython 5.1.0\n",
      "\n",
      "scipy 0.18.1\n",
      "sklearn 0.18.1\n",
      "numpy 1.12.0\n",
      "\n",
      "compiler   : GCC 4.2.1 (Based on Apple Inc. build 5658) (LLVM build 2336.11.00)\n",
      "system     : Darwin\n",
      "release    : 16.4.0\n",
      "machine    : x86_64\n",
      "processor  : i386\n",
      "CPU cores  : 8\n",
      "interpreter: 64bit\n"
     ]
    }
   ],
   "source": [
    "%pylab\n",
    "%matplotlib inline\n",
    "%load_ext watermark\n",
    "%watermark -v -m -p scipy,sklearn,numpy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "# visualize training history\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy\n",
    "\n",
    "# fix random seed for reproducibility\n",
    "seed = 7\n",
    "numpy.random.seed(seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# load pima indians dataset\n",
    "dataset = numpy.loadtxt(\"pima-indians-diabetes.csv\", delimiter=\",\")\n",
    "\n",
    "# split into input (X) and output (Y) variables\n",
    "X = dataset[:,0:8]\n",
    "Y = dataset[:,8]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# create model\n",
    "model = Sequential()\n",
    "model.add(Dense(12, input_dim=8, init='uniform', activation='relu'))\n",
    "model.add(Dense(8, init='uniform', activation='relu'))\n",
    "model.add(Dense(1, init='uniform', activation='sigmoid'))\n",
    "\n",
    "# compile model\n",
    "model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
    "\n",
    "# fit the model\n",
    "history = model.fit(X, Y, validation_split=0.33, nb_epoch=150, batch_size=10, verbose=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['acc', 'loss', 'val_acc', 'val_loss']\n"
     ]
    }
   ],
   "source": [
    "# list all data in history\n",
    "print(history.history.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAEZCAYAAAC5AHPcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11bf5cbd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# summarize history for accuracy\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='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAEZCAYAAAC5AHPcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11bf68150>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# summarize history for loss\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 left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<svg height=\"296pt\" viewBox=\"0.00 0.00 300.56 296.00\" width=\"301pt\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g class=\"graph\" id=\"graph0\" transform=\"scale(1 1) rotate(0) translate(4 292)\">\n",
       "<title>G</title>\n",
       "<polygon fill=\"white\" points=\"-4,4 -4,-292 296.556,-292 296.556,4 -4,4\" stroke=\"none\"/>\n",
       "<!-- 4764109264 -->\n",
       "<g class=\"node\" id=\"node1\"><title>4764109264</title>\n",
       "<polygon fill=\"none\" points=\"0,-243.5 0,-287.5 292.556,-287.5 292.556,-243.5 0,-243.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"83.6191\" y=\"-261.3\">dense_input_1: InputLayer</text>\n",
       "<polyline fill=\"none\" points=\"167.238,-243.5 167.238,-287.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"195.073\" y=\"-272.3\">input:</text>\n",
       "<polyline fill=\"none\" points=\"167.238,-265.5 222.907,-265.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"195.073\" y=\"-250.3\">output:</text>\n",
       "<polyline fill=\"none\" points=\"222.907,-243.5 222.907,-287.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"257.731\" y=\"-272.3\">(None, 8)</text>\n",
       "<polyline fill=\"none\" points=\"222.907,-265.5 292.556,-265.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"257.731\" y=\"-250.3\">(None, 8)</text>\n",
       "</g>\n",
       "<!-- 4764109008 -->\n",
       "<g class=\"node\" id=\"node2\"><title>4764109008</title>\n",
       "<polygon fill=\"none\" points=\"27.9932,-162.5 27.9932,-206.5 264.562,-206.5 264.562,-162.5 27.9932,-162.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"80.1191\" y=\"-180.3\">dense_1: Dense</text>\n",
       "<polyline fill=\"none\" points=\"132.245,-162.5 132.245,-206.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"160.08\" y=\"-191.3\">input:</text>\n",
       "<polyline fill=\"none\" points=\"132.245,-184.5 187.914,-184.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"160.08\" y=\"-169.3\">output:</text>\n",
       "<polyline fill=\"none\" points=\"187.914,-162.5 187.914,-206.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"226.238\" y=\"-191.3\">(None, 8)</text>\n",
       "<polyline fill=\"none\" points=\"187.914,-184.5 264.562,-184.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"226.238\" y=\"-169.3\">(None, 12)</text>\n",
       "</g>\n",
       "<!-- 4764109264&#45;&gt;4764109008 -->\n",
       "<g class=\"edge\" id=\"edge1\"><title>4764109264-&gt;4764109008</title>\n",
       "<path d=\"M146.278,-243.329C146.278,-235.183 146.278,-225.699 146.278,-216.797\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"149.778,-216.729 146.278,-206.729 142.778,-216.729 149.778,-216.729\" stroke=\"black\"/>\n",
       "</g>\n",
       "<!-- 4764109136 -->\n",
       "<g class=\"node\" id=\"node3\"><title>4764109136</title>\n",
       "<polygon fill=\"none\" points=\"27.9932,-81.5 27.9932,-125.5 264.562,-125.5 264.562,-81.5 27.9932,-81.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"80.1191\" y=\"-99.3\">dense_2: Dense</text>\n",
       "<polyline fill=\"none\" points=\"132.245,-81.5 132.245,-125.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"160.08\" y=\"-110.3\">input:</text>\n",
       "<polyline fill=\"none\" points=\"132.245,-103.5 187.914,-103.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"160.08\" y=\"-88.3\">output:</text>\n",
       "<polyline fill=\"none\" points=\"187.914,-81.5 187.914,-125.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"226.238\" y=\"-110.3\">(None, 12)</text>\n",
       "<polyline fill=\"none\" points=\"187.914,-103.5 264.562,-103.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"226.238\" y=\"-88.3\">(None, 8)</text>\n",
       "</g>\n",
       "<!-- 4764109008&#45;&gt;4764109136 -->\n",
       "<g class=\"edge\" id=\"edge2\"><title>4764109008-&gt;4764109136</title>\n",
       "<path d=\"M146.278,-162.329C146.278,-154.183 146.278,-144.699 146.278,-135.797\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"149.778,-135.729 146.278,-125.729 142.778,-135.729 149.778,-135.729\" stroke=\"black\"/>\n",
       "</g>\n",
       "<!-- 4764110160 -->\n",
       "<g class=\"node\" id=\"node4\"><title>4764110160</title>\n",
       "<polygon fill=\"none\" points=\"31.4932,-0.5 31.4932,-44.5 261.062,-44.5 261.062,-0.5 31.4932,-0.5\" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"83.6191\" y=\"-18.3\">dense_3: Dense</text>\n",
       "<polyline fill=\"none\" points=\"135.745,-0.5 135.745,-44.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"163.58\" y=\"-29.3\">input:</text>\n",
       "<polyline fill=\"none\" points=\"135.745,-22.5 191.414,-22.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"163.58\" y=\"-7.3\">output:</text>\n",
       "<polyline fill=\"none\" points=\"191.414,-0.5 191.414,-44.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"226.238\" y=\"-29.3\">(None, 8)</text>\n",
       "<polyline fill=\"none\" points=\"191.414,-22.5 261.062,-22.5 \" stroke=\"black\"/>\n",
       "<text font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"226.238\" y=\"-7.3\">(None, 1)</text>\n",
       "</g>\n",
       "<!-- 4764109136&#45;&gt;4764110160 -->\n",
       "<g class=\"edge\" id=\"edge3\"><title>4764109136-&gt;4764110160</title>\n",
       "<path d=\"M146.278,-81.3294C146.278,-73.1826 146.278,-63.6991 146.278,-54.7971\" fill=\"none\" stroke=\"black\"/>\n",
       "<polygon fill=\"black\" points=\"149.778,-54.729 146.278,-44.729 142.778,-54.729 149.778,-54.729\" stroke=\"black\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>"
      ],
      "text/plain": [
       "<IPython.core.display.SVG object>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import SVG\n",
    "from keras.utils.visualize_util import model_to_dot\n",
    "\n",
    "SVG(model_to_dot(model, show_shapes=True).create(prog='dot', format='svg'))"
   ]
  },
  {
   "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": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}