mimic_keras_pyopencl.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Mimicking Keras using PyOpenCL for convolutions etc\n",
    "\n",
    "Same goal as mimic_keras_predict but this time we're going to try to push convolution and pooling onto a GPU/FPGA using PyOpenCL."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "#System imports\n",
    "from __future__ import print_function\n",
    "\n",
    "#External imports \n",
    "from keras import layers, models\n",
    "from keras import backend as K\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "#Local imports \n",
    "from naive_prediction.data_utils import read_keras_model_from_file\n",
    "#from forward_pass import predict_with_keras_model\n",
    "from data import generate_straight_tracks\n",
    "from drawing import draw_2d_event, draw_2d_input_and_pred\n",
    "#import tests\n",
    "\n",
    "#For consistent printing\n",
    "import numpy as np\n",
    "import pyopencl as cl\n",
    "from pyopencl import array\n",
    "np.set_printoptions(suppress=True)\n",
    "\n",
    "#to test performance\n",
    "import time\n",
    "\n",
    "#for colors\n",
    "from termcolor import colored\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "%matplotlib notebook\n",
    "\n",
    "#pyopencl magic\n",
    "%load_ext pyopencl.ipython_ext"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Now we load our trained model and see how Keras predicts with it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "____________________________________________________________________________________________________\n",
      "Layer (type)                     Output Shape          Param #     Connected to                     \n",
      "====================================================================================================\n",
      "input_3 (InputLayer)             (None, 64, 64)        0                                            \n",
      "____________________________________________________________________________________________________\n",
      "reshape_5 (Reshape)              (None, 64, 64, 1)     0           input_3[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_27 (Convolution2D) (None, 64, 64, 8)     80          reshape_5[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_28 (Convolution2D) (None, 64, 64, 8)     584         convolution2d_27[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_7 (MaxPooling2D)    (None, 32, 32, 8)     0           convolution2d_28[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_29 (Convolution2D) (None, 32, 32, 8)     584         maxpooling2d_7[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_8 (MaxPooling2D)    (None, 16, 16, 8)     0           convolution2d_29[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_30 (Convolution2D) (None, 16, 16, 16)    1168        maxpooling2d_8[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_9 (MaxPooling2D)    (None, 8, 8, 16)      0           convolution2d_30[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_31 (Convolution2D) (None, 8, 8, 16)      2320        maxpooling2d_9[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_32 (Convolution2D) (None, 8, 8, 16)      2320        convolution2d_31[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_33 (Convolution2D) (None, 8, 8, 16)      2320        convolution2d_32[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_34 (Convolution2D) (None, 8, 8, 16)      2320        convolution2d_33[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "upsampling2d_7 (UpSampling2D)    (None, 16, 16, 16)    0           convolution2d_34[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_35 (Convolution2D) (None, 16, 16, 16)    2320        upsampling2d_7[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "upsampling2d_8 (UpSampling2D)    (None, 32, 32, 16)    0           convolution2d_35[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_36 (Convolution2D) (None, 32, 32, 8)     1160        upsampling2d_8[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "upsampling2d_9 (UpSampling2D)    (None, 64, 64, 8)     0           convolution2d_36[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_37 (Convolution2D) (None, 64, 64, 8)     584         upsampling2d_9[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_38 (Convolution2D) (None, 64, 64, 8)     584         convolution2d_37[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_39 (Convolution2D) (None, 64, 64, 1)     73          convolution2d_38[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "reshape_6 (Reshape)              (None, 64, 64)        0           convolution2d_39[0][0]           \n",
      "====================================================================================================\n",
      "Total params: 16,417\n",
      "Trainable params: 16,417\n",
      "Non-trainable params: 0\n",
      "____________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#load model, summary\n",
    "model = read_keras_model_from_file('/Users/Thomas/Desktop/opencl-cnn/model1.h5')\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Batch generator function, from heptrkx\n",
    "def batch_generator_convae(num_event, det_shape, num_tracks, binary=True):\n",
    "    \"\"\"\n",
    "    Generator function for batches of toy data for training.\n",
    "    This function generates a fixed number of tracks in a batch.\n",
    "    \"\"\"\n",
    "    while True:\n",
    "        events = sum(generate_straight_tracks(num_event, det_shape)\n",
    "                     for i in range(num_tracks))\n",
    "        if binary:\n",
    "            events[events > 1] = 1\n",
    "        # Input and target are the same here\n",
    "        yield events, events"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        // select the cell after this one\n",
       "        var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n",
       "        IPython.notebook.select(index + 1);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "num_test = 1\n",
    "det_shape = (64, 64)\n",
    "num_tracks = 8\n",
    "test_events, test_target = batch_generator_convae(num_test, det_shape, num_tracks).next()\n",
    "test_pred = model.predict(test_events)\n",
    "draw_2d_input_and_pred(test_events[0], test_pred[0], cmap='gray_r');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## The prediction looks good\n",
    "## Now let's set up PyOpenCL!\n",
    "Let's have a look at what device we want to use. In our case we'll just use a boring CPU, but this code should be runnable on GPUs and FPGAs as well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[<pyopencl.Platform 'Apple' at 0x7fff0000>]\n"
     ]
    }
   ],
   "source": [
    "platforms = cl.get_platforms()\n",
    "print(platforms)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[<pyopencl.Device 'Intel(R) Core(TM) i5 CPU       M 540  @ 2.53GHz' on 'Apple' at 0xffffffff>, <pyopencl.Device 'GeForce GT 330M' on 'Apple' at 0x1022600>]\n"
     ]
    }
   ],
   "source": [
    "devices = platforms[0].get_devices()\n",
    "print(devices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#I'll use my CPU for this example.\n",
    "ctx = cl.Context([devices[0]])\n",
    "queue = cl.CommandQueue(ctx)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### We need to create Kernels which can be run on GPUs/FPGAs\n",
    "Overall very similar to C code, these kernels allow our code to be parallelized. I've created 3 kernels for various activation functions in order to minimize number of conditionals. The kernel code currently contains only the convolution operation and none of the preparation (i.e padding). Will need to look into whether it makes sense to push the preparation onto an FPGA as well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Thomas/anaconda/lib/python2.7/site-packages/pyopencl-2017.1.1-py2.7-macosx-10.6-x86_64.egg/pyopencl/cffi_cl.py:1488: CompilerWarning: Non-empty compiler output encountered. Set the environment variable PYOPENCL_COMPILER_OUTPUT=1 to see more.\n",
      "  \"to see more.\", CompilerWarning)\n"
     ]
    }
   ],
   "source": [
    "%%cl_kernel -o \"-cl-fast-relaxed-math\"\n",
    "\n",
    "__kernel void convolute_2d(__global float *c,\n",
    "__global const float *a, __global float *b, __constant float *d, int filter_size,\n",
    "int output_width_height, int num_filters, int num_input_channels, int padded_matrix_dim)\n",
    "{\n",
    "    size_t j = get_global_id(0); //height\n",
    "    size_t i = get_global_id(1); //width\n",
    "    size_t k = get_global_id(2); //num filters (output depth)\n",
    "    size_t h;\n",
    "    size_t l;\n",
    "    size_t s;\n",
    "    float conv_result = 0;\n",
    "    for(h = 0; h < num_input_channels; h++) {\n",
    "        for(l = 0; l < filter_size; l++) {\n",
    "            if(num_input_channels == 1) {\n",
    "                for(s = 0; s < filter_size; s++) {\n",
    "                    conv_result += b[k + s*num_filters + l*num_filters*filter_size]*a[j+s + (i+l)*padded_matrix_dim];\n",
    "                }\n",
    "            } else {\n",
    "                for(s = 0; s < filter_size; s++) {\n",
    "                    conv_result += b[k + h*num_filters + s*num_filters*num_input_channels + l*num_filters*num_input_channels*filter_size]*a[h + (j+s)*num_input_channels + (i+l)*num_input_channels*padded_matrix_dim];\n",
    "                }\n",
    "            }\n",
    "        }\n",
    "    }\n",
    "    conv_result += d[k];\n",
    "    c[k + j*num_filters + i*num_filters*output_width_height] = conv_result;\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%cl_kernel -o \"-cl-fast-relaxed-math\"\n",
    "\n",
    "__kernel void convolute_2d_relu(__global float *c,\n",
    "__global const float *a, __global float *b, __constant float *d, int filter_size,\n",
    "int output_width_height, int num_filters, int num_input_channels, int padded_matrix_dim)\n",
    "{\n",
    "    size_t j = get_global_id(0); //height\n",
    "    size_t i = get_global_id(1); //width\n",
    "    size_t k = get_global_id(2); //num filters (output depth)\n",
    "    size_t h;\n",
    "    size_t l;\n",
    "    size_t s;\n",
    "    float conv_result = 0;\n",
    "    for(h = 0; h < num_input_channels; h++) {\n",
    "        for(l = 0; l < filter_size; l++) {\n",
    "            if(num_input_channels == 1) {\n",
    "                for(s = 0; s < filter_size; s++) {\n",
    "                    conv_result += b[k + s*num_filters + l*num_filters*filter_size]*a[j+s + (i+l)*padded_matrix_dim];\n",
    "                }\n",
    "            } else {\n",
    "                for(s = 0; s < filter_size; s++) {\n",
    "                    conv_result += b[k + h*num_filters + s*num_filters*num_input_channels + l*num_filters*num_input_channels*filter_size]*a[h + (j+s)*num_input_channels + (i+l)*num_input_channels*padded_matrix_dim];\n",
    "                }\n",
    "            }\n",
    "        }\n",
    "    }\n",
    "    conv_result += d[k];\n",
    "    if (conv_result < 0) {\n",
    "        conv_result = 0;\n",
    "    }\n",
    "    c[k + j*num_filters + i*num_filters*output_width_height] = conv_result;\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%cl_kernel -o \"-cl-fast-relaxed-math\"\n",
    "\n",
    "__kernel void convolute_2d_sigmoidal(__global float *c,\n",
    "__global const float *a, __global float *b, __constant float *d, int filter_size,\n",
    "int output_width_height, int num_filters, int num_input_channels, int padded_matrix_dim)\n",
    "{\n",
    "    size_t j = get_global_id(0); //height\n",
    "    size_t i = get_global_id(1); //width\n",
    "    size_t k = get_global_id(2); //num filters (output depth)\n",
    "    size_t h;\n",
    "    size_t l;\n",
    "    size_t s;\n",
    "    float conv_result = 0;\n",
    "    for(h = 0; h < num_input_channels; h++) {\n",
    "        for(l = 0; l < filter_size; l++) {\n",
    "            if(num_input_channels == 1) {\n",
    "                for(s = 0; s < filter_size; s++) {\n",
    "                    conv_result += b[k + s*num_filters + l*num_filters*filter_size]*a[j+s + (i+l)*padded_matrix_dim];\n",
    "                }\n",
    "            } else {\n",
    "                for(s = 0; s < filter_size; s++) {\n",
    "                    conv_result += b[k + h*num_filters + s*num_filters*num_input_channels + l*num_filters*num_input_channels*filter_size]*a[h + (j+s)*num_input_channels + (i+l)*num_input_channels*padded_matrix_dim];\n",
    "                }\n",
    "            }\n",
    "        }\n",
    "    }\n",
    "    conv_result += d[k];\n",
    "    //S(x) = 1 / 1 + e^-x\n",
    "    conv_result = 1 / (1 + exp(-conv_result));\n",
    "    c[k + j*num_filters + i*num_filters*output_width_height] = conv_result;\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Wrapping code for convolution\n",
    "Padding, turns ndarrays into arrays that can be used in openCL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def zero_pad_matrix(input_matrix, num_zeros):\n",
    "    \"\"\"\n",
    "    Pad the 3d (nxmxz) input matrix with p zeros\n",
    "    Assumes 'reshaped' matrix. Need to look at this more closely.\n",
    "    \"\"\"\n",
    "    num_zeros = int(num_zeros)\n",
    "    return np.pad(input_matrix, ((0, 0), (num_zeros, num_zeros),\n",
    "                                 (num_zeros, num_zeros), (0, 0)), 'constant')\n",
    "\n",
    "def keras_single_layer_output(model, layer_number, tinput):\n",
    "    \"\"\"\n",
    "    To test other functions\n",
    "    \"\"\"\n",
    "    get_layer_output = K.function([model.layers[layer_number-1].input],\n",
    "                                   [model.layers[layer_number].output])\n",
    "    return get_layer_output([tinput][0])\n",
    "\n",
    "\n",
    "def convolution_2d(input_matrix, filter_weights, stride, activation='relu', padding='same'):\n",
    "    weights_matrix = filter_weights[0]\n",
    "    bias_vector = filter_weights[1]\n",
    "\n",
    "    #pad matrix to maintain input dimensions\n",
    "    filter_dim = len(weights_matrix)\n",
    "    if padding == 'same':\n",
    "        num_zeros_padding = (filter_dim - 1) / 2\n",
    "        padded_matrix = zero_pad_matrix(input_matrix, num_zeros_padding)\n",
    "    else:\n",
    "        padded_matrix = input_matrix\n",
    "        \n",
    "    padded_matrix_dim = len(padded_matrix[0][0])\n",
    "    num_filters = len(weights_matrix.T)\n",
    "\n",
    "    try:\n",
    "        num_input_channels = len(input_matrix[0][0][0])\n",
    "    except:\n",
    "        num_input_channels = 1\n",
    "    output_matrix_width_height = padded_matrix_dim - filter_dim + 1\n",
    "    \n",
    "    a = cl.array.to_device(queue, padded_matrix)\n",
    "    b = cl.array.to_device(queue, weights_matrix)\n",
    "    c = cl.array.zeros(queue, shape=(1, output_matrix_width_height,\n",
    "                                     output_matrix_width_height, num_filters),\n",
    "                      dtype=np.float32)\n",
    "    d = cl.array.to_device(queue, bias_vector)\n",
    "    \n",
    "    if activation=='relu':\n",
    "        convolute_2d_relu(queue, (output_matrix_width_height, \n",
    "                             output_matrix_width_height,\n",
    "                             num_filters,), None, c.data, a.data, b.data, d.data, np.int32(filter_dim),\n",
    "                    np.int32(output_matrix_width_height), np.int32(num_filters), np.int32(num_input_channels),\n",
    "                    np.int32(padded_matrix_dim))\n",
    "    elif activation=='sigmoid':\n",
    "        convolute_2d_sigmoidal(queue, (output_matrix_width_height, \n",
    "                             output_matrix_width_height,\n",
    "                             num_filters,), None, c.data, a.data, b.data, d.data, np.int32(filter_dim),\n",
    "                    np.int32(output_matrix_width_height), np.int32(num_filters), np.int32(num_input_channels),\n",
    "                    np.int32(padded_matrix_dim))\n",
    "    else:\n",
    "        convolute_2d(queue, (output_matrix_width_height, \n",
    "                             output_matrix_width_height,\n",
    "                             num_filters,), None, c.data, a.data, b.data, d.data, np.int32(filter_dim),\n",
    "                    np.int32(output_matrix_width_height), np.int32(num_filters), np.int32(num_input_channels),\n",
    "                    np.int32(padded_matrix_dim))\n",
    "    \n",
    "    return c.get()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks good. Now let's test this convolution operation against Keras and make sure we're getting the same result!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def keras_get_layer_output(model, layer, test_input):\n",
    "    \"\"\"\n",
    "    Helper method, gives the output matrix from a Keras layer\n",
    "    \"\"\"\n",
    "    get_layer_output = K.function([model.layers[0].input],\n",
    "                                  [layer.output])\n",
    "    return get_layer_output([test_input])[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "input_layer = model.layers[1]\n",
    "output_layer = model.layers[2]\n",
    "test_input_matrix = keras_get_layer_output(model, input_layer, test_events)\n",
    "test_correct_output = keras_get_layer_output(model, output_layer, test_events)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "our_test_output = convolution_2d(test_input_matrix, output_layer.get_weights(), 1, activation='relu')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def compare_ndarrays(keras_ndarray, comparison_ndarray):\n",
    "    \"\"\"\n",
    "    Returns total difference between two ndarrays.\n",
    "    \"\"\"\n",
    "    if keras_ndarray.shape != comparison_ndarray.shape:\n",
    "        print(\"Unable to compare ndarrays with different shape.\")\n",
    "        print(\"Keras ndarray shape:\", keras_ndarray.shape)\n",
    "        print(\"Your ndarray shape:\", comparison_ndarray.shape)\n",
    "        return -1\n",
    "\n",
    "    if np.array_equal(keras_ndarray, comparison_ndarray):\n",
    "        print(\"Keras ndarray and comparison ndarray are the same.\")\n",
    "        return 0\n",
    "    \n",
    "    if np.allclose(keras_ndarray, comparison_ndarray, atol=0.001):\n",
    "        print(\"Every item in Keras ndarray and comparison ndarray within 0.001 of eachother.\")\n",
    "    print(\"Keras ndarray and comparison ndarray are NOT the same.\")\n",
    "    print(\"A total of\", np.sum(keras_ndarray == comparison_ndarray),\" out of\",\n",
    "                               keras_ndarray.size ,\" elements are the same.\")\n",
    "    return abs(np.sum(keras_ndarray - comparison_ndarray))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Keras ndarray and comparison ndarray are the same.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compare_ndarrays(test_correct_output, our_test_output)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Perfect, we're getting an exact match! Looks like our convolution operation is perfect!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Max pooling\n",
    "Well convolution works, now let's try and do the same for max pooling. Should be straightforward. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%cl_kernel -o \"-cl-fast-relaxed-math\"\n",
    "\n",
    "__kernel void max_pool_2d(__global float *c,\n",
    "__global const float *a, int kernel_size, int stride,\n",
    "int output_width_height, int num_filters, int input_w_h)\n",
    "{\n",
    "    size_t j = get_global_id(0); //height\n",
    "    size_t i = get_global_id(1); //width\n",
    "    size_t k = get_global_id(2); //num filters (output depth)\n",
    "    int l;\n",
    "    int s;\n",
    "    int i_i = i*2;\n",
    "    int j_j = j*2;\n",
    "    float curr_max = 0;\n",
    "    for (l = 0; l < kernel_size; l++) {\n",
    "        for (s = 0; s < kernel_size; s++) {\n",
    "            if (curr_max < a[k + (j_j+s)*num_filters + (i_i+l)*num_filters*input_w_h]) {\n",
    "                curr_max = a[k + (j_j+s)*num_filters + (i_i+l)*num_filters*input_w_h];\n",
    "            }\n",
    "        }\n",
    "    } \n",
    "\n",
    "    c[k + j*num_filters + i*num_filters*output_width_height] = curr_max;\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def maxpooling_2d(input_matrix, kernel_size, stride):\n",
    "    input_matrix_width_height = len(input_matrix[0])\n",
    "    input_matrix_depth = len(input_matrix[0][0][0])\n",
    "    output_matrix_width_height = int(((input_matrix_width_height - kernel_size) / stride) + 1)\n",
    "\n",
    "    c = cl.array.zeros(queue, shape=(1, output_matrix_width_height,\n",
    "                                     output_matrix_width_height, input_matrix_depth),\n",
    "                      dtype=np.float32)\n",
    "    a = cl.array.to_device(queue, input_matrix)\n",
    "    \n",
    "    max_pool_2d(queue, (output_matrix_width_height, \n",
    "                         output_matrix_width_height,\n",
    "                         input_matrix_depth,), None, c.data, a.data, np.int32(kernel_size),\n",
    "                np.int32(stride), np.int32(output_matrix_width_height), np.int32(input_matrix_depth),\n",
    "                np.int32(input_matrix_width_height))\n",
    "    \n",
    "    return c.get()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "input_layer_pool = model.layers[3]\n",
    "output_layer_pool = model.layers[4]\n",
    "test_input_matrix_pool = keras_get_layer_output(model, input_layer_pool, test_events)\n",
    "test_correct_output_pool = keras_get_layer_output(model, output_layer_pool, test_events)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "our_test_output_pool = maxpooling_2d(test_input_matrix_pool, 2, 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Keras ndarray and comparison ndarray are the same.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#now we check to make sure pooling works consistently with keras' pooling\n",
    "compare_ndarrays(test_correct_output_pool, our_test_output_pool)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Great! Max pooling works using our parallelized opencl kernel.\n",
    "### Upsampling, reshaping arrays\n",
    "Does it make sense to upsample on the FPGA? Maybe not for now, values are simply being mapped to one another (assigning one value in the input array to multiple in the output array). Numpy will probably do this more efficiently on a CPU than we will on an FPGA. Will want to look into this however.\n",
    "\n",
    "Sample applies to reshape."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def upsampling_2d(input_matrix, kernel_size):\n",
    "    \"\"\"\n",
    "    Increase width and height of input matrix.\n",
    "    \"\"\"\n",
    "    output_matrix = np.repeat(np.repeat(input_matrix, kernel_size, axis=1), kernel_size, axis=2)\n",
    "    return output_matrix\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 64, 64, 1)\n",
      "(1, 64, 64, 1)\n",
      "True\n"
     ]
    }
   ],
   "source": [
    "def reshape(inputs, target_shape):\n",
    "    return np.reshape(inputs, (-1,) + target_shape)\n",
    "\n",
    "#def reshape_tf(inputs, target_shape):\n",
    "#    return tf.reshape(inputs, (1,) + target_shape).eval()\n",
    "\n",
    "#def reshape_keras(inputs, target_shape):\n",
    "#    return K.reshape(inputs, (1,) + target_shape)\n",
    "\n",
    "print(reshape(test_events, (64, 64, 1)).shape)\n",
    "#print(reshape_tf(test_events, (64, 64, 1)))\n",
    "#print(reshape_keras(test_events, (64, 64, 1)))\n",
    "print(test_input_matrix.shape)\n",
    "print(np.array_equal(test_input_matrix, reshape(test_events, (64, 64, 1))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Making a full prediction using parallelized convolution and pooling\n",
    "Let's time it to see how we compare to Keras! This is still all being done on a CPU."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_layer_type(keras_layer):\n",
    "    \"\"\"\n",
    "    return string describing the type of a layer\n",
    "    \"\"\"\n",
    "    config = keras_layer.get_config()\n",
    "    return config['name'].split('_', 1)[0]\n",
    "\n",
    "def predict_with_keras_model(keras_model, event):\n",
    "    prediction = event\n",
    "    for layer in keras_model.layers:\n",
    "\n",
    "        #get layer information via keras\n",
    "        layer_type = get_layer_type(layer)\n",
    "        layer_config = layer.get_config()\n",
    "\n",
    "        #perform appropriate operation\n",
    "        if layer_type == 'convolution2d':\n",
    "            prediction = convolution_2d(prediction, layer.get_weights(), 1, activation=layer_config['activation'], padding='same')\n",
    "        if layer_type == 'maxpooling2d':\n",
    "            prediction = maxpooling_2d(prediction, layer_config['pool_size'][0], layer_config['strides'][0])\n",
    "        #not parallelized\n",
    "        if layer_type == 'reshape':\n",
    "            if(layer_config['target_shape'] != (64, 64, 1)):\n",
    "                prediction = reshape(prediction, layer_config['target_shape'])\n",
    "        if layer_type == 'upsampling2d':\n",
    "            prediction = upsampling_2d(prediction, layer_config['size'][0])\n",
    "    return prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "A single prediction took 0.227004051208  seconds.\n"
     ]
    }
   ],
   "source": [
    "start = time.time()\n",
    "our_prediction = predict_with_keras_model(model, test_input_matrix)\n",
    "end = time.time()\n",
    "print(\"A single prediction took\", end-start, \" seconds.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Our prediction\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        // select the cell after this one\n",
       "        var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n",
       "        IPython.notebook.select(index + 1);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Keras prediction\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        // select the cell after this one\n",
       "        var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n",
       "        IPython.notebook.select(index + 1);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        // select the cell after this one\n",
       "        var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n",
       "        IPython.notebook.select(index + 1);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Our prediction\")\n",
    "draw_2d_input_and_pred(test_events[0], our_prediction[0], cmap='gray_r');\n",
    "print(\"Keras prediction\")\n",
    "draw_2d_input_and_pred(test_events[0], test_pred[0], cmap='gray_r')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There was some issue with numpy's reshape module so I've been forced to use Keras' reshaped input. \n",
    "Oh well, aside from that it works! Now we've just got to run this on an FPGA!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def full_run_comparison(keras_model, test_input, cheaters_input):\n",
    "    \"\"\"\n",
    "    From start to finish compare the output of a Keras prediction to\n",
    "    a prediction done using forward_pass.py\n",
    "    \"\"\"\n",
    "    previous_layer_output_keras = test_input\n",
    "    previous_layer_output_self = cheaters_input\n",
    "\n",
    "    for layer in keras_model.layers:\n",
    "\n",
    "        #get layer information via keras\n",
    "        layer_type = get_layer_type(layer)\n",
    "        layer_config = layer.get_config()\n",
    "\n",
    "        #perform appropriate operation\n",
    "        if layer_type == 'convolution2d':\n",
    "            previous_layer_output_self = convolution_2d(previous_layer_output_self, layer.get_weights(), 1, activation=layer_config['activation'], padding='same')\n",
    "            self_out_with_keras_input = convolution_2d(previous_layer_output_keras, layer.get_weights(), 1, activation=layer_config['activation'], padding='same')\n",
    "        elif layer_type == 'maxpooling2d':\n",
    "            previous_layer_output_self = maxpooling_2d(previous_layer_output_self, layer_config['pool_size'][0], layer_config['strides'][0])\n",
    "            self_out_with_keras_input = maxpooling_2d(previous_layer_output_keras, layer_config['pool_size'][0], layer_config['strides'][0])\n",
    "        elif layer_type == 'reshape':\n",
    "            if layer_config['target_shape'] != (64, 64, 1):\n",
    "                previous_layer_output_self = reshape(previous_layer_output_self, layer_config['target_shape'])\n",
    "                self_out_with_keras_input = reshape(previous_layer_output_keras, layer_config['target_shape'])\n",
    "            else:\n",
    "                self_out_with_keras_input = previous_layer_output_self\n",
    "        elif layer_type == 'upsampling2d':\n",
    "            previous_layer_output_self = upsampling_2d(previous_layer_output_self, layer_config['size'][0])\n",
    "            self_out_with_keras_input = upsampling_2d(previous_layer_output_keras, layer_config['size'][0])\n",
    "        else:\n",
    "            continue\n",
    "\n",
    "        #perform keras prediction for this layer\n",
    "        previous_layer_output_keras = keras_get_layer_output(keras_model, layer, test_input)\n",
    "\n",
    "        #now let's see how we compare to keras\n",
    "        print(\"\\n===========================================================================\")\n",
    "        print(\"Forward pass on a\", layer_config['name'],\"layer.\")\n",
    "        if np.array_equal(self_out_with_keras_input, previous_layer_output_keras):\n",
    "            print(colored(\"PASS: This layer successfully produces the same results as Keras.\", 'green'))\n",
    "        else:\n",
    "            print(colored(\"FAIL: This layer does no perform in the same way as Keras.\", 'red'))\n",
    "        if previous_layer_output_keras.shape != self_out_with_keras_input.shape:\n",
    "            print(colored(\"CRITICAL ERROR: This layer returns the wrong output shape.\", 'red'))\n",
    "        print(\"Using self output as input\", np.sum(previous_layer_output_self == previous_layer_output_keras),\n",
    "                                                   \" of\", previous_layer_output_keras.size, \" correctly predicted\")\n",
    "        print(\"Using keras output as input\", np.sum(self_out_with_keras_input == previous_layer_output_keras),\n",
    "                                                    \" of\", previous_layer_output_keras.size, \" correctly predicted\")\n",
    "        if np.allclose(previous_layer_output_self, previous_layer_output_keras, atol=0.001):\n",
    "            print(colored(\"SUCCESS: All output values within 0.001 of eachother.\", 'green'))\n",
    "        else:\n",
    "            print(colored(\"FAIL: All output values NOT within 0.001 of eachother.\", 'red'))\n",
    "        print(\"===========================================================================\")\n",
    "    return previous_layer_output_self, previous_layer_output_keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "===========================================================================\n",
      "Forward pass on a reshape_5 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 4096  of 4096  correctly predicted\n",
      "Using keras output as input 4096  of 4096  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_27 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 32768  of 32768  correctly predicted\n",
      "Using keras output as input 32768  of 32768  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_28 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 28106  of 32768  correctly predicted\n",
      "Using keras output as input 28106  of 32768  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a maxpooling2d_7 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 6373  of 8192  correctly predicted\n",
      "Using keras output as input 8192  of 8192  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_29 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 4237  of 8192  correctly predicted\n",
      "Using keras output as input 5175  of 8192  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a maxpooling2d_8 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 848  of 2048  correctly predicted\n",
      "Using keras output as input 2048  of 2048  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_30 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 2180  of 4096  correctly predicted\n",
      "Using keras output as input 2284  of 4096  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a maxpooling2d_9 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 405  of 1024  correctly predicted\n",
      "Using keras output as input 1024  of 1024  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_31 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 294  of 1024  correctly predicted\n",
      "Using keras output as input 330  of 1024  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_32 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 429  of 1024  correctly predicted\n",
      "Using keras output as input 485  of 1024  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_33 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 256  of 1024  correctly predicted\n",
      "Using keras output as input 360  of 1024  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_34 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 282  of 1024  correctly predicted\n",
      "Using keras output as input 387  of 1024  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a upsampling2d_7 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 1128  of 4096  correctly predicted\n",
      "Using keras output as input 4096  of 4096  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_35 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 2181  of 4096  correctly predicted\n",
      "Using keras output as input 2357  of 4096  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a upsampling2d_8 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 8724  of 16384  correctly predicted\n",
      "Using keras output as input 16384  of 16384  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_36 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 2960  of 8192  correctly predicted\n",
      "Using keras output as input 3751  of 8192  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a upsampling2d_9 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 11840  of 32768  correctly predicted\n",
      "Using keras output as input 32768  of 32768  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_37 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 15471  of 32768  correctly predicted\n",
      "Using keras output as input 19313  of 32768  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_38 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 19401  of 32768  correctly predicted\n",
      "Using keras output as input 21883  of 32768  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a convolution2d_39 layer.\n",
      "\u001b[31mFAIL: This layer does no perform in the same way as Keras.\u001b[0m\n",
      "Using self output as input 304  of 4096  correctly predicted\n",
      "Using keras output as input 299  of 4096  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n",
      "\n",
      "===========================================================================\n",
      "Forward pass on a reshape_6 layer.\n",
      "\u001b[32mPASS: This layer successfully produces the same results as Keras.\u001b[0m\n",
      "Using self output as input 304  of 4096  correctly predicted\n",
      "Using keras output as input 4096  of 4096  correctly predicted\n",
      "\u001b[32mSUCCESS: All output values within 0.001 of eachother.\u001b[0m\n",
      "===========================================================================\n"
     ]
    }
   ],
   "source": [
    "#This function tells us layer by layer how our layers are performing in comparison to Keras'.\n",
    "ourout, kerasout = full_run_comparison(model, test_events, test_input_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Keras prediction vs our prediction\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        // select the cell after this one\n",
       "        var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n",
       "        IPython.notebook.select(index + 1);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        // select the cell after this one\n",
       "        var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n",
       "        IPython.notebook.select(index + 1);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Keras prediction vs our prediction\")\n",
    "draw_2d_input_and_pred(kerasout[0], ourout[0], cmap='gray_r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "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": 1
}