Quantized training bottleneck.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Quantized training bottleneck\n",
    "\n",
    "In order to enable low-bit training, we need to understand which part of the process prevents training. This can be the forward pass, the backwards pass, or both.\n",
    "\n",
    "We swap each part of the implementation individually with a floating point version and find the lowest bitwidth"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<script>\n",
       "code_show=false; \n",
       "function code_toggle() {\n",
       " if (code_show){\n",
       " $('div.input').hide();\n",
       " } else {\n",
       " $('div.input').show();\n",
       " }\n",
       " code_show = !code_show\n",
       "} \n",
       "$( document ).ready(code_toggle);\n",
       "</script>\n",
       "<form action=\"javascript:code_toggle()\"><input type=\"submit\" value=\"Click here to toggle on/off the raw code.\"></form>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from matplotlib import cm\n",
    "import seaborn as sns\n",
    "import itertools\n",
    "import pickle\n",
    "\n",
    "\n",
    "from activations import *\n",
    "import train_integers\n",
    "\n",
    "%matplotlib notebook\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "\n",
    "def save_object(data, filename):\n",
    "    with open(filename, 'wb') as output:\n",
    "        pickle.dump(data, output, pickle.HIGHEST_PROTOCOL)\n",
    "        \n",
    "        \n",
    "def load_object(filename):\n",
    "    try:\n",
    "        with open(filename, 'rb') as input:\n",
    "            return pickle.load(input)\n",
    "    except Exception as err: \n",
    "        print err\n",
    "        return None\n",
    "    \n",
    "\n",
    "# button for hiding code\n",
    "from IPython.display import HTML\n",
    "\n",
    "HTML('''<script>\n",
    "code_show=false; \n",
    "function code_toggle() {\n",
    " if (code_show){\n",
    " $('div.input').hide();\n",
    " } else {\n",
    " $('div.input').show();\n",
    " }\n",
    " code_show = !code_show\n",
    "} \n",
    "$( document ).ready(code_toggle);\n",
    "</script>\n",
    "<form action=\"javascript:code_toggle()\"><input type=\"submit\" value=\"Click here to toggle on/off the raw code.\"></form>''')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Here is a grid search of fraction widths and learning rates when both the forward pass and backwards pass are computed with fixed point values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\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",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\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 backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\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 * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\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'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\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 * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\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\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\"];\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",
       "    var width = fig.canvas.width/mpl.ratio\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 + '\" width=\"' + width + '\">');\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 width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\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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\" width=\"750\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.datasets import load_digits, fetch_mldata\n",
    "\n",
    "mnist = fetch_mldata(\"MNIST original\")\n",
    "inputs = mnist.data / 255.0\n",
    "targets = mnist.target\n",
    "\n",
    "mnist_acc = load_object(\"mnist_accuracies.pickle\")\n",
    "if mnist_acc is not None:\n",
    "    accuracies, learning_rates, fractions = mnist_acc\n",
    "if mnist_acc is None: \n",
    "    accuracies = np.zeros((20 - 5, 24 - 8))\n",
    "    learning_rates = np.zeros((20 - 5, 24 - 8))\n",
    "    fractions = np.zeros((20 - 5, 24 - 8))\n",
    "    for lr in range(5, 20):\n",
    "        for fraction in range(8, 24):\n",
    "            print(\"Fraction {}, LR {}\".format(fraction, lr))\n",
    "            acc = train_integers.test_classification(inputs, targets, fraction=fraction, lr=lr, \n",
    "                      conf=[784, 10], iter=10000, smooth_window=1001, func=relu, \n",
    "                      func_der=relu_derivative, verbosity=\"none\")\n",
    "            accuracies[lr - 5, fraction - 8] = acc[-1]\n",
    "            learning_rates[lr - 5, fraction - 8] = lr\n",
    "            fractions[lr - 5, fraction - 8] = fraction\n",
    "    \n",
    "    save_object((accuracies, learning_rates, fractions), \"mnist_accuracies.pickle\")\n",
    "\n",
    "fig = plt.figure(figsize=(7.5, 5))\n",
    "ax1 = fig.add_subplot(111, projection='3d')\n",
    "surface = ax1.plot_surface(learning_rates, fractions, accuracies, cmap=matplotlib.cm.coolwarm, linewidth=1, antialiased=True)\n",
    "\n",
    "# axis labels\n",
    "ax1.set_xlabel(\"Learning rate (1/2^value)\")\n",
    "ax1.set_ylabel(\"Fraction bitwidth\")\n",
    "ax1.set_zlabel(\"Accuracy\")\n",
    "ax1.set_title(\"MNIST accuracy depending on fraction bitwidth & learning rate\")\n",
    "\n",
    "cbar = fig.colorbar(surface)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# And here is the same grid search when we use floating point values for the backwards pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\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",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\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 backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\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 * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\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'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\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 * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\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\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\"];\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",
       "    var width = fig.canvas.width/mpl.ratio\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 + '\" width=\"' + width + '\">');\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 width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\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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\" width=\"750\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.datasets import load_digits, fetch_mldata\n",
    "\n",
    "mnist = fetch_mldata(\"MNIST original\")\n",
    "inputs = mnist.data / 255.0\n",
    "targets = mnist.target\n",
    "\n",
    "mnist_acc = load_object(\"mnist_bw_float_accuracies.pickle\")\n",
    "if mnist_acc is not None:\n",
    "    accuracies, learning_rates, fractions = mnist_acc\n",
    "if mnist_acc is None: \n",
    "    accuracies = np.zeros((20 + 2, 24 - 1))\n",
    "    learning_rates = np.zeros((20 + 2, 24 - 1))\n",
    "    fractions = np.zeros((20 + 2, 24 - 1))\n",
    "    for lr in range(-2, 20):\n",
    "        for fraction in range(1, 24):\n",
    "            print(\"Fraction {}, LR {}\".format(fraction, lr))\n",
    "            acc = train_integers.test_classification(inputs, targets, fraction=fraction, lr=lr, \n",
    "                      conf=[784, 10], iter=10000, smooth_window=1001, func=relu, \n",
    "                      func_der=relu_derivative, verbosity=\"none\")\n",
    "            accuracies[lr + 2, fraction - 1] = acc[-1]\n",
    "            learning_rates[lr + 2, fraction - 1] = lr\n",
    "            fractions[lr + 2, fraction - 1] = fraction\n",
    "    \n",
    "    save_object((accuracies, learning_rates, fractions), \"mnist_bw_float_accuracies.pickle\")\n",
    "\n",
    "fig = plt.figure(figsize=(7.5, 5))\n",
    "ax1 = fig.add_subplot(111, projection='3d')\n",
    "surface = ax1.plot_surface(learning_rates, fractions, accuracies, cmap=matplotlib.cm.coolwarm, linewidth=1, antialiased=True)\n",
    "\n",
    "# axis labels\n",
    "ax1.set_xlabel(\"Learning rate (1/2^value)\")\n",
    "ax1.set_ylabel(\"Fraction bitwidth\")\n",
    "ax1.set_zlabel(\"Accuracy\")\n",
    "ax1.set_title(\"MNIST accuracy depending on fraction bitwidth & learning rate\")\n",
    "\n",
    "cbar = fig.colorbar(surface)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lets also compare a full floating point implementation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Errno 2] No such file or directory: 'mnist_float_accuracies.pickle'\n",
      "LR -2\n",
      "LR -1\n",
      "LR 0\n",
      "LR 1\n",
      "LR 2\n",
      "LR 3\n",
      "LR 4\n",
      "LR 5\n",
      "LR 6\n",
      "LR 7\n",
      "LR 8\n",
      "LR 9\n",
      "LR 10\n",
      "LR 11\n",
      "LR 12\n",
      "LR 13\n",
      "LR 14\n",
      "LR 15\n",
      "LR 16\n",
      "LR 17\n",
      "LR 18\n",
      "LR 19\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\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",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\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 backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\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 * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\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'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\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 * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\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\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\"];\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",
       "    var width = fig.canvas.width/mpl.ratio\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 + '\" width=\"' + width + '\">');\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 width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\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,iVBORw0KGgoAAAANSUhEUgAAAu4AAAH0CAYAAABiuKiqAAAgAElEQVR4nOzdeZwcVb3//zfB/XpRv3rBK/dnhUBQULy4Il4X3K9eRcQFvIIMuKCAC8K1sgAJa2STLSxhkSHIrhCwsu8JSZgQkpAQQvbJvkz2bWYy0/35/XGqZ2o6PTPdk5muru7X8/GoR6arq6s/3amuftfpU6ckAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADyVS2pNuYaulO1Dn49JmlwsQspsr6SxknaJfd6z4i3nBZTwqnYquTeh0/lsewU9XyN+W6Dg8Nl85FZ9n1dqqj7ZOoodb3l6qyKtwwAla5Kbmdkkj6f4/7DJK0N7w+y7ss87rIO1hv94hus3F8U35U0VdIWSfslrZT0tKT/Du+fEnmujqbB7bxG9JxqEdzLwUxJGyVdIukcSf9RxOc+Ue797Z3jvilKXnD/gNzrObkb6zmU4D5AuQ/EMsseSnDvJelyScsk1UtaIeleSe8sYB2ZOkpdbyU7uHf0OQOQIFVyO6N6SffkuP+08P4GtR/cN0l6Rzvr7Sy4Xx7OmyLpUkkXSrpZ0jy5ECVJX5cLE5npjvAx12fN/1hHLxQ9olrlH9zfJulNRa+keN4u93m6Lqbn/2H4/KfluO8t4VRsVco/uGfX+Cl1f8DLdxscrIND8F617ktzLXsowf3ScB3PSfqV3D55pQoLh5k6St1hcv8Ph8ddSBd19DkDkCBVch/mf0iq08FfDvdLmiMXZnIF93nhv39sZ70dBfc3yf00P66d2o5sZ36574AOkwtTSVCt8g/u5e6Dcp+ny/NY9l964PlL8fNcpfyDe7aeCO75GqziBvcaSa/J7bMyeqmwA91MHcXWE9tyMRVafyl+zgB0QZXch/mHktKSvhW57y2StsuF8lrlDu5DJU2Ua3WPhs3MejsK7u8Pbw8qsOau7IC+IOkZSWskNcp1/7lNuQPyh+W66tTJ/RKxRK4lKepoSQ9J2hCub5XcT8SZlrfByv1lVBXO7x2ZVyv33n5T7iCpQdIfwvvOlzRJrhtRo6TXJf2mndf4LbkuR3sk7Zb0sqT/De+7WlKTpH/L8bj7Je2Ua03qyBlyX9IN4b/fV+6g2yusf1G47GZJwyS9J2u5WrnX/Q1J88NlX5d0Zo7nfrek2+X+3xolLZfkh8+V0VutAfRXcj/bN8q9D58+hNeT3U1hcDjvuHD5nXIHoA/r4F+e3i7pTklb5f5fXpDbdvLt+nCk3Ha2OazzVUnnZS3TW4W97qjMa4lOtVn3nSjpcUk75A7UJffrVrVc62qD3Of/r5Lem+M5OvqsVOV4/uhne4oO7irT0++JInV9UW7b3Sb3mRqug7fjaI2ntfN6qiT9TlJKblvOuCy8/y+ReYfLbSs3Rubl2l4+H76ehvD1XaiD9zu5aqkO78ssm8923J5Zcp/ZQ5GpI9s5kl6R2wdvl/SkpP8va5l89+vVcgcwx0oaJff+jgjvmyL3+T9R0mS57prrJf0pax29dfABWWa9R4fr2yv3vXGLDm6Zf6+kR+W2o52SHpH0nznWmUtVuNyX5H4Z3yL3eZQkL5y3RO692ib3nvTO8fj2PmeS+/6YLmmf3PszUtJHOqkLQAyq1BqwZ8h9MWV8T+6L5gPqOLh/QQe3ukfXmzFYbYN7L7md5BxJ/6+AmrsS3O+U2xH1l/sif1BSs9wOLupjcl9eWyXdEC57o6QFkWU+ILdj3yf3JXGhpGvkvsAyX8qDVVhwXyb35TQkXF/mtc2W+yL9g1z/47Hh4y/Osd60pIVyfVovkvSAWv8/jwsfd0nW4zIHZw/lqDXqG3LbwkK5n8evk/vyeU0HB90H5A4S7g9fy5/lvtBmS3pz1uteIvcFNCRc74Lweb4eWe4dcuFsq9wB1IVyX3ppuTCf0Tt8jXPl3s8/Sfo/uS/StVnPXcjraS+4z5X7peo34Ws2tQ1bkvRUOH+43P/JU3IHKfkE97fLbVMH5ILdbyVNCx/7+y6+7mwfk9u2TC6cn6PW/tCZ17lILpT8JnwNkguc0yRdKemXcv8P++VaYKOtr519Vvood9e3o8LHT1Hb4F6M90Rq/ZwuCNf/W7l9XUru4Dj6GqM1HhW+JyYX+DOvp4+kj4fzvxN57IhwnS9H5mVa7P8nMi97ezlJ7v1eLamfpCvkDp5eVdv9zjlywX5apJZTw/sGK//tuD2/Cpe/MM/lc8nUETVQ7vP9ZFjXVXL/d6vU9sAn3/16tdz7sDz8+0JJ54b3TZHbRtfIbce/kWuMMrVtyOqt3MG9Xm6/8ZCkX0v6e7hctIGll9x5JM2S7pLbf49T674gus5cqtT6WZwitx/3w/t+GK7narnP4vVy+/RatR6AdfY5O1fu/R4drvtPcu/1DtEnHig5VWoN2BfLtQZkWiuelmvtlToO7gqX2xh5bHS9GYN18E+zV4fz9sq1hAyQ9IlOau5KcM/Vst5Pbmf1wci8qXLvwQezlo1+UT8i92Wb62f0zHKDVVhwN7kW93zqHiPXwpbxrrDml3Rwq3m07pnhMlHfV37v5Ty5FtN3ReZ9XW1baCXXCmhqbenP+GaO+bXhvGgL+xHh88yNzLtCbvvom7XOIXJfhJlWuN7h+raqbavo6To4MOX7eqT2g3v2wc6z4XNnfCJc7ras5R7Osc5cfh8u99PIvDfL/T/ukfSv4bzeyv9155J5fHZXmcFqDfTZcm2XZ4fLfyEyL5/PSkef5ylqG9yL9Z5UhcvNUduQ/3/h/NM7qLG9rjK95BoFMqH4sLC+p+W248wJnZfq4Jb57O3lObnAGN1PnRCuJ3u/01lXmc62444MkWvpbpbbl3RFpo4ML1zfgKzlPirXIBCdn+9+vTp8jiE5lp8S3nduZN5b5L7P/h6Z11u5g7vJHaxFzZXbdjLO1MEHl73UeoAQXWcuVeFy03VwS36u9+CzOvg1tfc5e6dcQL8/a/5Rco0Z2fMBxKxKrQH73+R2jD+S+wLcL+kX4XK16ji4fzG8fWmO9WYM1sHBXZJ+IrdDSqn1J7y5cl9EuRxqX71/CWvI1Py9cP6/hbdvb+dxUuuX74gOlpEKD+4rO1mf5ELm++Ral0ytoTPzfnQ2hN+vw+WOjcz7u1xL02E5H+H8u9r/0luktkH3Drmd/b+FtUanPXItehm1ci1d2c/95/D53h/eflWuJSh7fV9V2xDXO7x9d9b63hPO/10XXo/UfnDP7nKROVHviPD2gPB29gFHJtAPVsfGyoWHXlnzMwE5Ez57K7/X3Z7M49sL7l/s5PFvk/v/yKwnE07y/awUEtyL9Z5Uhcv9Kmv+O+X2kfd1UGNHfdxHy3UvkVzXDJPbHqK/Mj0rt81HRbeXw+X2zU/kWP9IFR7cO9uO2/M7udD+KbkuII1yv2RFNUq6tpP1ZOqIPn9a7lfC7M/865LGt7Oe9vbrUmvAzm6Qkdz/3R4dvB96Xm0bEHqr/eCe3QXxDrlW74z75X4lyu6ClAn0VepYVbjczzpZ7s1yXXLeJxfGo40G7X3OMo03X9bB7/dYuV+sAJSQKrUN2KPlWnPOk9vpZlp9atVxcJfatrpnr1fq/GSoI+S+vB4Ll1uu3P2uuxLcPyi3k92u1oODzJTZGZ4S3v5FjsdnHBUu09kIHINVWHCf2M56/kvSBLmuBtl1Z76E/PD2cZ3U9B65n4uvCm+/K7ydK8BGZVpvLshx37NqG3RH5agzOj0fWbZW7heObBeEy342vL2/k3VmDhZ7h7d9HczUei5FIa8n89jBkduDw3lHZS1XFc73wtvD5AJZ9ol6R+RYZy5vyHVxyJbpF5vpLtVb+b3u9mQe315wz+5XLLmubXfI9TPP/v/IbF/5flYKCe7Fek+q1Bpmsq2R+9WrvRo7Cu4D5ALc2+W6a2wI58+V60Ikuf7LQ7MeF91eMucGXaOD/UWFB/fOtuNc3i4XDDMHRm+S++7Yp9Zhhfuq7cFUezJ1ZNyjjj/v0YOafPbrCpdp0sEHfJL7v1ucY361XHeRjN7KHdzr83hNY+W6NWX7WI515lIVLveFHPe9XW5bWCt3wBN9D/4aWa69z9mf1PH7vauT2gAUWZXaBuxz5cLcLLVtKatV58H9S+G8P+RYr1TYKAbV4bJfynFfocH9cLm+1HVyX+Tfk/Q1uYOT6E6zO4P7IOUO7j9X7uCe/d5KrmW8Qa7/4oWSvh3Wnflyzqwj3+AuuRb2JVm1dHYCUiFBd4xcmPtaO9N/RpatVX7BvUGuP2h768wcwPRW7gAqtQ0+3RXcs7fjKrX9fyl2cO/sdbenvccPVvuf1wlyB1RXy7XYfV2t3aEyz1cKwb2r70mVeia4Z1qDvyLpb3LnPEjuIGiy3InxJumsDmru7uDe2XacS+ZXo2hXjH+RO8dhV3j/zXL7grd2sJ5oHRn3yQXQbyr35z2zX8h3vy61nkSayxS5PurZqtV2X9C7gPVmv6buCu65upw9KLefuVXus/R1ufdhq9r+v7f3OesXzj9Hud/v7OUBxKxKbXcI71RrC+ePI8vVqvPgLrkvn41yJ+YcSnC/JFz27Bz3FRrcT9bBLTBSa5/mqvB2d3aV+V24rndnzb9G+Qf3zEmD2T/vXp+1jny7ykitfXw/LfcLydyOF5dUWNeSu+X6p+YznGWt8usqs0iuD3Nneiu/sNZdXWU6Czw91VXmLLVtyeyt4gb3THeTq7LmZ1pYM8+X72flB8o/uBfrPakKl+tKV5lPqv0w9ja5XzIHyQW5zMniP5Db714cPvbfO6i50K4ye9T9wT1zom12P/T3yrVe14XP+3t1LlNHRuY8guM7eVy++3Up/uDeXV1lcgX3nWrbsi657axZbf/f2/uc/Sicn93NCUCJqtLBO4Tz5L5YouGrVvkF90yre2Z8946C+zvUOsJBtswX0Mdz3FdocD8pXP68yLzD5F5P9k6zu05O/Z9w3dGT2P5F7ss63+D+23BZLzLvXXI/r0fXcURYc406PjlVcn0g6+Ra3lM6ePz99uR7Mmfm//+GHOt4k9oeyNSGy+Y6OXVeZF7m14tcJ+++W60t2r2Vf1jrjpNTOws8mQB3qCen/iQy702SXlTuEzGLFdwzvxgMypp/d47ny+ez8t9q/8BzinKfnNrT70lVuFx7J6dG+09n15hpNf+DcntR7pcDU+svUEeGt5fIdRHsrOZCTk7dpNwHT4PV9eD+pnC9G3Tw9TYyv2g1y41m0plMHRnHho99TAfvvw5T65CjhezXqxVvcM+E5kM9OTXX52ib3D4lKrOdVkfmtfc5O0LuAHuKco+2lGsIYQAxqlL7O4SoWuUX3KXWs/Q7C+7vC2/PkgsBF8gFyczwbs+1U0uhwf3Ncl+GdXItRJfI/TKQayiu/5QLAJnhIDPDa82PLHO0XKtfZoi7X4X1v6bWYPpmuZBeJ9eH8DK51tw5yj+4f0iudW6BXEucH76OTN3RdWS6vSyUO3n113JjZT+SY713qfWLNbtlrz3/rbbDJ16r9odPvC9c/yi58HKx3K8Y6+X+7zJq1XY4yD+odTjIaEh/h9x4zk1yJ7f+Wu79rJb70sxsT72Vf1gr5PV0NbhLrUPDRYeDzBzUZgffbJmhDxvlxoW+RK2frVxDHxazq8xUue3/Orlf155T7mEu8/msvF9uW5wlF8LOVmsYnKLcw0H29HtSFS6XGQ7yErnPTUruRPr2hoOU3Gd/h1w4/3n4eo6J3D8kXPdOtf3lIBPms0NYrpo/JhfcV8vtFwYq93CQkmsE2Su3bz1brkugdGjBXXKf5ZRcd5hr5F7rLXL9zReE/y5U5ye5ZuqIynTfmCEXQn8tNxrPUrX+nxayX69WvMH9cLmGlehwkGPVui84Tx2rUvvf04+E671d7vP1sFx/9+yuMh19zv5XrfvDgeF6rgvry/UdDyBGVer+4H6a8gvub5LrT/5cuP4GuS/4uXI75/Yudd6Vk1NPkBuNYI/cjv5+td+/8CNyfZ13yH05vqGD+5N+UG6HuUWtF0AZmlXzJ+SGX2yU+4K9VB1fgCmX78p9GdfLnSj1J7mLMuX6Yv2u3BfdfrkWlBrl7mr06fDxY9t5zvacKReaGuQOQtq7YJHkDnjmhLXslvsiv1FtDxRq1XoBplfD9S5W23Cf8U65A6llcu9nndxrvUytrUS9VVhYy/f1HEpwf4fcdrFNbtt7Tq4LgCn3iZPZjpT7GbxOrQdxVVnL9Fbxg/vRav2M7JQb0jDTBSn7+fL5rPwinJ9pMT4tnD9FuS/A1NPvSVW4XOYCTNvl/v/+poOvOZGrxtPltqkmHbyP+bZaD2yjMmOo5zr3IlfNX5T7jDWq/QswSa4BYKpau0BWh/Mzy3Y1uEvuZMkxcvubBrng10/uACtzcDxWHV9JNVfNkvt8TpcLxnvl9g1D1bYLTb779WrFG9wl9z4/ptYLMD0s6XPhctnnNGSrUvvf0+9W6+dhj9z/x4fC2quzlm3vc6bw7zFhbfVyB0UPy/1yCACIUeZEvnM7W7CH1ar9A5Zylumb+9POFgRQ1s6Q2xf8V9yFAABK11C5lpl/ibmOWpV/cM91km61XEtkrmEWAZSn7H3B4XJ93HfluA8AAH1XrntG5nLxcatV+Qf3QZJekOsm9Vu1jnM/LM6iABTdg3JdZS6R6+I3Q25f0D/OogAApatWru/iCLWOvhGnWpV/cP+63Cgi2+UOmJbLhfmO+vwCKD//K3eS/S658xIWqXU4UAAAAAAAAAAAAKAILlbr8IM1kj7TwbJVah3aMDM1ZC1zmNxwgRvluiJM0MFXSwQAAABQgLPk+pedL+lEuXFfd+jgq79lVMn1SXt/ZDoqaxlfbuzT78mNIfu8pJU6+CqWHTlMbmzkI5iYmJiYmJiYKnw6WgdfPRcVqEZtLxrUS+6Kjv3aWb5KLpS35zC5lvbohT7eJdcqn+viN+05Wge37DMxMTExMTExVep0tFDR3iJ3xbAzsuY/ItdKnktV+JjVcpcRfl7uypoZfeQ2rpOzHjdV0h0F1HaEJFu7dq3t2rWLiYmJiYmJiakip7Vr12aC+xEF5CiUoQ/IbQinZs2/Sa4lPpdTJf1MLph/SdI/5brO/Ed4f+aSxf+e9binJT3VQS1v1cE/CdmuXbsMAACgUu3atYvgDkldC+7Z3iw3FvO14e2uBvfByvGzEMEdAABUMoI7MrrSVSaXZyQ9Ef7d1a4ytLgDAABkIbgjqkbSXZHbvSStU/snp2Y7XNIbar1kfObk1Msiyxyhwk9OPUIEdwAAUOEI7og6Sy5UnyfpBEnD5IaDzAzxOFzSkMjyV0n6hlzL+ifkWtrr5YaSzPDDdZwu6SS5S8oXOhwkwR0AAFQ8gjuyXSI3SkyjXAv8KZH7pkiqjty+LbLsJkkjJX08a32ZCzBtkjsomCDp+AJrIrgDAICKR3BHEhDcAQBAxSO4IwkI7gAAoOIR3JEEBHcAAFDxCO5IAoI7AACoeAR3JAHBHQAAVDyCO5KA4A4AACoewR1JQHAHAAAVj+COJCC4AwCAikdwRxIQ3EtM/YFme2DaCltVtzfuUgAAqBgEdyQBwb3EXD/ydfP8wL5521RLpdJxlwMAQEUguCMJCO4lZFXdXjtuwEjz/MA8P7DRCzfEXRIAABWB4I4kILiXkF8+8rJ5fmAfvmK0eX5g/337NFrdAQAoAoI7koDgXiJmLKszzw+sT/+R9vKqbfaRq8aEre4b4y4NAICyR3BHEhDcS0BzKm3fvG2qeX5gV41YaGZmt4x9g1Z3AACKhOCOJCC4l4DHa1ab5wd20qAxtn1vo5mZ7djXSKs7AABFQnBHEhDcY7ar/oB94ppx5vmBPTR9ZZv7bh7jWt2/dfs0S6dpdQcAoKcQ3JEEBPeY3TDKDf/45Vsm24HmVJv7tu9tbXUf8xqt7gAA9BSCO5KA4B6j2q17re+AUeb5gU1cvCnnMrS6AwDQ8wjuSAKCe4wuHD7HPD+wcx58qd1Qvn1vo514pRseciyt7gAA9AiCO5KA4B6Tmcu3mucHdky/wJZs2t3hsjeNWUyrOwAAPYjgjiQguMegOZW2b90+zTw/sCueW9jp8rS6AwDQswjuSAKCewyenO2Gf/zooDG2LRz+sTOZVvdv30GrOwAA3Y3gjiQguBfZ7voD9slrx5vnB/bAtBV5Py7a6j5uUe4TWQEAQNcQ3JEEBPci+/No13J+2s2TrbEp1fkDIm4cTas7AAA9geCOJCC4F9Gabftahn8c34VW8220ugMA0CMI7kgCgnsR/eZvbvjHnz7Q/vCPncm02P/PnbS6AwDQXQjuSAKCe5G8tKJ1+MfFG7v+fm/b22gnhK3uXWm1BwAAByO4IwkI7kXQnErbt+9wwz8OeHbBIa+PVncAALoXwR1JQHAvgqdeXuOGf7xqjG3d03DI64u2uk94nVZ3AAAOFcEdSUBw72F7GprsU9e54R/vn5r/8I+dGTLKtbp/587ptLoDAHCICO5IAoJ7D8tcOOlLN00qePjHjmzd00CrOwAA3YTgjiQguPegNdv2Wd+BbvjHsa9t7Pb10+oOAED3ILgjCQjuPeiix14xzw/s7GGzeiRYb93TYB++wrW6T1xMqzsAAF1FcEcSENx7yOxV28zzA+vdL7BF63vu/b1h1Ovm+YF99y5a3QEA6CqCO7JdLKlWUoOkGkmfyfNxZ8ttSCOy5leH86PTmAJrIrj3gFQqbd+5c7p5fmD9/vFqjz4Xre4AABw6gjuizpLUKOl8SSdKul/SDklHdvK43pLWSZqm3MF9tKT3R6b3FFgXwb0HPDNnrXl+YB+5aoxt2X3owz925oaRrtX9dFrdAQDoEoI7omokDY3c7iVpvaR+HTzmcEkzJP1cLqTnCu7Z8wpFcO9mexua7NPh8I/3TllelOesi7S6T1q8uSjPCQBAOSG4I+MtkpolnZE1/xFJz3fwuKslPRf+Xa3cwX2npC2Slki6V9J7C6yN4N7Nbhn7hnl+YF+4cZI1NDUX7XlpdQcAoOsI7sj4gNyGcGrW/JvkWuJz+bxcF5n3hberdXBwP1vS6ZJOkjsoeF3SbLmW+va8VW6DzExHi+DebdZu32fHh8M/jl64oajPTas7AABdR3BHRqHB/V8lrZL0rci8anXeLaZP+Dxf7WCZwTr4hFaCeze55PG55vmB/fi+mbG0el+faXUf+iKt7gAAFIDgjoxCu8qcLLfhNEemdDg1Szq2g+eqk3RhB/fT4t5D5tS2Dv+4cN3OWGqo29NgH7rCtfhPeoNWdwAA8kVwR1SNpLsit3vJdYXJdXLq2yR9NGsaIWli+Pdb2nmO/5AL96cXUBd93LtBKpW20+9ywz/+6ZmeHf6xM7S6AwBQOII7os6SG7/9PEknSBomNxzkUeH9wyUN6eDx1WrbVeadkm6W9Fm5ISO/KukVSUvlWtXzRXDvBv94xQ3/eOKVo23z7vpYa9myu7XVfTKt7gAA5IXgjmyXSFotN557jaRTIvdNkQvn7alW2+D+dklj5UaUOSB3Yaf71XogkC+C+yHa19hkn7neDf949+RlcZdjZmbXBYvM8wP7Hq3uAADkheCOJCC4H6Jbxy0xzw/sv/480eoPFG/4x47Q6g4AQGEI7kgCgvshWLdjf8vwjyMXFHf4x87Q6g4AQP4I7kgCgvsh+N0TbvjHH90bz/CPHYm2uk9ZsiXucgAAKGkEdyQBwb2L5tRubxn+ccHaeIZ/7My1/3St7mfcTas7AAAdIbgjCQjuXZBKpe17Q180zw/s8qfnx11OuzbvrqfVHQCAPBDckQQE9y54bu468/zATrhytG3eFe/wj525Jmx1/z6t7gAAtIvgjiQguBdoX2OTnXL9BPP8wIZOKo3hHzuyeXd9ywm0U2l1BwAgJ4I7koDgXqDbxrvhHz83pHSGf+wMre4AAHSM4I4kILgXYMPO/S19xv/56vq4y8lbtNV92lJa3QEAyEZwRxIQ3AvwhyfnmecH9oN7ZiSu5frqF1yr+5kJrB0AgJ5GcEcSENzzNHe1G/7R8wN7de2OuMsp2OZdtLoDANAegjuSgOCeh3Q6bWfc7YZ//ONTpTv8Y2dodQcAIDeCO5KA4J6HEfPc8I8fvmK0bSrx4R87Em11n760Lu5yAAAoGQR3JAHBvRP7G5vt1Bvc8I93TlgadzmHbPALryW2nz4AAD2F4I4kILh34o4JS83zAzv1hgmJGf6xI7S6AwBwMII7koDg3oGNO+vtw1eMNs8P7Pn5yRn+sTODnnet7j+8l1Z3AADMCO5IBoJ7By59al5ZXrho06566xu2ur+4jFZ3AAAI7kgCgns75q/Z0TL847w1yRv+sTO0ugMA0IrgjiQguOeQTqftzHtmmOcHdumT8+Iup0fQ6g4AQCuCO5KA4J7DC/PXtwz/uGHn/rjL6TGZVvcf3TuTVncAQEUjuCMJCO5Z6g802+eGTDTPD+z28ckf/rEjG3e2trrPoNUdAFDBCO5IAoJ7lrsmuuEfP3vDBNvfmPzhHztDqzsAAAR3JAPBPWLTrno74Uo3/OOIeeviLqcoNu6st74Dwlb35bS6AwAqE8EdSUBwj7js6fnm+YGdUWbDP3bmqhELXav7fbS6AwAqE8EdSUBwDy1Yu7Nl+MdXVm+Pu5yiotUdAFDpCO5IAoK7ueEff3ivG/7x90/MjbucWERb3QEAqDQEdyQBwd3Mglc3mOcH9qErRtn6HeU7/GNHNuzc39LqPnP51rjLAQCgqAjuSIKKD+7R4R//Mm5J3OXE6sqw1f3HtLoDACoMwR1JUPHBfeikZWu7A2QAACAASURBVOb5gZ1y/QTb19gUdzmxotUdAFCpCO5IgooO7pt31duJ4fCPz85dG3c5JeGK51yr+1nDaHUHAFQOgjuSoKKD+/8944Z/PH3oi5ZKMQyimdn6HbS6AwAqD8EdSVCxwX3Trnrr3c8N/zintrKGf+wMre4AgEpDcEcSVGxwH7dok3l+YN+8bWrcpZScaKv7rBW0ugMAyh/BHUlQscH9tvFLzPMDu/SpeXGXUpIGPrfAPD+ws4fNirsUAAB6HMEdSVCxwf0Xj7xsnh/Yg9NXxl1KSVq/Y78dN2Akre4AgIpAcEe2iyXVSmqQVCPpM3k+7my5DWlE1vzDJF0jaaOkekkTJPUtsKaKDe6ZsdsJpe2j1R0AUCkI7og6S1KjpPMlnSjpfkk7JB3ZyeN6S1onaZoODu6+pJ2SvifpY5Kel7RS0tsKqKsig/v2vY3m+e7E1F31B+Iup2Sti7S6v8QBDgCgjBHcEVUjaWjkdi9J6yX16+Axh0uaIennkqrVNrgfJtfSfnlk3rvkWvPPLqCuigzuLy6rM88P7As3Toq7lJI34FnX6v6T+2l1BwCUL4I7Mt4iqVnSGVnzH5FrJW/P1ZKeC/+uVtvg3kdu4zo56zFTJd3RwTrfKrdBZqajVYHBfdjU5eb5gf360Tlxl1Lyoq3uNSu3xV0OAAA9guCOjA/IbQinZs2/Sa4lPpfPy3WReV94u1ptg/vnwnX+e9bjnpb0VAe1DA4f12aqtOD+uyfmmucHdueEpXGXkgj9w1b3/32AVncAQHkiuCOj0OD+r5JWSfpWZF61uie40+JuZl+9dYp5fmATF2+Ku5REWLt9X0urO+8ZAKAcEdyRUWhXmZPlNpzmyJQOp2ZJx6rrXWWyVVwf9/2NzXZMeMXUTbvq4y4nMa4c4a6mevzAUTZ1yZa4ywEAoFsR3BFVI+muyO1ecl1hcp2c+jZJH82aRkiaGP79FrWenHpZ5HFHiJNTO/XK6u3m+YF98tpxlk6n4y4nMRqamu3n1bPN8wPrO3CUTX5jc9wlAQDQbQjuiDpLLlSfJ+kEScPkhoM8Krx/uKQhHTy+WrmHg9wh6XRJJ4X3MxxkJx6dVWueH9i5D9XEXUriNDal7Jfhhav6DhhFtxkAQNkguCPbJZJWy43nXiPplMh9U+TCeXuq1f4FmDbJHRRMkHR8gTVVXHDv9w93ouWfRy+Ou5REOtCcsl8/Osc8P7DjBoy0cYsI7wCA5CO4IwkqLrifftd08/zA/vnq+rhLSawDzSm76LFXzPMDO7b/SBu9cGPcJQEAcEgI7kiCigruTc0pO37gKPP8wFbW7Y27nERrak7Zbx93w2r26T/SRi7YEHdJAAB0GcEdSVBRwf2NjbvN8wP7yFVjLJXixNRD1ZxK26VPzmsJ7y/M51cMAEAyEdyRBBUV3P/xylrz/MB+eO+MuEspG82ptF329Hzz/MCO6RfYiHnr4i4JAICCEdyRBBUV3K/55yLz/MAGPf9a3KWUlVQqbf7fX20J73+fszbukgAAKAjBHUlQUcH9rGEzzfMDe2r2mrhLKTupVNr6P+tG7OndL7CnXuY9BgAkB8EdSVAxwT2dTttJg8aY5we2cN3OuMspS+l02q54zl1h1fMDe7xmddwlAQCQF4I7kqBigvuabftaxh5vbErFXU7ZSqfTNuj511rC+6OzauMuCQCAThHckQQVE9xHL9xgnh/Yt26fFncpZS+dTrecT+D5gT0yc1XcJQEA0CGCO5KgYoL7LWPfMM8P7PKn58ddSkVIp9N2w8jXW8L7Q9NXxl0SAADtIrgjCSomuF/w8Gzz/MAefpEAWSzpdNpuHL24Jbw/MG1F3CUBAJATwR1JUDHB/ZTrJ5jnBzZ71ba4S6ko6XTabg1/7fD8wO6dsjzukgAAOAjBHUlQEcF9656GlmEK9zQ0xV1ORbpt/JKW8D500rK4ywEAoA2CO5KgIoL71CVbzPMDO+3myXGXUtHunLC0JbzfPn5p3OUAANCC4I4kqIjgfs/k5eb5gV302Ctxl1Lx7p68rCW83zr2DUun03GXBAAAwR2JUBHB/eLHXjHPD+zuyXTRKAXDpi5vCe83jVlMeAcAxI7gjiSoiOD+5Zsnm+cHNmXJlrhLQejB6StbwvuQUYR3AEC8CO5IgrIP7nsbmqx3PxcQt+xuiLscRFTPWNUS3q8LFhHeAQCxIbgjCco+uL+8apt5fmCfvm583KUgh+GzalvC++AXXiO8AwBiQXBHEpR9cM+06lb9tSbuUtCOx15a3RLerxyxkPAOACg6gjuSoOyD+/89M988P7Cbx7wRdynowFOz17R0aRrw7AJLpQjvAIDiIbgjCco+uH/7jmnm+YGNWrAh7lLQiWfmrG0J7/7fXyW8AwCKhuCOJCjr4N7YlLK+A0aZ5we2euu+uMtBHp6du9aOCcP75U/Pt2bCOwCgCAjuSIKyDu6L1u8yzw/so4PG0G86QZ6fv9769B9pnh/YpU/NI7wDAHocwR1JUNbB/emX15jnB/bj+2bGXQoKFLy6oSW8/+6JudbUnIq7JABAGSO4IwnKOrgPev418/zArn5hUdyloAtGL9xgx4bh/eLHXiG8AwB6DMEdSVDWwf1H9840zw/s73PWxl0KumjsaxvtuAEuvP/mb3PsAOEdANADCO5IgrIN7qlU2j5y1Rjz/MAWbyy/11dJJry+qeUk418Nf9kamwjvAIDuRXBHEpRtcF9Vt9c8P7C+A0fRSlsGJi3ebH0HuvD+8+rZ1tDUHHdJAIAyQnBHEpRtcA9e3WCeH9h37pwedynoJlOWbLHjw/B+/sOzrf4A4R0A0D0I7kiCsg3uN45e3HIhH5SP6Uvr7ENXuPD+s4dqCO8AgG5BcEcSlG1w/9lDNeb5gQ2fuSruUtDNZiyvsw9fMdo8P7BzHnyJ8A4AOGQEdyRB2Qb3T1473jw/sDm12+MuBT3gpRVb7YQrXXj/yf2zbH8j4R0A0HUEdyRBWQb3zbvrzfMD690vsH2NTXGXgx4ye9U2OzEM7+f9tYbRZgAAXUZwR7aLJdVKapBUI+kzHSx7pqQ5knZK2idpvqRzs5apltvAotOYAmsqy+A+6Y3N5vmBfeWWyXGXgh42e9W2lj7vv350DhdpAgB0CcEdUWdJapR0vqQTJd0vaYekI9tZ/jRJ35d0gqRjJf1eUrOkb0aWqZY0WtL7I9N7CqyrLIP70EnLzPMD++3jc+MuBUUwbemWlnHeL31ynqVS6bhLAgAkDMEdUTWShkZu95K0XlK/AtYxV9K1kdvVkkYcYl1lGdx/87c55vmB3TdledyloEjGvrbR+vR3V1gd+NwCS6cJ7wCA/BHckfEWudbyM7LmPyLp+Twef5ikr8p1mfl6ZH61XFeaLZKWSLpX0nsLrK0sg/sXb5pknh/Y9KV1cZeCIhoxb5317heY5wd2/cjXCe8AgLwR3MtDH0kXyLWMD8ia8vUBuQ3h1Kz5N8m1xLfnXZL2SmqS6xd/Qdb9Z0s6XdJJcgcFr0uaLenwDtb5VrkNMjMdrTIL7rvqD5jnu/C2fW9j3OWgyJ6oWd3y/3/7+KVxlwMASAiCe/JdIBea6yS9JmlhZFpQwHq6Gtx7STpO0smSLpNrXT+tg+X7hM/z1Q6WGayDT2gtq+D+0oqt5vmBnXrDhLhLQUwenL6yJbw/MG1F3OUAABKA4J58tSqsZb09h9pVJuNBSWM7WaZO0oUd3F/2Le4PhaHt59Uvx10KYnTnhKUt4f2xl1bHXQ4AoMQR3JNvt1wrdneokXRX5HYvSetU2Mmpf5U0pYP7/0NSWq77TL7Kro/7H5+ab54f2K3jlsRdCmKUTqdtyKjFLeP5Pzd3XdwlAQBKGME9+R6W9MtuWtdZcv3Uz5Mb4nGY3HCQR4X3D5c0JLJ8f7kTUfuEy18m123nF+H975R0s6TPSuot1z3mFUlL5VrV81V2wf2bt001zw9szGsb4y4FMUun03bliIXm+YH16T/SRi9kmwAA5EZwT6aLItOf5EZseVBuHPWLsqZCXSJptdx47jWSToncN0VulJiM6yQtk1QvabukmXLhP+Ptct1mtkg6INet5361Hgjkq6yCe0NTsx0bDgm4dvu+uMtBCUil0i2/wvQdMMqmLNkSd0kAgBJEcE+mtXlOa+IqsJuVVXBfsHaneX5gHxs8lqEA0aKpOdUytv+HrhhlL63YGndJAIASQ3BHEpRVcH9ythsK8Cf3z4q7FJSYxqaUVf21xjw/sI9cNcbmr9kRd0kAgBJCcE++AXJdUrK9Td0z2kwpKKvgnunPfF2wKO5SUILqDzTb2cNmtfwq8/qG8tjuAQCHjuCefClJR+aY/97wvnJQVsH9zHtmmOczggjat7ehyc64+0Xz/MA+ee04W7FlT9wlAQBKAME9+dKS/i3H/C/JjZdeDsomuDen0nbClaPN8wNbuml33OWghO3cd8C+dfs08/zAPnvDBFuzjROZAaDSEdyTq05utJZU5O/MtE3uYkr3xVZd9yqb4L58y56Wkw+bU5yYio5t3dNgX7llsnl+YF+8aZJt3lUfd0kAgBgR3JPr53LjpaclXRrezkznSvpCfKV1u7IJ7s/PX2+eH9j3hr4YdylIiI076+3zN040zw/sa7dOsW17G+MuCQAQE4J78n1V0pvjLqKHlU1wv2HU6+b5gQ14dkHcpSBB1mzbZ6dcP8E8P7Bv3zHNdu4/EHdJAIAYENyT6R0FTOWgbIL7OQ++ZJ4f2GMvrY67FCTMss177BPXjDPPD+zMe2bYvsamuEsCABQZwT2Z0nJ92/OZykFZBPd0Om0fD4PXPMbnRhcsWr/LTho0xjw/sP99YJbVH2iOuyQAQBER3JPpq5HpAkmbJN0s6cxwulnSRknnx1VgNyuL4L5h537z/MD69B9J4EKXvbJ6u50Yjkz08+rZdqA5FXdJAIAiIbgn33hJ5+SYf46kSUWupaeURXAfv2iTeX5gX//LlLhLQcLNXL7Vjh84yjw/sIsfe4URigCgQhDck2+/pL455vcN7ysHZRHc75iw1Dw/sD88OS/uUlAGJr2x2Y4bMNI8P7D/e2a+pQjvAFD2CO7Jt1TSkBzzh4T3lYOyCO6/Gv6yeX5gD0xbEXcpKBOjFmywY/oF5vmBDXr+NUunCe8AUM4I7sn3HUkNkubJXXDpPklzw3nfjbGu7lQWwf2//uzG4p6xvC7uUlBG/j5nrXm+C+83jVkcdzkAgB5EcC8PvSXdJOmFcLoxnFcuEh/cd+470BKudu5jDG50r+Gzalu2r6GTlsVdDgCghxDckQSJD+4zlteZ5wf2X3+eGHcpKFP3TVneEt4ffnFl3OUAAHoAwT2ZTpTUK/J3R1M5SHxwf2DaCvP8wH41/OW4S0EZu3Xckpbw/tTsNXGXAwDoZgT3ZEpLOjLydyr8N511mwswlYg/PDnPPD+wOyYsjbsUlLF0Om3X/HOReX5gx/QL7IX56+MuCQDQjQjuyXSspMMif3c0lYPEB/ev/2WKeX5gE17fFHcpKHPpdNr6/WOBeX5gx/YfaeMXsc0BQLkguCMJEh3c6w80W5/+brztjTvr4y4HFaA5lbbfPzHXPD+wvgNH2YvLGMkIAMoBwT35Vkn6q6TzJHkx19JTEh3c563ZYZ4f2MevGcc42yiapuaU/fIRd+2AD18x2l5etS3ukgAAh4jgnnxVcsF9pVy/9lWSqiWdL+mY2KrqXokO7n97yQ3Vd86DL8VdCipMQ1OznfPgS+b5gX30qjG2cN3OuEsCABwCgnt5+f8knSPpcUlNkprjLafbJDq493/W9Te+YdTrcZeCCrS/sdl+dO9M8/zATr56rC3ZtDvukgAAXURwLw9vk/QVSddImi6pXtJCSXfFWVQ3SnRw/97QF83zA3ueET4Qk931B+y7d003zw/s09eNt9qte+MuCQDQBQT35Jsmab+kBZLulPQDSe+LtaLul9jg3pxK24euGGWeH9jyLXviLgcVbPveRvvGX6aa5wf2uSETbf2O/XGXBAAoEME9+XZKqpM0XNIFkvrEW06PSGxwX7ppt3l+YCdcOdpSKU5MRbw27663026ebJ4f2Gk3T7bNuxnlCACShOCefL0kfULSHyU9L2m7pNVyQf78GOvqTokN7s/NXWeeH9iZ98yIuxTAzMzW7dhvnxsy0Tw/sG/eNtV27GuMuyQAQJ4I7uXn43KjyjSJK6fG7rrAXcXyyhEL4y4FaLGqbq996rrx5vmBff/uF62xKRV3SQCAPBDck+9jkn4n6VlJW+VGkon2dy8HiQ3uP7l/lnl+YE/OXh13KUAbSzbttpMGjWHEIwBIEIJ78qUkzZN0u6TvS/p/8ZbTIxIZ3NPptH1s8Fjz/IDxs1GSRi/cYJ4fmOcHNnXJlrjLAQB0guCefO+Ju4AiSGRwX7t9n3l+YMf2H2kNTc1xlwPkNCC8zsAnrx1vW3Y3xF0OAKADBPfkW6rcrezvDu8rB4kM7mNe29hyAiBQquoPNLcME3nuQzWMfgQAJYzgnnxpSUfmmH+UpMYi19JTEhncbx23xDw/sD8+NT/uUoAOLdm0244f6K43MGzq8rjLAQC0g+CeXN8Op7Skn0Ruf1vSdyXdIWlJbNV1r0QG959XzzbPD+yh6SvjLgXo1N9eqm3p2jV/zY64ywEA5EBwT650OKUif2emJkkrJJ3ehfVeLKlWUoOkGkmf6WDZMyXNkbsI1D5J8yWdm7XMYZKukbRRUr2kCZL6FlhTIoP7qTdMMM8P7KUVW+MuBehUOp22Xz86xzw/sC/eNMl21x+IuyQAQBaCe3IdLulNktZKen94OzN11Vly3WvOl3SipPsl7VDurjiSdJrcSDYnSDpW0u/lhqP8ZmQZXy7Yf09u6MrnJa2U9LYC6kpccN++t7FltA4CEJJi574DLRdn+v0Tcy2dpr87AJQSgjuiaiQNjdzuJWm9pH4FrGOupGvDvw+Ta2m/PHL/u+Ra888uYJ2JC+7Tl9a1tFwCSTKndpv16T/SPD+wZ+asjbscAEAEwT2ZLlJri/VFnUz5eotca/kZWfMfkWsl78xhkr4q12Xm6+G8PnIb18lZy06V64PfnrfKbZCZ6WglLLjfN2W5eX5gv/nbnLhLAQp254Sl5vmBnXDlaFuxZU/c5QAAQgT3ZFor6b2Rv9ub1hSwzg/IbQinZs2/Sa4lvj3vkrRXrl99g6QLIvd9Llznv2c95mlJT3WwzsHh49pMSQruv318rnl+YEMnLYu7FKBgzam0nTVspnl+YN++YxrXIQCAEkFwR0ZXg3svScfJtapfJtef/bTwvq4G98S3uH/llsnm+YFNemNz3KUAXbJxZ72dfLW78u/VLyyKuxwAgBHc0epQu8pkPChpbPh3V7vKZEtUH/d9jU3Wu587MXXz7vq4ywG6bMLrm1pOsp64eFPc5QBAxSO4l4fz5IZi3B9O8yVVdWE9NZLuitzuJWmdCjs59a+SpoR/Z05OvSxy/xEq85NT59RuN88P7FPXjY+7FOCQDX7hNfP8wD5+zTjbtIsDUQCIE8E9+QbJnRB6s9y46meGf+8N7yvEWXKh+jy5IR6HyQ0HeVR4/3BJQyLL95c7EbVPuPxlcn3dfxFZxg/XcbqkkySNUJkPBzl85irz/MDO+2tN3KUAh6yhqdm+dfs08/zAzh42y5pTDBEJAHEhuCffFkk/zTH/p+F9hbpE0mq58dxrJJ0SuW+KpOrI7eskLZO7sNJ2STPlwn9U5gJMm+QOCiZIOr7AmhIV3P2/v2qeH9iNoxfHXQrQLZZv2WMnXDmaE64BIGYE9+TbqdxXIj1e0q4i19JTEhXcv3PndPP8wIJXN8RdCtBtnn55jXl+YH36j7Q5tdviLgcAKhLBPfnulusak+1GSfcVuZaekpjgfqA5ZX0HjDLPD2xV3d64ywG6TTqdtt894YY5/dyQibZzP1cEBoBiI7gn002R6XZJuyW9KhfU75M7OXWXpNviKrCbJSa4L964yzw/sI9cNcZS9AVGmdldf8C+cOOklouLpdNs4wBQTAT3ZJqe5zQtrgK7WWKC+9/nrDXPD+xH986MuxSgR8xbs8OO7T/SPD+wx2tWx10OAFQUgjuSIDHB/eoXFpnnBzbo+dfiLgXoMfdNWW6eH9iHrhhlSzftjrscAKgYBHckQWKC+4/vc5eJf/rlNXGXAvSYVCpt5zz4knl+YN/4y1SrP9Acd0kAUBEI7kiCRAT3dDptHx00xjw/sEXrS7tW4FBt3l1vn7x2nHl+YAOfWxB3OQBQEQjuSIJEBPfVW/eZ5wfWd8AoO9CcirscoMdNXbLFPD8wzw9s9MKNcZcDAGWP4I4kSERwH7Vgg3l+YP9z57S4SwGK5oaRr5vnB/axwWNt3Y79cZcDAGWN4I4kSERwv3nMG+b5gf3pmVfjLgUomsamlJ1+l7vo2A/vnWFN/NoEAD2G4F4e+ki6QFI/SQOypnKQiOBe9dca8/zAHpm5Ku5SgKKq3brXPnKVO7/j1nFL4i4HAMoWwT35LpDUJKlO0muSFkamBTHW1Z0SEdw/fd148/zAXl7F5eBReUbMW2eeH9gx/QKbtWJr3OUAQFkiuCdfrcqnZb09JR/ct+xuMM8PrHe/wPY2NMVdDhCLy56eb54f2CnXT7DtexvjLgcAyg7BPfl2y3WVKWclH9ynhKNrfPnmyXGXAsRmb0OTffmWyeb5gf28+mVLp9NxlwQAZYXgnnwPS/pl3EX0sJIP7ndPXmaeH9jFj70SdylArBau22l9B4wyzw+sesaquMsBgLJCcE++P0naIulBSb+XdFHWVA5KPrhf9Ngr5vmB3TN5edylALF7aPpKd02DgaO4GBkAdCOCe/Kt7WBaE2Nd3ankg/tpN7vuAVOXbIm7FCB26XTaLnh4tnl+YF+5ZbLta+S8DwDoDgR3JEFJB/c9DU0tV4/cuqch7nKAkrBtb6N95vrxXNsAALoRwR1JUNLBffaqbS0jaQBoNWN5nfXu5w5qX5i/Pu5yACDxCO7l4X8lzZO0T9J+SXMl/STWirpXSQf3h190/XkveHh23KUAJSdzReGPXjXG1mzbF3c5AJBoBPfk+4NcWL9V0pnh9Be5EP+7GOvqTiUd3C8Px66+dewbcZcClJym5pSdec8M8/zAvjf0RTvQnIq7JABILIJ78q2SVJVj/vmSVha3lB5T0sH9W7dPM88PbPTCjXGXApSktdv32UmDxpjnB/bn0YvjLgcAEovgnnyNko7LMb+vpIYi19JTSja4Nzal7LgBI83zA7oBAB0YuWBDy9WFpy+ti7scAEgkgnvyLZLUL8f8/pJeK3ItPaVkg/vCdTvN8wM7adAYrhIJdKLfPxaY5wf2qevGWx0jMAFAwQjuyfcjSc2SArmw3j/8u0nSD2KsqzuVbHB/avYa8/zAzho2M+5SgJK3v7HZvv6XKeb5gf3soRpLpTjYBYBCENzLw2ckPSnp1XB6UtKnY62oe5VscB/0/Gvm+YFd889FcZcCJMIbG3fb8QNHmecH9sC0FXGXAwCJQnBHEpRscP/hvW60jH+8sjbuUoDEeHRWrXl+YMcNGGkL1u6MuxwASAyCezK9I+vvjqZyUJLBPZVK24lXjjbPD+yNjbvjLgdIjHQ6bb8a/rJ5fmBfummS7WloirskAEgEgnsypSQdGf6dDm+3N5WDkgzuK+v2mucHdvzAUdbE2NRAQXbsa7RTb5hgnh/YH56cF3c5AJAIBPdk+qqkN4V/fy283d5UDkoyuP/z1fXm+YGdftf0uEsBEmn2qm12TL/APD+wv8+huxkAdIbgnnwfaGf+YR3clzQlGdz/PHqxeX5g/f6xIO5SgMS6Y8JS8/zATrhytK3YsifucgCgpBHcky/abSbqvaKrTI8696Ea8/zAHp1VG3cpQGI1p9L24/tmmucH9j93TrOGpua4SwKAkkVwT760cgf3D0raV+RaekrJBfd0Om2fvHaceX5gc1dvj7scINE27Nxv/3n1WIZWBYBOENyT66ZwSkkaGrl9k6RbJc2QNDO26rpXyQX3TbvqzfMDO6ZfYPsbaSEEDtW4RZvM811/90mLN8ddDgCUJIJ7ck0Pp7SklyK3p0uaKOkhSR/qwnovllQrqUFSjdzFndrzy/D5doTThBzLV8ttYNFpTIE1lVxwn7jYhYyv3Tol7lKAspG5oNnHrxln63fsj7scACg5BPfke1Td9593lqRGSedLOlHS/XKBPFdXHEl6TNJFkk6W9GFJD0vaKenoyDLVkkZLen9kek+BdZVccL8zPKHud0/MjbsUoGzUH2i2/759mnl+YH0HjrJLHp9r05fWWSqVjrs0ACgJBPfke6ekd+eY/+7wvkLUyHW7yeglab2kfnk+/nBJuyX9LDKvWtKIAuvIVnLB/cLhc8zzAxs2dXncpQBlZVXdXvufO6e1dJvx/MD+688T7fbxS20drfAAKhzBPflGSbokx/yLJAUFrOctkpolnZE1/xFJz+e5jn+VVC/pO5F51XKt8FskLZF0r9yIN4UoueD+hRsnmecH9uKyurhLAcpOOp22BWt32sDnFthHB41pCfC9+wV27kM1Fry6gdFnAFQkgnvybZfr1pLtBElbC1jPB+Q2hFOz5t8k1xKfj3skrZD0tsi8syWdLukkuYOC1yXNlmudb89b5TbIzHS0Sii476o/0BIkduxrjLscoKzVH2i2Z+eutbOGzWzTCn/y1WPt6hcW2Rsbd8ddIgAUDcE9+fZJ+miO+R+VtL+A9RxqcO8ndxDxsU6W6xM+T0dXdR2sg09oLZngPmvFVvP8wD43ZGLcpQAVpXbrXrtpzGL7zPXj24T404e+aI+9tNp21x+Iu0QA6FEE9+SbKun2HPPvlPRiAes5lK4yl8t1h/lUns9VJ+nCDu4v6Rb3B6evNM8P7BePvBx3KUBFampO2cTF4VybaAAAIABJREFUm+xXw1+2Y/uPbAnwH75itP3xqflWs3KbpdOc0Aqg/BDck+8LckM3TpI0MJwmhfO+VOC6aiTdFbndS9I6dXxy6p8k7ZL02Tyf4z/khrA8vYC6SqqP+6VPzTPPD+y28UviLgWoeFt2N9iwqcvtK7dMbtMK/+WbJ9s9k5fb5t31cZcIAN2G4F4ePinpKbmTP+dLGi43PGOhzpIL/OfJ9ZEfJjcc5FHh/cMlDYks78sNH/kDtR3uMTOazTsl3SwX6nvLdY95RdJSuVb1fJVUcP/mbVPN8wMbt2hT3KUACKXTaZtTu83+75n5dsKVo1sCfJ/+I+3n1S/buEWbrKk5FXeZAHBICO7Idomk1XKBvEbSKZH7psiNEpNRqxx90eX6qEvS2yWNlRtR5kC4/P1qPRDIV8kE9/oDzdYn/GmeC8QApWlPQ5M9OXu1ff/uF9u0wn/quvE2ZNRiW7FlT9wlAkCXENzLy5slvSNrKgclE9xfXbujZUQL+tACpW/ppt12XbDIPnHNuDYh/kf3zrRn5qy1fY1NcZcIAHkjuCff2+VOTt0gKZVjKgclE9wfr1ltnh/YTx94Ke5SABSgsSlloxdusKq/1tgx/VoD/EeuGmP9/rHA5q3ZwcE4gJJHcE++uyS9Idc/fb+kX8h1VVkv6dz4yupWJRPcBz63wDw/sBtGvh53KQC6aMPO/XbXxKUtF1LLTN+8bao9OH2lbdvL9RkAlCaCe/KtkfTl8O89ko4L/z5PhV05tZSVTHA/I+wzO2LeurhLAXCIUqm0zVheZ797Yq71HTiqJcD3HTDKLvrbKzZlyRZrTtEKD6B0ENyTb5+kD4Z/r5f06fDvYyTtjaWi7lcSwb05lbYPX+FGq1i2mas1AuVk574D9sjMVfbtO6a1aYX/3JCJduu4JbZm2764SwQAgnsZWCg3lrskTZS70qkkXSwX5MtBSQT3ZZv3tFzkhVY4oHwtXLfTrhyx0E4aNKYlwPfuF9g5D75kq7cS4AHEh+CefJdL+n349zfkxmHfL3di6h/jKqqblURwHzFvnXl+YGfc/WKsdQAojvoDzTZi3jr7yf2zWgL8J68db6+t3xl3aQAqFMG9/PSR9GNJn4i7kG5UEsH9hpGvm+cHNvC5BbHWAaD4VtXttf++fVrLSDQzltfFXRKACkRwT7Y3y13gqG/chfSwkgjuP33gJfP8wB6vWR1rHQDisav+gP34vpktJ7COWrAh7pIAVBiCe/JtVetIMuUq9uCeTqft5KvHmucH9uraHbHVASBe9Qea7cLhc1r6vT86qzbukgBUEIJ78t0h6fq4i+hhsQf39Tv2m+cH1qf/SKs/0BxbHQDi15xKW79/LGjp937b+CVcvAlAURDck+92Sbsk1Ui6W25UmehUDmIP7uMWbWq5QAsApNNpu3XckpbwPvC5BYw2BaDHEdyTb3oH07QY6+pOsQf328a7L+hLn5oXWw0ASs8jM1dZ734uvP/mb3OsoYlf5AD0HII7kiD24P6LR142zw/swekrY6sBQGkKXt1gfQe4K6+ePWyW7a4/EHdJAMoUwT25+kg6LO4iiiT24P65IRPN8wObtWJrbDUAKF0zltXZiVe6Kyt/+45ptmV3Q9wlAShDBPfkSkk6MnL7KUlHxVRLT4s1uG/f29jSj3XnflrSAOS2cN1O++S148zzA/viTZOsduveuEsCUGYI7smVVtvgvkeuFb4cxRrcZyyrM88P7PM3Tozl+QEkx6q6vfb5Gye2XGV14Tqusgqg+xDck4vgXiT3T11hnh/YhcPnxPL8AJJl8656rrIKoEcQ3JMrJenfIrf3SDomplp6WqzB/fdPzDXPD+zOCUtjeX4AyZN9ldWRXGUVQDcguCdXWtJISc+GU5OksZHbmakcxBrcv3brFPP8wCYu3hTL8wNIJq6yCqC7EdyT6+E8p3IQW3Df39hsx4RjNG/aVV/05weQbFxlFUB3IrgjCWIL7nNXbw9PMhvHly2ALuEqqwC6C8EdSRBbcH90Vq15fmDnPlRT9OcGUF6Gc5VVAIeI4I4kiC24Z37i/vPoxUV/bgDlh6usAjgUBHckQWzB/fS7ppvnB/bPV9cX/bkBlKcZy+rsI1eNMc8P7Fu3T7PNuzl/BkB+CO5IgliCe1Nzyo4f6FrGVtZxBUQA3Sd6ldUv3MhVVgHkh+COJIgluL+xcbd5fmAnXjnaUpxIBqCbrarba1+4cRJXWQWQN4I7kiCW4P7s3LXm+YH94J4ZRX1eAJVj826usgogfwR3JEEswf3afy4yzw/sqhELi/q8ACoLV1kFkC+CO5IgluB+9rBZ5vmBPTV7TVGfF0Dl4SqrAPJBcEcSFD24p9NpO2mQG/WBfqcAiqE5lbb+z7ZeZfUv47jKKoC2CO5IgqIH9zXb9pnnB3bcgJHW2JQq2vMCqGxcZRVARwjuSIKiB/fRCze2jLEMAMUWvcrqrx+dY/UHuMoqAII7kqHowf3WsW+Y5wd2+dPzi/acABAVvcrqWcNmcpVVAAR3JELRg/sFD882zw/s4RdXFu05ASAbV1kFEEVwR7aLJdVKapBUI+kzHSz7S0nTJe0Ipwk5lj9M0jWSNkqqD5fpW2BNRQ/up1w/wTw/sNmrthXtOQEgF66yCiCD4I6osyQ1Sjpf0omS7pcL5Ee2s/xjki6SdLKkD0t6WNJOSUdHlvHDed+T9DFJz0taKeltBdRV1OC+dU9Dy5BsexqaivKcANCRtldZHcdoV0CFIrgjqkbS0MjtXpLWS+qX5+MPl7Rb0s/C24fJtbRfHlnmXXKt+WcXUFdRg/vUJVvM8wM77ebJRXk+AMgHV1kFQHBHxlskNUs6I2v+I3Kt5Pn4V7nuMN8Jb/eR27hOzlpuqqQ7OljPW+U2yMx0tIoY3O+ZvNw8P7CL/vZKUZ4PAPK1q/6AnTWMq6wClYrgjowPyG0Ip2bNv0muJT4f90haodZuMJ8L1/nvWcs9LempDtYzOHxcm6lYwf2Sx+ea5wc2dNKyojwfABQi+yqrD7+40lKM9Q5UBII7Mg41uPeTtF2uH3tGV4N7rC3uX75lsnl+YJPf2FyU5wOAQmVfZfVrt06xZ+eutaZmLhgHlDOCOzIOpavM5XInoH4qa35Xu8pkK1of970NTS0XPdmyu6HHnw8AuiqdTtv9U1fYRweNaQnwn79xov3tpVou2ASUKYI7omok3RW53UvSOnV8cuqfJO2S9Nkc92VOTr0sMu8IlfDJqXNqt5nnB/bp68b3+HMBQHfYVX/Ahk5aZp+4ZlxLgP/0dePtgWkrbF8jI2MB5YTgjqiz5EL1eZJOkDRMbjjIo8L7h0saElnelxs+8geS3h+Z3pm1zA5Jp0s6SdIIlfBwkI/MXGWeH1jVX2t6/LkAoDvtb2y2h6avbLkOhecHdvLVY+2OCUtt536uugqUA4I7sl0iabVcIK+RdErkvimSqiO3a5XjJFK5k0szMhdg2iR3UDBB0vEF1lS04P6nZ141zw/s5jFv9PhzAUBPaGhqtidqVtsXb5rUEuA/ctUY+/PoxVa3hy6AQJL9/+3deZhU1Z3/8Y8kajJxSSbr6IwFAsZdEreRcR016vgbRyeJmhhHHU18Rh2HJCbFpmBQiChiFBXBpSGKC3HDarZmaUDAZm2Qfe1uaPalm633+v7+OLeb6urq6m7orlu36/16nvNA3bp161unquv51KlzTxHcEQQpC+43v+jWSB7PEmsAAq6qusY+WbzZrn8+ty7An9FnvPX7dJkV7z3kd3kAjgDBHUGQkuBeWV1jXXuPt1A4YoW7DrbpfQFAqtTURG3Ssq12y0uz6gJ8l97Z9sexS2zjzgN+lwegBQjuCIKUBPflxaUWCkfs3H4TLRplTWQA7Us0GrVZa3bW/YBTKByxTj0j9siYRbZya2qW2wVwdAjuCIKUBPcP5hdZKByx24fPadP7AQC/LSjYbfe+mVcX4EPhiN2fNd8WFe7xuzQASRDcEQQpCe79Pl1moXDEnhy3vE3vBwDSxbLiEnvo7YV1v18RCkfslyPn2ux1O/nmEUhDBHcEQUqC+8+Hu6+Pxy7Y1Kb3AwDpZu32/fa79/Pt9F7ZdQH+tpc/tykrthHggTRCcEcQtHlwr6mJ2rlPuF8fXLGFuZ4AMlPR7oPW9+MvrWuf8XUB/oahM2xcfrFV1xDgAb8R3BEEbR7co9Gord+x38blF1tldU2b3Q8ABMH20jIbmL3Czn58Ql2Av+bZ6fb+/CLeIwEfEdwRBClbxx0AcNjegxX2/OTVdn7/SXUBvvugqZY1e6OVVVb7XR6QcQjuCAKCOwD4aH95lQ3PXWcXDsipC/AXDsixV3PX2b6ySr/LAzIGwR1BQHAHgDRQVllto+ZstO6DptYF+PP6TbQhk1fbngMVfpdXT01N1PaXV9m20jJbu32/5RfttcVFe62GufoIMII7goDgDgBppLK6xj6YX2TXPDu9LsCf9fgEezp7hW0vLTvi41bXRK3kUKUV7z1kq7fts4WFe2zmmh02fukW+2B+kb31+QYbNm2tDRq/0vp+/KX99r3F9sCo+faLEXPtlpdm2TXPTbdLns6xc56YWG+Jy9h2z5t5VsU8fQQUwR1BQHAHgDRUXRO1z5YU240vzKwLxl37jLc+Hy+1z5YU23vzCu31WRvsL1PW2NPZK6zXR0vt0XcX2X+/Nc9uHz7Hbn5xpl01eJpdOCDHzuw7IWHQPtrWqaf7RezLBk6xrr3dajl9Pl7KMpcIJII7goDgDgBpLBqN2tSV2+y2lz9vlbDdpXe2dXtykv3Ln6faDUNn2E9fmW3/9UaePfT2QvvD2Hx7ctxyGzJplQ3PXWdvf1FgnyzebDnLt9nc9bvsy80ltmHnAdu+r8wOVVTXC+gTl22tG4l/fdYGH3sMODIEdwQBwR0AAiAajdqcdbvsN6Pn28+Hz7H73ppnj4xZZD0/XGpPRZbb0JzVNnLmens3r9DG5RfbtFXbbd7G3ba8uNQKdx203QcqrLyqbVereW3GOguFI9axZ8Rylm9r0/sCWhvBHUFAcAcAtIpoNGo9P1xaNy//y80lfpcENBvBHUFAcAcAtJrK6hq7a+QXFgpH7JKnc2xryZGfUAukEsEdQUBwBwC0qpJDlXbdkFwLhSP2b3+ZaQfKq/wuCWgSwR1BQHAHALS6ot0H7cd/mmyhcMTuz5pn1azxjjRHcEcQENwBAG1iQcEe69rHLRP5p8+W+10OkBTBHUFAcAcAtJlx+cV1S1GOnlvgdzlAowjuCAKCOwCgTb00dY2FwhE7vVe25a7e4Xc5QEIEdwQBwR0A0Kai0aj97v18C4Ujds4TE23V1n1+lwQ0QHBHEBDcAQBtrqKqxm4fPsdC4Yh1HzTVtu9jmUikF4I7goDgDgBIiT0HKuzqZ6dbKByxW4Z9bocq2vaXXIGWILgjCAjuAICU2bDzgF3w5CQLhSP2P28vsBqWiUSaILgjCAjuAICU+mL9LuvSO9tC4Yj9ecJKv8sBzIzgjmAguAMAUu5vCzbVLRP5/rwiv8sBCO4IBII7AMAXQyatslA4Yp17ZdvstTv9LgcZjuCOICC4AwB8EY1G7ZExiywUjth5/Sba2u37/S4JGYzgjiAguAMAfFNWWW23vfy5hcIRu+KZabb7QIXfJSFDEdwRBAR3AICvdu0vt8ufmWqhcMR++spsK6tkmUikHsEdQUBwBwD4bs22fXZuv4kWCkfs0XcXWTTKMpFILYI7goDgDgBIC7PW7LTOvdwykUNzVvtdDjIMwR3xHpZUIKlcUp6kS5Lse46kD739TVKPBPv0966LbataWBPBHQCQNsbkFdYtE/nxos1+l4MMQnBHrDskVUi6T9LZkkZI2ivpe43sf7GkZyXdKWmrGg/uyyT9IKZ9p4V1EdwBAGllYPYKC4Uj1rX3eJu3cbff5SBDENwRK0/SsJjLHSQVS+rZjNsWqPHgnn+UdRHcAQBppaYmag+OXmChcMS6PTnJNu484HdJyAAEd9Q6TlK1pFvjto+S9Gkzbl+gxoP7QUlbJG2Q9I6k01pYG8EdAJB2DlVU27+/NMtC4Yhd8+x0KzlY6XdJaOcI7qh1itwL4bK47YPlRuKbUqDEwf0mST+XdL6kGyTNkVQo6cQkxzpe7gVZ204VwR0AkIa2l5bZZQOnWCgcsTtfm2sVVTV+l4R2jOCOWm0V3ON9U1KppPuT7NNfDU9oJbgDANLSii2ldvbjEywUjthjH+SzTCTaDMEdtdpqqkwi8yUNSnI9I+4AgECZtnK7derpVpp5efpav8tBO0VwR6w8SS/FXO4gabOO7uTUeCdI2iPp0RbUxRx3AEDay5q9sW6ZyMiSLX6Xg3aI4I5Yd8it336PpLMkvSa3HOT3vetHq/5I+XGSunlti9zSkN0kdYnZ5zlJV0nqKKm7pBxJOyV9twV1EdwBAIHQ79NlFgpH7Iw+421R4R6/y0E7Q3BHvEfkTh6tkBuBvzTmulxJWTGXOyrBXHRvv1rvyYX6CrnR+/ckdW5hTQR3AEAgVNdE7b635lkoHLELB0y2ot0H/S4J7QjBHUFAcAcABMb+8iq78YWZFgpH7Prnc620jGUi0ToI7ggCgjsAIFCK9x6yi5/KsVA4Yr96/QurqmaZSBw9gjuCgOAOAAicpZtK7My+bpnI3h8tZZlIHDWCO4KA4A4ACKSJy7ZaR2+ZyJEz1/tdDgKO4I4gILgDAAJrxIz1FgpHrGPPiE1evs3vchBgBHcEAcEdABBY0WjUen641ELhiJ3Zd4J9ubnE75IQUAR3BAHBHQAQaJXVNfar17+wUDhilzydY1tKDvldEgKI4I4gILgDAAKvtKzSrhuSa6FwxG56YaYdKK/yuyQEDMEdQUBwBwC0C0W7D9qFAyZbKByx+7PmWXUNK82g+QjuCAKCOwCg3VhYuMe69hlvoXDEnhy33O9yECAEdwQBwR0A0K58tqTYQmG3TOToORv9LgcBQXBHEBDcAQDtzrBpay0Ujlinni688wNNaArBHUFAcAcAtDvRaNT+MDa/buT9obcXWmlZpd9lIY0R3BEEBHcAQLsUjUZt5Mz11rlXtoXCEbvimWm2dBPrvCMxgjuCgOAOAGjXFhXuse6DplooHLGuvcdb1mymzqAhgjuCgOAOAGj3Sg5W2gOj5tdNnXlw9AIrOcTUGRxGcEcQENwBABkhGo3aG7M2WJfeburM5c9MtfyivX6XhTRBcEcQENwBABklv2ivXf6MmzrTpXe2vTFrA1NnQHBHIBDcAQAZp+RQpT04ekHd1Jlfj5pvJQeZOpPJCO4IAoI7ACAjRaNRy5q90br2dr+02n3QVFtUuMfvsuATgjuCgOAOAMhoSzeV2JWDp1koHLHOvbJtxIz1TJ3JQAR3BAHBHQCQ8UrLKu2hdxbWTZ3577fm2Z4DFX6XhRQiuCMICO4AAJibOvPXuQXWtY+bOnPZwCm2oGC332UhRQjuCAKCOwAAMZYVl9jVz063UDhip/fKtldz11lNDVNn2juCO4KA4A4AQJz95VX2v2MW1U2duefNPNvN1Jl2jeCOICC4AwCQQDQatTF5hXVTZy59eorN28jUmfaK4I4gILgDAJDEii2ldk3M1Jlh09YydaYdIrgjCAjuAAA04UB5lfV4b3Hd1Jm738iznfvL/S4LrYjgjiAguAMA0AzRaNTen1dkP+zrps5c/FSOzV2/y++y0EoI7ggCgjsAAC2waus++9fn3NSZTj0j9uKUNVbN1JnAI7gjCAjuAAC00MGKKvvd+/l1U2fuGvmF7djH1JkgI7gjCAjuAAAcoQ/mF9mZfSdYKByxi57KsdnrdvpdEo4QwR1BQHAHAOAorNm2z65/Prdu6szQnNVMnQkggjuCgOAOAMBROlRRbX8Ye3jqzC9GzLXt+8r8LgstQHBHEBDcAQBoJR8u3GRnPe6mzlw4YLLNWsPUmaAguCPew5IKJJVLypN0SZJ9z5H0obe/SerRCsdMhOAOAEArWrt9v90wdIaFwhHr2DNiQyatYupMABDcEesOSRWS7pN0tqQRkvZK+l4j+18s6VlJd0raqsTBvaXHTITgDgBAKyurrLaeHy6pmzpz+/A5tq2UqTPpjOCOWHmShsVc7iCpWFLPZty2QImD+9EcsxbBHQCANvLJ4s12tjd15sd/mmy5q3f4XRIaQXBHreMkVUu6NW77KEmfNuP2BWoY3I/0mMfLvSBr26kiuAMA0GbW79hvN74ws270ffDElVZVXeN3WYFRUxO1QxXVbX4/BHfUOkXuhXBZ3PbBcqPmTSlQw+B+pMfs792uXiO4AwDQdsoqq633R0vrwvsNQ2fYwPErbMKXW207U2jqKTlUadNXbbchk1fbr17/ws59YqI9N2lVm98vwR210im4M+IOAIBPPltSbOc8MbEuwNe27oOm2kPvLLSRM9fbwsI9Vl7V9iPM6aCmJmqrt+2zd/MK7Q9j8+3aIbkN+iYUjtg9b+a1eS0Ed9RKp6ky8ZjjDgBACm0vLbMP5hdZr4+W2o0vzLROPRsG1a69x9t/DPvc+o9bZp/mF1vR7oMWjQZ/ZZqSQ5WWu3qHPV87mt6v4YeYUDhiVw6eZj3eW2yj52y0LzeXpGRqEcEdsfIkvRRzuYOkzTr6k1OP9Ji1CO4AAPjoQHmVzV6304ZNW2v3Z823CwdMThhmL3oqx349ar69Mn2dzV2/yw5WVPldelK1o+nvzSu0P45dYtc1Mpp+Zt8JdvvwOfbnCStt8vJttnN/uS/1EtwR6w65tdbvkXSWpNfklm78vnf9aEmDYvY/TlI3r22RWxqym6QuLThmcxDcAQBII9Fo1Ip2H7RPFm+2fp8us1temmWde2U3CLyn98q2m16Yab0/WmpjF2yydTv2+zoqn86j6c1BcEe8RyQVyq29nifp0pjrciVlxVzuqAQnkXr7NfeYzUFwBwAgzZVVVtuCgt02cuZ6e+jthXbZwCkJQ/H5/SfZPW/m2Qs5a2zG6h1WcqiyTeqpqYnamrjR9I4Jpvyky2h6cxDcEQQEdwAAAmhrSZmNX7rFns5eYT97dbad0Wd8wjB/7ZBce+yDfBuTV2grt5Ye0a+4lpZV2ozVO2xozmq7+408O6+J0fRRaTaa3hwEdwQBwR0AgHagsrrGlm4qsVFzNlqP9xbblYOnJQzXZz8+wX4xYq4NnrjScpZvs11xo+Dxo+nXPx/80fTmILgjCAjuAAC0U7v2l9uUFdvs2Ymr7Jcj59b9imuikfL/HbMo6Wj6Fc9Ms/97d1EgR9Obg+COICC4AwCQIaprorZq6z4b462bnu4rvaQSwR1BQHAHACCDlRyqtJlrdtiwaWvb7Wh6cxDcEQQEdwAAkPEI7ggCgjsAAMh4BHcEAcEdAABkPII7goDgDgAAMh7BHUFAcAcAABmP4I4gILgDAICMR3BHEBDcAQBAxiO4IwgI7gAAIOMR3BEEBHcAAJDxCO4IAoI7AADIeAR3BAHBHQAAZDyCO4KA4A4AADIewR1BQHAHAAAZj+COICC4AwCAjEdwRxAQ3AEAQMYjuCMITpJkmzZtstLSUhqNRqPRaLSMbJs2bSK4I+2dKvcipdFoNBqNRqO5bASkpWPkXqAntXGr/YCQivtqD43+oq/oL/8bfUV/0Vfp0VLZX6fKZSMgo50k90d3kt+FBAT91Xz0VcvQX81HX7UM/dV89FXL0F9AivFH1zL0V/PRVy1DfzUffdUy9Ffz0VctQ38BKcYfXcvQX81HX7UM/dV89FXL0F/NR1+1DP0FpNjxkvp7/6Jp9Ffz0VctQ381H33VMvRX89FXLUN/AQAAAAAAAAAAAAAAAAAAAAAAAAAAAEC6elhSgaRySXmSLvG1mvTUS9J8Sfsl7ZD0iaQf+lpRcPSUWyrsBb8LSWOnSnpb0m5JZZK+lHSRrxWlr69IGiBpo1xfrZf0uPg1RUm6UtJnkrbI/c3dGnf9MZL+JGmrXN9NkdQ1lQWmmWT9daykZ+T+Fg96+4yWdEqKa0wXTb22Yg339umRgrqAjHOHpApJ90k6W9IISXslfc/PotLQREn3SjpH0gWSsiUVSvqGjzUFwcVyAWuJCO6N+ZbcB+e35D40d5L0E0mdfawpnfWWtEvSzZI6SvqZ3AfqR32sKV3cJOkpSbcpcbgKSyqR9B+Szpf0qaQNkr6WwhrTSbL+OllSjqTb5QZp/lluYGtBimtMF029tmrdJilfUrEI7kCbyJM0LOZyB7k/uJ7+lBMY35V787rS70LS2AmS1ki6TlKuCO6N+bOkWX4XESARSW/EbftQ7hsLHBYfro6RG2l/LGbbyXLftN6ZwrrSVVOjyJIbiDBJp7V9OWmtsb46VdJmuQGuAhHcgVZ3nKRqNfwDHCU3EoPGdZF78zrX70LS2ChJQ73/54rg3pgVcv00Vm4a1mJJv/a1ovTWWy4UnOFdvkDSdkl3+VVQmooPV6d727rF7TdD0l9SVVQaa05wv05SVPxCaKK+6iBpmqT/8y4XiOAOtLpT5P4AL4vbPlhuJB6JdZAb9fvc70LS2J1yc0Nrv4LPFcG9MeVeGyjpR5J+Izf/+B4/i0pjHeS+pYhKqvL+7eVrRekpPlx197b9Q9x+H0h6P1VFpbGmgvvXJC2U9E5qyklrifqql6TJOnyuSYEI7kCrI7gfmVfl3pT+0ec60tU/yY2Anh+zLVcE98ZUSpoTt+1FSXN9qCUI7pS0yfv3PEl3y53Uywed+gjuLZMsuB8raZykRWK0XWrYVxdK2qb6J+4WiOAOtDqmyrTcMLnQ0MnvQtLYrXJv7NUxzeRGRqvlVgXBYYWSXo/b9j9y55qgoU1yK2HF6itplQ+1pDOmyrRMY8H9WEkfy51g/+2UVpS+4vuqhw6/v8e+59fIBXgArShP0ksxlzvInVzCyan1HSMX2ouV2cunNceJcnP/Y9tfoCgJAAAIlklEQVR8SX8V5wQkMkYNT04dqoaj8HB2y32widVL7kRoHNbYyam/j9l2kjg5tVai4F4b2pfJLUgAJ76vvq2G7/nFclPaWDYZaGV3yL1x3yPpLEmvyS0H+X0/i0pDr8gto3aVpB/EtK/7WVSA5IqpMo25WG6udm+5k55/KbduNCdbJpYlN7hQuxzkbZJ2yq25nelOkBtR7yYXrn7r/b92FZSw3Pv7LXLTjD5RZi8Hmay/jpX75nmT3AnQse/7x/lRrM+aem3FKxBTZYA284jc1/UVciPwl/pbTlqyRtq9PtYUJLkiuCfz/+RO5i2XtFKsKpPMiXKvpUId/gGmp5SZYSre1Ur8PpXlXV/7A0zb5F5rU3R4dZ5MdLUa76+OjVxn3u0yzdVK/tqKVyCCOwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABBo98r94m6m+LakHXI/XpNKQ+R+DOYjSV9po/voLym/lY95o3fMDq18XAAAgLSUJffT7eno65K+53cRSXSUC7zdWul4z0saGbftRUkL5X6ZOVnwvUru5+UlqZek+ZL2y30Q+ETSDxu5XW9vv99I2prg/ltLf7V+cJfc47y7DY4LAACQdrKU+uB+XIrvr6WaW19HtV5w/ztJpZL+OW77i5IeljRayYPv85Je8f4/Ue7binMkXSApW1KhpG/E3eY3kvZIutS73FVSkaRBR/IAmtBfbRPcH5YL7wAAAO1elpIH929Kel3STkn7JE2TC4O1Okv6VNJ2SQfkQtR1cccokPS4XPjc591nR7nQ+5+Spks6JGmJpMtibnev6k+V6S8X/u72jlkq6T1JJ8bsc6KkdyQdlBtB/q2kXEkvJHmMtcd9QNJGSVFv+42SPvdq2C0p4j3eWhbXcmOue0DSSknlklZJeijJ/UvSz+RGx5uqsTHrvHoT+a5X35Vx97dV9Z9LSTpN0lpJv/MunySpTNJNcfvdJjdS/3fe5WckrZF7HjdIGiDp2CT156rhc/KJ3Guj1vGSnpNULPd85km6OkG9pvrPCwAAQLuUpeTBPUfSOEkXyY3IPidpl6S/966/QNKDks71rh8gF/ROizlGgVzI/r1cwOqsw8F9paSbJZ0haay371e9292rhsF9v6QPvfu7Qi58Ph2zz0jvGNd6+3wk92GhqeB+QNIEST+SdL63/adyHyy6yI2qj5O0VIfnVF/sPYZrJf0gpk/ukrTFu20n79/dku5JUsNfvPtPVmNjwf0cucfY2DcFXbw6z01y/GTGSvpr3La/xW3rK6m73PP675K2SfpjzPX91fLgPlLSbLnnubOkx+Q+CHWNu902udcKAABAu5alxoP75XKB+/i47evkplk0ZpmkR2IuF0j6OG6fjnJh8v6YbWd72870Lt+rhsH9oOqPsA+W9IX3/xMlVcqNJtc62btNU8G9Um5kOpnvqH4Arn0M8VNl1kn6Rdy2vpLmJDn2J5LeaKLGxoJ7b7lwnUgHuW8KPk9y7Kbcqvqj67Wj8I2N8EsuZC+IudxfLQvup0mqlnRK3D5TJA2M27ZIUr8ktQAAALQLWWo8uD8sqUZuNDq21chNjZCkE+RG4VfKheza6wfHHKdAUp+4Y3eUC70Xx2z7lupP6bhXDYP78rjj/FZuaobkRv9N9Uf7JRfsmgruaxNs7yrpXe/4++Qem0n6t7jHEBvcv+FtO6T6fVYuN52oMZMkvdxEjY0F9zw1foLmq3L9/49Jjt2U4+Tmwt/pXb5P7rF8NWafO+RGx7fp8OONnfrTXy0L7jfL9WP8a69K0vtxt5utw69HAACAditLjQf3sKTNclMt4tt3vH2GS1ovN+f5PO+6fNUPZQWSesQdu6Maht5vetuu9i7fq8Rz3GP18I4vHV1wTxSKV8kF6mslnSU3JcXkRqAbewzf97bdpYZ91ilJDe9IGnMENf6D3LcFf5/gumFyK80ku9/mGiE3VUhy06dejLnuMrnR8T46PKXqcSV/7qbJTQ+Kla3Dwf0O75g/VMN+/EHc7VbKjfADAAC0a1lqPLhfLxeeOia5/ZdyIa3WCXKBzY/gXjtV5qcx158sN1Lb0uD+ba+WK2K2Xa76wf0U7/KFcbctVv0+aY7HEtTQVI2Sm7KUG7ftGLnQXqyG88GP1FVyfXuO3Dcql8Zc93u5D2+xXlfy5+59SR/EXP6K3Mo3Wd7lM9Sw/xP5mlfXtU3sBwAAEHhZcqu6dItr/yQXAGfJBa6fyIXt7nIng17k3f4jSYu921wgNyobfzJogVIT3CV3QuMGSdfIhcy/efUMjX/gTRy3g9xJuH+VG+X9V0nzVD+4f1VuSkwfuZH2k73tD3jbH5ULoOfJTS+pXaklkfPkpoF8K2577YmxwyWt1uHnp/ZE1EiC474i129XyY1O17avJ7n/phwjt1Rkvtwc/li3eLXfKXcS6aNyJ+Mme+4elDv34Ga5cxpGyJ1PkRWzz9tyq/zUnuR7idwa9TfH7HO16s+/BwAAaLey1HBZQ5MbMZXcKPaLcqO3lXLh7W25YC+5AD5NLqgWyc2Lz5V/wT3RcpB5Sr42eaLjSm5ZyxVy87WXyAXh2OAuuZBeJDcKnRuz/ZdyH2gq5OaHz5CbTpRMnlygjZWrxM9PR7n59GVy4T5Wov1NR7/yyjPecZ5McN1guQ86++WW6Oyh5M/dsXIfMHbLzZfvqYaryhzr3ddGudfeFrkPiufF7POa3IcaAAAABNw35ALk/U3tmAZulvug0KGpHT3/6e2fqb4jF/xbYw4/AAAAUuxHcksxdpb0Y7lR3BIdPpk23fXQ4W8zmvITuTXTM9VFciexAgAAIIB+JGmh3Ampe+RWQDkv6S0AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACQ9v4/NXORlqHNgBkAAAAASUVORK5CYII=\" width=\"750\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f227203bc10>"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.datasets import load_digits, fetch_mldata\n",
    "reload(train_integers)\n",
    "\n",
    "mnist = fetch_mldata(\"MNIST original\")\n",
    "inputs = mnist.data / 255.0\n",
    "targets = mnist.target\n",
    "\n",
    "mnist_acc = load_object(\"mnist_float_accuracies.pickle\")\n",
    "if mnist_acc is not None:\n",
    "    accuracies, learning_rates = mnist_acc\n",
    "if mnist_acc is None: \n",
    "    accuracies     = np.zeros(20 - 5)\n",
    "    learning_rates = np.zeros(20 - 5)\n",
    "    for lr in range(-2, 20):\n",
    "        print(\"LR {}\".format(lr))\n",
    "        acc = train_integers.train_float_classification(inputs, targets, lr=lr, \n",
    "                  conf=[784, 10], iter=10000, weight_variance=0.1, smooth_window=1001, func=relu, \n",
    "                  func_der=relu_derivative, verbosity=\"none\")\n",
    "        accuracies[lr - 5] = acc[-1]\n",
    "        learning_rates[lr - 5] = lr\n",
    "    \n",
    "    save_object((accuracies, learning_rates), \"mnist_float_accuracies.pickle\")\n",
    "\n",
    "fig = plt.figure(figsize=(7.5, 5))\n",
    "ax1 = fig.add_subplot(111)\n",
    "surface = ax1.plot(accuracies)\n",
    "\n",
    "# axis labels\n",
    "ax1.set_xlabel(\"Learning rate (1/2^value)\")\n",
    "ax1.set_ylabel(\"Fraction bitwidth\")\n",
    "ax1.set_title(\"MNIST accuracy depending on fraction bitwidth & learning rate\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  Lets see them side by side:\n",
    "\n",
    "As we can see, the training is far more successful when using floating points, even though only for the backwards pass. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\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",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\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 backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\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 * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\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'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\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 * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\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\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\"];\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",
       "    var width = fig.canvas.width/mpl.ratio\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 + '\" width=\"' + width + '\">');\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 width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\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,iVBORw0KGgoAAAANSUhEUgAAAu4AAAH0CAYAAABiuKiqAAAgAElEQVR4nOydd5wcdf2HH4I0xZPfT6mBVELvPQGS0ItKkSqChCJIkd7rhcAFQnKigAi2gA2sFFEsaKgaeicgHQKKImJBQGB+f3xnf9nb7N3tzOzu7Ow+z+v1fiU7O7szc7f3+Tz7nQYiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiIiJSVMYDNwAvAxGwSw2vmQjcB7wNPAVMatC6iYiIiIhIzA7AucCu1CbuI4F/AzOAVYEjgXeB7Rq4jiIiIiIiUkYt4n4B8EjFtKuBmxqyRiIi0vIsCnTVMYs2d/VFRBLRKjWvFnG/FbioYtoBwBsplykiIgVm0WWWWjAiNJB65RWUdxFpTRpV85akr8wvUsO61CLuTwKnVkzbMX7tYjUsQ0RE2oguIHr+3hHR60+Oypzn7x1RamRdOW+XiEg1GlXzKtNdw7oo7iIikoguIHrtyZHRf18ZnTmvPTlScReRVqZRNW8ojRlx91AZERH5fxR3EekkWqnm1Xpy6sMV076HJ6eKiHQkXUD06hPDo7deHpk5rz4xXHEXkVYm75q3OLBOnAg4Nv7/sPj5qcBVZfOXLgc5DVgFOBwvByki0rF0AdGfnhgWvfnyiMz50xPDFHcRaWXyrnkTqX5M/Mz4+ZnArCqvuZ9wA6an8QZMIiIdS95NTESkmVjzRESksHQB0ctPLB/96+VhmfPyE8vbxESklbHmiYhIYekCohfnDI3emLtC5rw4Z2iaJnYE8BzwFjAb2GiAeRcCziLsLn4LeBDYPsV2i0hn0go1T0REJBV5N7G9CMdtHgCsBlwBvA4s1c/8FwBzCdcxHgUcBvwHWDf1T0BEOom8a56IiEhqws1I5iwXvT53+cx5fs5ySZvYbOCSssdDCGJ+Sj/zv0wYoS/nx8B3at9kEelg8q55IiIiqekComfnLBv9de7QzHl2zrJJmtjChMuaVV7H+Ergun5e8xpwUMW07xAOtRERGYw8a56IiEgmGtXEarmL4HLxvGMrpk8jjMRX43vAo8AYwuj8NsCbhMNtREQGQ3EXEZHC0gVET89ZJnp17nKZ8/ScZapdnzgCuqssO424LwlcC7xHGK1/AriUcJy7iMhgNKrmKe4iItJwuoDoyceXjl55adnMefLxpZOMuKc5VKbEovEyFiCcsPpo6p+AiHQSjap5iruIiDScvJvYbODissdDgJfo/+TUShYCngJ6apxfRDqbvGueiIhIarqAaM7jS0dzX1o2c+Ykb2J7Ea7Hvj+wKnA54XKQS8fPXwVMLZt/Y+BThEtBbg7cDDwDLJHx5yAinUHeNU9ERCQ1XUD06ONLRS+8tEzmPPr4Umma2JHA84QTTGcT5LzELGBm2eMJwGME2f8rQeyXS7vxItJxtELNExERSYVNTEQ6CWueiIgUli4geuixpaJnX1wmcx56zCYmIi2NNU9ERApLFxA98NhS0dMvLpM5D9jERKS1seaJiEhhsYmJSCdhzRMRkcLSBUT3PbZ09OSLy2bOfY95hQURaWmseSIiUli6gOieR5eO5rywbObc86hNTERaGmueiIgUFpuYiHQS1jwRESksXUA0+9FlokdfWC5zZj+6jE1MRFoZa56IiBSWLiC689Flo4deGJo5dz66rE1MRFoZa56IiBQWm5iIdBLWPBERKSxdQHT7I8tFDzy/fObc/shyNjERaWWseSIiUli6gOiWR4ZG9z6/Qubc8shQm5iItDLWPBERKSw2MRHpJKx5IiJSWLqA6LePrBDd9fzwzPntIyvYxESklbHmiYhIYekCopsfHhb94bkRmXPzw8NsYiLSyljzRESksNjERKSTsOaJiEhh6QKiXz08PLrjuZGZ86uHh9vERKSVseaJiEhh6QKiXzw0Mrr12dGZ84uHRtrERKSVseaJiEhhsYmJSCdhzRMRkcLSBUQ3PjQqmvXsmMy58aFRNjERaWWseSIiUli6gOj6h0ZHNz+7UuZc/9Bom5iItDLWPBERKSw2MRHpJKx5IiJSWLqA6KcPjol+9cwqmfPTB8fYxESklbHmiYhIYekCoh8/uFJ00zOrZs6PH1zJJiYirYw1T0RECotNTEQ6CWueiIgUli4g+uGDq0Q3PrN65vzwwVVsYiLSyljzRESksHQB0dUPrBZd//SamXP1A6vZxESklbHmiYhIYbGJiUgnYc0TEZHC0gVE33tgjejap9fOnO89sIZNTERaGWueiIgUli4g+vb9a0Y/fmqdzPn2/WvaxESklbHmiYhIYbGJiUgnYc0TEZHC0gVEM+9fO/rBU+tlzsz717aJiUgrY80TEZHC0gVE37xv3ejqP26QOd+8b12bmIi0MtY8EREpLDYxEekkrHkiIlJYuoDoa/etH333jxtlztfuW98mJiKtjDVPREQKSxcQXX7f+tFVT26cOZfbxESktbHmiYhIYbGJiUgnYc0TEZHC0gVEl923YTTzybGZc9l9G9rERKSVseaJiEhh6QKiS+7dOPrGE5tmziX3bpymiR0BPAe8BcwGNhpk/mOAJ4D/AC8CXwQWTbjdItKZtELNExERSUXeTWwv4G3gAGA14ArgdWCpfubfhyD4+wAjgG2Bl4HetD8AEeko8q55IiIiqekCoi/fu0n0tSc2y5wv37tJ0iY2G7ik7PEQYC5wSj/zXwLcXDFtBnB77ZssIh1M3jVPREQkNV1A9MV7xkVfnTM+c754z7hSExsav3cpi1RZ9sLAu8AuFdOvBK7rZ333Af7OvMNpRgGPA6dl+SGISMfQqJqnuIuISMNpVBOrTHeVZS8XPze2Yvo0wkh8fxwFvAP8N379ZVl+ACLSUSjuIiJSWLqAaPo9m0WXzpmYOdPv2SzJiHsacZ8I/Ak4GFgT2BV4ATgz249BRDqERtU8xV1ERBpOFxBNu3vz6OLHt8icaXdvnqSJpTlU5jbgwopp+wJvEo6PFxEZiDxrnoiISCbybmKzgYvLHg8BXqL/k1PvBS6omPZpgrgvWPNWi0inknfNExERSU0XEJ1/94Toose3ypzz756Q5nKQbwH7A6sClxMuB7l0/PxVwNSy+buBfwB7AyOBbYCngGuy/RhEpEPIu+aJiIikpguIeu7aIup9bJvM6blrizRN7EjgecL13GcDG5c9NwuYWfb4A8DZBFn/D+H49kuBJVJuv4h0Fq1Q80RERFJhExORTsKaJyIihaULiKbctWV04WPbZs6Uu7a0iYlIK2PNExGRwtIFRJNnbx1d8Oj2mTN59tY2MRFpZax5IiJSWGxiUmIW8EjeK9FEZgLPpXxtN+FzXiT6uxFaf/NekmFZSwM/Al6L3+sYwj0YovjfPLHmiYhIYekCorNmbx31PLpD5pzVeU3scML2DnSn16Iwi9YS9+UIorlOg95/Jp0t7uPix9VO7M4q7t8D/gmcQLjPwio0R9wH2qYS1jyRDmUWrdXkGs1MOrvJDTZv2iY3In79pJSvz0oXEJ3xh22jcx/5eOac8YdtO62J3QE8S9jmFXNel6zMorVq2gY09m9jIarf0bcWPgAsWsd1aQaLEta7xAmEn++IKvNmFfc/Ad+pmDaRxov7QNtUwponkgBHpxqHo1P1pVmjUyNILycfjNdpYsplg00sCyMJ27or8CrhMpVFZhatVdOSivsHG7cqbUkjxf39Kq+fiOIuUjgcnWocjk7Vl2aNTi0QLyvNXT8/RrI9A9XoAqJTf799NPnhT2bOqb/fvpOa2BnA34CFga8AT/Yz3xDgaOBhws2m/gLcRPibLWdf4C7CXWBfB24Ftk24TksA7wFHlU37GEGkXiN83kpcRhgZLTGLUNNWA34Xr8dc4KQqy1kEmEy4nv7bwIvANOavEaW/jV3i934beBTYfpDtmBi/tjKTKtZ1fcLP6U3govi5nYEbgZfj5T0NnMn8f2Mz6TsYMSJexgnAIfHr3gbuBjaseG038w9GJNnWicA9hM/D08Ch/bxnJUcRfr/lAwjHx6/rLZu2IOFQlfK7DJfXitKyKjMixbaUM6mf94X+xX0Pwh2R/wP8lTBSP7RinrUIv69nCD+zPwHfBD5aNs9g27QNcDvwd/KveUcQPntvEQYyNxpk/mOAJ+Kf0YvAFyleT5UC4uhUY3F0qrE0cnQqLXUT91N+v0N09sM7Zc4pv9+hk8T9ceDr8f83J2x3peABfCt+7ucEgT8euJZw46kSZ8fz3EH4rB0FfBc4P8V6PUg4MbDELgTZi4DVy6Y/Avyw7PEsgqi/QJDgw4Cb49ftUDbfEOCXwL8JAnEIcDHw33i7yomABwgSfQZh+5+OX/tR+mdpgmxHhLvp7htnVNm6vkLoJV+O12Hn+LmfEu6kewLweeAH8ftcWLGMmVQX9/uAPxK+sJxI+KL1ImHwokQ31cW9lm1dlyBszwInA6cRfu4PVHnPStaN5/lE2bRrCb/fu8umlfrBxyvWrzv+/1qE49BLJ46Wfr4fSrgtlYyK3ycCflX2vlBd3CfF0+6K12Mq4UvYs8z/5eRWwmfic4TP55sE6S19GR1om1Zn3pewE8m35u0Vr8sBhC/JVxC+qC/Vz/z7ED4v+xA+o9sSfi+9/cwvUjccnXJ0Ku22TqQ9R6eg+qEyM4F/EUadro3//xdgOvN+L6XXVaa0zssQhPGleH1eAa5j/i8eins61ids59bx4wUIf9cXVcy3RTzfl6q8R6m+rEj4nP6EUP+qzZOES+hbq2YAtwB/JogswP8S6lx57ZsVr+t+ZdMWJnx2yr8I7Buv72YVyz00fv24smkR4fM3umzaWvH08i8u1RhoMKK0rodWeW6xKtO+SpDO8po7k+o17a/A/5RN34n5Zbmb6jWtlm29Pl6X5cqmrUj44jNYTRsCvMG8WrVAvL4/AN4FFo+nH8v8ta/yS/5ggxFpf2+l1w92qMxChM/kw/QdPf54PN/ksmnVfqd7x/NtXjatv206Jp7+MfKvebPp+7MZQnCHU/qZ/xLCF+hyZhD2Hog0FEenHJ1Ks63tPDoF/Yv7fwifuW8Qfi8/iuc7LJ7nQ/H0iCB8pXVaK37+DsIu4SnAQcCpwG+B8RXL7wKiE+/8eHTGQ7tkzol3frxTxL2XIMflX3CnV5l2CUGQ/3eA9yrJRr3OT9krfr+V48d3ET4HPyLUSZgno+uWvW4W4Qts5ZeF6wh/5+WPHyFIUHnGxO95etm8EWFgoJI3GHzEcDBxf4vwxWIgPhyv22fi91q77LmZVK9pl1a8x//E08u/5HRTvaYNtq0LEgZOvltlvuurvGc1fgH8Pv7/avFr1iPUtW3i6T8h9LbK9esuezyYuKf9vZVeP5i4j6VvTSvnccJgTTUWJfxOR8SvP7rsuf62aVI8/SDgI+RX8xYmfMHapWL6lYS/q2rsQ6jlpcNpRhF+PqfVsDyR1Dg65eiUo1PVGUF1cY8IX8bKuY++zay/Q2WWYN7ekMHoAqLj7/hEdNqDu2bO8Xd8ohPEfUHCl7TvEz6LpexB2PbyPX+/IOz1GIjLCJ+/wSS0VpaL1+Ngwhe8/8brdBRhoAHCF/M36FtDZxGEoJKZhC/OJR6j+t6eUsrrd0TYvkqeIwzSDMRgNe3pfl63OmFA4o0q61b+xXUm1WvayVXeM6Lv4Z3dVK9pg23rssw/mlyit8p7VuM04B1C7T6U8FmEUB/Oif//KvOLc9Kalvb3Vnr9YOJeGjHfssrrf0oYBCrxv4TP1Z+Z/3d6Vtl8/W3TYoQR6lLPakTNG0qoe6VUOyes9Lc5tmL6NAa+aMdRhN95qe9V+92I1BVHpxydcnSqOiPoX9yXrJj3S4TDzUr0J+6LEL5I/Iy+X6qqobgnZxsGFtcry+bNQ9whnMR3FWGw5F3C33ZpD9Rw4A/xupUzi+rn7cyk79/+HOCh+L2rZeWyeasJHPH7zRxkGwaradXWdQmCmD1DGIn9RLxOJ9FXGmHgw/8qqfw766Z6TRtsW+sh7uOZJ7zfIewxhVAffke4ZnpE6G2V69dd9jjNeTvPMfjvrb/XTyS9uP+G0AsmE86T2wbYjmTbNATYKl6vRtS8ynRXWYc04j6R4EoHA2vG2/8C8w/siNQNR6ccnerG0an+GEF1cf9PlXm76bvNA52cegzh7+QdwnkNJxGOe6+kC4iOuWOn6OQHd8ucY+7YqRPEfSZh5G/3Kvke8A/m7cnKYzACwpeHZwl/O6W9NEMIu9wPJXwuTq14zSxqE/cbCXW6lj2cWQSwtKd2UpXnZlF9XXdh/toF4YTGVhD3BQl/21kGIxYlfDE/G3ieeXv1diPI7RHx+yxbZf26yx6XzvcZUWUZzRD3gQ6VeYx5n9vSYNBZFfOUBr+6y6YNtE0lGlXzahlxT3OozG3Mf+jqvoTfdeVRByJ1wdEpR6e6cXSqP0ZQXdz/VWXebmoXdwiH7hxPuLrD24STuNetmKcLiI66fefoxAd2z5yjbt+53cV9MYKYf6Of58fR9/NUz8P/liV8XhdicA6OlzuHcG5NiZ8TLisXMf/he7OorabtH7/+kCrzLsa8cz8g299G6W/zmCrPzaL6un4yfs2EsmkLA/fTGjUN4AbSH/5X4nbC7zZi3p7RpeLHTxAuhFBJ5TaUzpGp9oWxGeJeOjn1QfpK7g70rftdzD8YBGFvb63bVP7FOe+aN5twnluJIQTv6e/k1Hvpe+EEgE8TxD3NZYRFBmUmjk45OuXoVH+MIL24f7TKevbHGIIsVN7NMO8mVjRKh9bt3M/zQwh7cK4vm3ZV/JqfE/bkHQP8mL7nP5wTz3MH4bN2JKEuTS2bZyaDjyaWWJl5gyOfKpt+SjztLeYfEZxFbTVtCKGuvU/Yk3ok4Yv/ZYSrcZVfASzL38ZChC+bcwgnFe5NuKzwQOv6UcLhZM8BxxHOX7mPeSezTxxgu0bQHHFfn1CTniUMkpxKOOG+9OWiFqbG8/6dvl/2SjJfbU9f5TZsGE+7kXCu1t70PeG+0eIO804a/QPhM9RDqFPP0veco1vi6ecSRuh/yrzfaS3bdBHhczCF8HnNs+btRfj72x9YlXBBidcJF5mAUC/K/+67CZ5U+vxvQ/hidg0iDcDRKUenwNGpgRhBenFfLH5ceZL3B5n/5hxDCMdJ/rBiehcQHXn7rtHxD+yZOUfevmu7i/v1hC+TA90H4VuEL/ulKwotSPg7eZwgbK8Sast6Fa87gCAXbxHkcxbzTuiHZOIO807kK78+9KbxtFurzD+L2moahLp6Ujx/aX3vIRzOUP67zyqAOxEurVr6e580yLpC6Cu/Z94lei8gHP7YKjUNwh7A+wifh6cIX0ymU/0QuWrsyLwvg+V8LZ5+YJXXVPuSfwZhYKl0JbURZfM2Q9wB9mTe5/41qt+AaSih579O+LLyA+btke2umLfaNm1JuKLYXMLPPO+adyRhIOltwgj8xmXPzaLvz/gDhIGnpwifjxcIexuq3cVbJDOOTjk6BY5ODcQI0os7BKl5BTg8Xqc1CF8uXiN8xr5AGKH6Vfza3Spe3wVEh932qeiY+/fKnMNu+1S7i7tIo7iWcGldaSzWPJEBcHQq4OiUo1P9MYJs4j6W8Fl6u2ydPxqvz+Px+/ydsCt6jyrvaRMTaT6Vl+EdQ+iDX8thXToNa56ISA44OlUfuoDo0Ft3i466b+/MOfTW3WxiIoPzCmFP4OcIx22/RviSPSbPleoQrHkiIg3G0anG0QVEh9yyR3TkvftkziG37GETExmcbxH2yL1FuBTvTcy/V1kagzVPRKTBODrVOGxiItJJWPNERBqMo1ONowuIDrplz+iwez+TOQfdsqdNTERaGWueiIgUli4gOmDWntGh9+ybOQfMsomJSEtjzRMRkcJiE2s+CzD/7dGNMekylNpucFiiC2ueiIgUlC4g2v93e0efu/uzmbP/7/a2iQ3OUObdN8IYkz2VN2waCGueSEYcfTKmfkk1+rTf7z4dHXT3/pmz3+8+3SlN7AjmnXcxG9gowWu7gGgzdowmsrMxJmU2Y8c09caaJ5IRR5+MqW8Sjz7ZxBKxF+GGVwcAqwFXEO5avNRALyqjC4gmsnO09QK7G2NSZiI7p6k31jyRjDj6VMqGpyXOhI1OT5UNd56SPLskz+bjz0qczSacnS4Tm5MJm5zRktl0gxNTN7F9frtPNOmuSZmzz2/36YQmNpu+d8odQrjr8Ck1vl5xN6YOySLu1jyR9NjEShl7TuJsNW5Kqozd/cLk2SN5ttiyJ3EmbjU1XbZuTrba7NyWzIRNzkjdxPa+ed/os7MPzJy9b9633ZvYwsC7wC4V068EruvnNYsw/+FM1jxjMiaLuFvzpJ1IeuzmHsCceP6HgR0TLk9xL0VxV9wV91ZnOcL2ja2YPo1QL6vRTZVDmqx5xmSL4i6S/NjNcYTRpxOBVYEphFvRr5FgmYp7KYq74p6TuO95837Rvn84KHP2vHm/dm9iacTdEXdjGpAs4m7Nk3Yh6bGb1wA/q5j2B+CrCZapuJeiuCvuOYn7Hjd/NtrnDwdnzh43f7bdm1iaQ2UqseYZU4dkEXdrnrQDaRrSC8AxFdMmAw8OsBxHn/qL4q64K+5FYDZwcdnjIcBLeHKqMU2N4i6dTppdwO8An66Ydjjw5wGW043He1aP4q645yTuu/1m/2jv338uc3b7zf6d0MT2IpzTsz/hEMHLCYcULl3j6xV3Y+qQLOJuzZN2oFni7oh7f1HcFfecxH3XXx8Q7XnnoZmz668P6JQmdiTwPOGcoNnAxgleq7gbU4dkEXdrnrQDzTpUphKbWCmKu+KuuHcC1jxj6hDFXST5sZvXADdUTLsTT05NF8Vdcc9J3Hf+1YHR7nd8PnN2/tWBadYhySVoZ1H9brE3Jlhe3ljzjKlDsoh7zjVPpG4MduzmVcDUsvnHAf8FjgdWIRy/7uUg00ZxV9xzEvdP/uqg6FN3HJY5n/zVQUnXIeklaP8XWKYsqxP2FE5KsM15Y80zpg7JIu451jyRujPQsZuzgJkV8+8BPBHP/wjegCl9FHfFvfPEPeklaCs5BvgH8KEa528FrHnG1CGKu0g+tHwT236FoxMnjRhv+qsTE+eEB3ZvWjb55UmJs/Vvj0mcnW87LFUOu+czifOF+/ZOnAPv2j9x0qxb0hw0a8/UTezjvzw42uX2wzPn4788OMk61OOa6A8TRumLRMvXPGOKkCzinlPNE2kLWr6JKe6Ke7uL+w43fS7a6bYjMmeHmz5XWoeh9L161CJVlp3mSlblbBS/fqBj4luRlq95xhQhWcS9ATVPcZeOoeWbmOKuuCvuiZtYZbqrLDuruF8OPJRgW1uFlq95xhQhirtIPrR8E1PcFfd2F/ftfnFI9Ilbj8yc7X5xSJIR9yyHynwIeAM4OsG2tgotX/OMKUKyiHsDap7iLh1DyzcxxV1xb3dx3+bnh0Y73vKFzNnm54emOTk1ySVoS0wiXAHrowm2tVVo+ZpnTBGSRdxzrHkihaflm5jirrgr7g1rYkkvQVviNuDqBNvZSrR8zTOmCFHcRfKh5ZuY4q64t7u4b/3zQ6Ptbzkqc7ZO18SSXoJ25XgZ2yRYRivR8jXPmCIki7jnXPNECk3LNzHFXXFvd3Hf8sbPR9vOOjpztrzx8zaxwWn5mmdMEZJF3K15Iulp+SamuCvuirtNrI60fM0zpghR3EXyoeWbmOKuuLe7uE/82WHR1r87JnMm/uwwm9jgtHzNM6YIySLu1jyR9LR8E1PcFfd2F/fxNxwebfnbYzNn/A2H28QGp+VrnjFFSBZxt+aJpKflm5jirrgr7jaxOtLyNc+YIkRxF8mHlm9iirvi3u7ivtn1R0QTbz4ucza7/gib2OC0fM0zpgjJIu7WPJH0tHwTU9wV93YX902vOzKa8JvjM2fT6460iQ1Oy9c8Y4qQLOJuzRNJT/Oa2IJ7pcp2Sx2WOBvuNz1x0kjkN54YlyqXzRmfON96YhPTwvnKvRvYxIqB4m5MHaK4i+SD4q64m5zFfex1X4g2/80JmTP2ui/YxAZHcTemDski7tY8kfQo7oq7yVncN772qGjTX5+YORtfe5RNbHAUd2PqkCzibs0TSY/irrgbxb2TUNyNqUMUd5F8UNwVd5OzuG/006Ojcb86KXM2+unRNrHBUdyNqUOyiLs1TyQ9irvibnIW9w1+cky0yS9PzpwNfnKMTWxwFHdj6pAs4m7NE0mP4q64G8W9k1DcjalDFHeRfFDcFXeTs7iv/+Njo41vOiVz1v/xsTaxwVHcjalDsoi7NU8kPYq74m5yFvf1fnRctOEvTs2c9X50nE1scBR3Y+qQLOJuzZN24FTgbuCfwKvAtcDKg7xmEuEDW563Ei5XcVfcjeLeSSjuxtQhirt0OjcRRHx1YG3gRuB54EMDvGYS8AawTFmWTrhcxV1xNzmL+7o/Oi7a4OenZs66NrFaUNyNqUOyiLs1T9qRJQkfxvEDzDMJ+HvG5SjuirvJWdzX/tHx0Xo/Py1z1v7R8TaxwVHcjalDsoi7NU/akRUJH8Y1BphnEvAuYWT+ReA6woj9QCxC+ICXMhTFXXE3invnoLgbU4co7ibTuNIAACAASURBVCLzGAL8DLh9kPnGAp8F1gEmADcQDp1ZfoDXdDP/cfHRxCGfSi3Wjc52HzskcTbcd3rijJyRPJMf/kSq/PmlZZuSv760XOK88tKyqZJmWe2UZx9fJnUTW+uHJ0Tr3nh65qz1wxNsYoOjuBtTh2QRd2uetBuXAc8xsIBXYyHgKWDKAPNUH3FX3BV3xT03cV/jBydGa//sjMxZ4wcn2sQGR3E3pg7JIu7WPGkko4ADgVOA0yrSCC4hHPYyMuXrfwh8P8H8oYkp7oq74q64p6fZdTILirsxdYjiLq3IgcB/gb8AjwAPl+WhOi9rAYK0zwXGpHyPBYE5QG+C1yjuirvinrO4r37NidFaN5yROatfk0sTa2adrAeKuzF1SBZxL3jNkxbmOZo3YvQVwhViJtD38o6Llc1zFTC17PFZwLaE0a71CCPt/wFWS7BcxV1xV9xzFvfVrj4pWvP6MzNntatPyqOJPUfz6mQ385+jMyfheyjuxtQhWcS94DVPWph/EKS4Gcx3wmicSWXzzAJmlj3+IuGKMm8DfyJc+33dhMtV3BV3xV1xz0Iz62Q3YVS/fHDjYwnfQ3E3pg5R3KUV+RbwubxXosEo7oq74p6zuK/6/ZOjNa47K3NW/f7JeTSxZtbJbuCBjO+huBtTh2QR94LXPGkxDi/LScCrwNeBoyueOzyvFawzirvirrjnLO4rf++UaLVrz86clb93SrOaWF51shv4N/Ay8AzwXWBYwvdQ3I2pQ7KIewFrnrQwL9aYF/JawTqjuCvuirvinpS86uQOwB7AWsB2wJ2EwwU/PMBr8rvpnDFtHMVdJB8Ud8Vdcc9Z3Ff67inRqj89O3NW+m7HNbElCDedO2iAebqpdtM5xd2YTMki7tY8aRSn0feqLiUWpTWvT5wGxV1xV9xzFvcx3zklWuUn3Zkz5ju5NLG86+Td9L3aViWOuBvTgGQR94LXPGlh3gOWqjL9o/Fz7YDirrgr7op7FvKsk4sDfwOOSvAaj3E3pg5R3KUVeR9Yssr0CYSbjbQDirvirrjnLO4rfvvUaOUfT86cFb99ah5NrJl1cnr8viOAccCv42VUW35/KO7G1CFZxL3gNU9akL8QrpLwXtn/S3kNeBf4am5rV19aXty3WXif5NlocuJsuN/0xLn12VGp8s7LyfPPucOakrdeHpkqabYpTdKsWzN+bnPnLJ+6iY3+9qnRSj+enDmjm9vE8qiTVxOuKPM28FL8eHTC91DcTWGz7eL7t0y2/NCnO63mSQtzEHAwYSTp2PhxKfsBm+e3anVHcVfcFXfFPQ1FrZOKuyls8pZ1xV1ana2AhfJeiQajuCvuinvO4j7qqtOiMT86J3NGXXVaHk2saHVScTeFTd6yXi9xL3jNkxbjgwnSDijuirvinre4X3laNOaH52TOqCub1sSKXCcVd1PY5C3rdRP34tU8aWHeJxyzWUvaAcVdcVfcFfekFLlOKu6msMlb1ttE3I8AngPeAmYDGw0y/xLApcArhPNrngR2TLA8aTBbleVA4E/AhcCn4lxI+OUdkNcK1hnFXXFX3PMW95mnRyv+YErmjJp5erOa2HWEE1DfIdTIv1GcOqm4m8Imb1mvm7jnV/P2Isj3AcBqwBXA61S/pC3AwoR7RdwIbEq4otUEYO0E2yxN5NfAvlWm7wv8tsnr0igUd8Vdcc9Z3Ed+6/Ro9DVTMmfkt3JpYndT/UZLrVonFXdT2OQt6/US9xxr3mzgkrLHQ4C5wCn9zP954GmKdR5PR/MmMKbK9DHxc+2A4q64K+6dK+71aGJFq5OKuyls8pb1Fhb3ofS9Q/IiVZa9MOEytbtUTL+SsBexGj8HvkMY1Pgz8AhhoGLBBNssTeRJqt9Ke2r8XDuguCvuinvO4j7im2dEo64+N3NGfPOMPJrYX4Hzq8zbqnVScTeFTd6yXi9xb0DNq0x3lWUvFz83tmL6NMIgRjXmEA4j/AawPmEv5WvA2Qm2WZrIJwi/sPsJNxL5KnBfPO2TOa5XPVHcFXfFPW9x/8aZ0ajvn5c5I75xZh5N7B/AfylOnVTcTWGTt6zXTdzrX/NqGaxIU/OeBF6g7wj7cYRzeKRFGUH4pV4f54J4WruguCvuinv7iXuzm9irFKdOKu41ppZaO3rqjEHz55eWHTS1/v2/98qKg6Ze72MGzutPjmqlmlfLOqTZy3gL8JuKaTvEy1y45q0WqSOKu+KuuOcs7sO/fmY08nvnZc7wr9vEakBxV9xzl952SBZxz6nmQRiUuLjs8RDgJfo/r6eHcNWtIWXTjgZeTrDN0mBWY94vaLVB0g4o7oq74p63uH/tzGjkd8/LnOFfa1oTW515dbKHMOJelDqpuCvuuUtvOySTuOdX8/YiHMa3P7AqcDnhSlpLx89fRd9zG1cgHA54MbAS8HHC+T2nJ9hmaTDvM+9SaKWbjLxflvfK/m0HWl7c02T7kcclzgb7z0icEV+cnippxPCNucsnThoxzrsZFDUd1sQiYBNCE4voWyNbvU4q7op77vWiHVLQmgdwJPA84VK4s4GNy56bBcysmH8s8AdCrXwaryrTcowGFij7/0BpBxR3xd2GlnMTG3bFWdGI7/RkzrArzmpWE7ufeU1sOuHSj0Wpk4q74p57vWiHFLjmiRQaxV1xt6G1QhP7dk/m2MRqQnFX3HOvF+0Qa560Is8C3yTsRh7e4GV1M/8l3OYM8po9mHd5toeBHVMsV3FX3G1oNrEsNLNO1gPFXXHPvV60Qzq45kkLM4nQkJ4hHK/5LGG38QHAyDovq5twM5NlyvKxAeYfR7gixImEY1OnAO8AayRcruKuuNvQcm5iK1x+djT8qqmZs8LlZ+fRxCbRvDpZDxR3xT33etEO6eCaJwVhBWBf4HuEm428W+f37wYeSDD/NcDPKqb9gXDzkyQo7oq7DS3vJvbVs6PhV07NnBW+mnsTa3SdrAeKu+Kee71oh1jzpFVZFNgSOAe4DfgP4bCUiwd6UQq6gX8Trg36DPBdYNgA878AHFMxbTLw4CDLWYS+N2YZiuKuuNvQbGLZaFadrAeKe63ivtDeg2aHW74waGq5ZGvef7/Gmiftwa3Am8BDwJeB3Rj48JUs7EA4Zn0tYDvgTsKVHj7cz/zvAJ+umHY44fqiA9FNlVuiK+6Ku+KeZxPrjoZfeX7mrPDV7jyaWDPrZD1Q3BV3Y82TNuXvwF8I1zI+EBjVxGUvAbwBHNTP82nF3RF3xV1xb7Umdll3NHzm+ZmzwmW5NLE862QaFHfF3VjzpE0ZAqwHHEe4BfjfCKPgVxFOvGo0d9P35iflpD1UphKPcVfcFXebWBbyrpNJUdwVd2PNkw5hXcLVEv5L4+8IuDihAR7Vz/PXADdUTLsTT05V3BX34jWxr3RHw791fuas8JWWaGLNrJNpUNwVd2PNkzZlLYI4/wT4K+EKCeXHcdaT6cAEYAThUo+/Jux+XjJ+vvLW4+MIjfF4YBXCseteDlJxV9yL2MQunRwN/+YFmbPCpZPzaGLNrJP1QHFX3I01T9qU9wi39r4I2BX43wYu62rCFWXeBl6KH5ffLnwW8996fA/gifg1j+ANmBR3xd0m1vwm1sw6WQ8Ud8XdWPOkTfmfvFegCSjuirvinnMTW/6SydGwb1yQOctfkksTK1qdVNwVd2PNkzblSaqPHi0RP9cOtKW4b7vYvomzzUaTE2fUhTNS5dZnRyXOX19aLnH+Nndo4tRys5R6Jc02tWqefXyZ9E3s4snRsK9fkDnLX5xLEytanVTcFXeTt7gXu+ZJC/M+sFSV6UsTDk9pBxR3xV1xV9yzULQ6qbjXUdzXPKp30NRSd/KWUKO4S7HZMc77hGul71iWTwJfIhxb3g4o7oq74p63uH/5nGjY16ZlzvJfPqeZTayodVJxV9xN3uJezJonLcz7cd4r+38p/wWeBnbKbe3qi+KuuCvuOYv7Cl86Jxp+xbTMWeFLTW1iRa2TirvibnIW94LWPGlhFgQ+ALwILBM/LqXdUNwVd8VdcU9DUeuk4q64G8VdpLAo7oq74p63uF90TjT88mmZs8JFNrEaUNwVd5O3uFvzpI4cDixa9v+B0g4o7oq74p63uH9xSjT8qxdmzgpfnNKsJlbkOqm4K+4mb3EvXs2TFuZF4KNl/+8vL+SydvVHcVfcFXfFPSlFrpOKu+JuFHeRwqK4K+6Ke97i3jslGn7ZhZmzQq9NrAYUd8Xd5C3u1jyR1Cjuirvinre4z5gSDf/KhZmzwgybWA0o7oq7yVvcrXnSQPYHHgDejPMAMCnPFaozirvirrgr7lmpV50cD9wAvEzYjl0qnl8AOAd4BfgP8BtgTMJlKO51FPdNd5k2aJ59cZlBk7eEGsVd2oOzgX8DFwKfinMh8K/4uXZAcVfcFfe8xX36lGj4pRdmzgrTc2li9ayTOwDnArtSXdxPBv4O7AysBVwHPMO8E2VrQXFX3E3e4l7smictzKvAZ6pM/0z8XDuguCvuinve4n7hudHwS6ZnzgoXnptHE2tUnawU9wUII+0nlE37CPAWsHeC91XcFXeTt7gXu+ZJC/N3qu+GXQl4o8nr0ijaUtzTZPthxyTOhO3OT5U0sn/j06snzisvLZs4eQtwI9KMLyFPPb50pzaxRtXJSnEfFU9bp2K+W4AvJXhfxV1xN4q7tCmXEnb5VnIB8NUmr0ujUNwVd8U9Z3EfNu3caMTF0zNn2LRcmlij6mSluI+Lpy1bMd8PgGsGeJ9FCD+PUoaiuCvuJldxL3jNkxZjWlkuAv4BPEhoQF8lnHT1BvDFvFawzijuirvinre4X3BuNOLL0zNn2AVNa2LNqJP1Evfu+HV9orgr7iZHcS9ezZMW5rYac2teK1hnFHfFXXFX3JPSjDpZr0NlHHFX3I3iLtI2KO6Ku+KuuLci/Z2cenzZtC48OVVxN4p7e9Q8kZpQ3BV3xT1ncR9+wbnRyC9Nz5zhxW9iixNG1NchbMex8f+Hxc+fDLwO7ASsCVyLl4NU3E3hxN2aJ5IexV1xV9wV91ZhIlWORwdmxs+XbsD0J8JI+28IV69JguJeR3HffrkjB81udxw6aGq9XG3esmoUd5G8UdwVd8U9b3E//7xo5EUzMmf4+efZxAZHcVfcTd7ibs0TSY3irrgr7op7J6G4K+5GcRfJzHNU30V8aT/zT6oy71splqu4K+6Ke97iPvW8aOQXZ2TO8Kk2sRpQ3BV3k7e4W/OkgYwCDgROAU6rSD1ZElimLFsTPowT+5l/EuE6yeWvWTrFchV3xV1xz1vce86LRvbOyJzhPbk1sWbVyXqguCvuJm9xL37NkxblQOC/wF+AR4CHy/JQg5d9EfAU4WSsakwi3Go8K4q74q64K+5ZyLNOpkFxV9yN4i5tynPkM2K0MPDXQZY9CXgXeB54EbgOWL2G965+MxLFXXFX3HMT9xHnnReNmjEjc0acl6qJHUGodW8Bs4GNBph3EtUP6WvFkfX+UNwVd5OzuOdc86SN+QdhF3Cz2ZMg5csNMM9Y4LOEaxxPAG4gHDqz/CDv3U21238r7ulkf/QJqbL5Jy5InGEzpybO9/+4fuI88vxyqfL8i8u0bJ58IV0ef2HZmnP3oxnE/dzzolHTZ2TOiHMTN7G9gLeBA4DVgCsI10lfqp/5JzH/IXr/JJ86mRbFXXE3eYt7fjVP2pxvAZ/LYbm/JIh4EhYiHFozZZD5HHFX3BV3xb3EbOCSssdDgLmEY9WrMYn5D9HLq06mRXFX3I3iLm3KScCrwNeBo4HDK9IIhgPvATuneO0Pge8nfI3HuCvuinve4j7lvFSHT1VmxJRETWxhwp69XSqmX0k49K4ak5j/EL0ZNL9OZkFxV9xN3uKeT82TDuDFAfJCg5bZDbwCfCDh6xYE5gC9CV+nuCvuinvO4j7ynPOi0dNmZM7Ic/6/iQ2l7561Raose7l43rEV06cRRuKrUe0QvfeBl2luncyC4l5Hcd9m4X0GzdqHzRg0L724TE156+WRgyZvoe2UZBH3BtQ8xV1yYQhhJOv8Ks9dBUwte3wWsC3h2NL1CCPt/yEcp5oExV1xV9zbT9wr011l2WnEvZJaD9FrJRR3xd0o7iJ1YVvCB3ClKs/NAmaWPf4iQfLfBv4E3Aism2KZirvirrjnLe6Te6LRF/RmzsjJPUlG3NMcKlONNIfo5YnirribvMW9/jVPcZf/Zx/gfuDfwJvAfcCnc12j+qK4K+6Ke97i3t0TjT6/N3NGdiduYrOBi8seDwFeov+TUyspHaL3c4pTJxV3xd3kLe751Txpc44hNKEZwKfi9BKa01E5rlc9UdwVd8W9c8V9L8L12/cHVgUuJ1wOsnQX5loO0XuHcJheUeqk4q64G8Vd2pRnCVdRqOQA4JnmrkrDUNwVd8U9Z3EfdXZPtOLU3swZdXaqJnYk8w67mw1sXPbcLAY/RG8uxaqTirvibnIW95xrnrQxbwMrVpk+hjBK1Q4o7oq74p63uJ/VE63Y05s5o87KpYkVrU4q7oq7yVvci13zpIV5lOrHep4KPNLkdWkUirvirrgr7lkoWp1U3BV3o7hLm7IH4aoLPyM0oVPj//8X2C3H9aonirvirrjnLe5n9kRjzuvNnFFn5tLEilYnFXfF3eQt7sWuedLibARcDTwY52pgw1zXqL4o7oq74p63uJ/RE405tzdzRp2RWxMrUp1U3Jss7tsve8SgOeLeT9eUf84dNmjyFtpOSSZxL37NE8kNxT2HbDfmxMQZv8MFiTP8yqmJ840nxqXK/c8vnzj3PrdC4jz0/NDESbNu9z+/fHTXc8Nqzm8fXt4mVgwUd8XdKO7SRnyw4v8DpR1Q3BV3xT1vcT+9JxozpTdzRp3etCZW5DqpuCvuJm9xL17NkxbmPWCp+P/vx4/7SzuguCvuinvO4j76tJ5opXN6M2f0aU1rYkWuk4q74m5yFvcC1jxpYbYCPhD/f+v4cX9pBxR3xV1xV9yTUuQ6qbgr7kZxlzZluX6mLzDAc0VDcVfcFXfFPQtFq5OKu+JuFHdpU8p3B5fzUVpzF3AaFHfFXXHPW9xP7YlWmtybOaNPzaWJFa1OKu6Ku8lb3Itd86SFeZ/qDWkY8O8mr0ujUNwVd8Vdcc9C0eqk4q64G8Vd2oxpcd4DLil7PA2YAdwB3Jnb2tUXxV1xV9xzFvcVT+mJVu7uzZwVT2lqEytqnVTcFXeTs7gXtOZJC3NbnPeBP5Q9vg24GfgGsHJua1dfFHfFXXFvBXE/uzdzmtzEilonFfcmi/t2yxw+aLYfdkxN8c6prZPM4l68micF4Nu0/4dBcVfcFXfFPQtFq5OKu+JuFHdpUxYHlqgyfYn4uXZAcVfcFfe8xf3knmjls3ozZ8WTc2liRauTirvibvIW92LXPGlhfg4cWWX64cDPmrwujUJxV9wV95zFfcxJPdEqZ/ZmzpiTcmliRauTirvibnIW94LXPGlh/gasVmX6qsBfm7wujUJxV9wVd8U9C0Wrk4q74m4Ud2lT/g2sUWX6GsCbTV6XRqG4K+6Ke97ifmJPtMoZvZkz5sRcmljR6qTirribvMW92DVPWphbgIuqTP8ycHuT16VRKO4FSRrZn7DN1MQZ9s3zU2Xyw59InBufXj1xfvH0qolz8zMrNTzXPzg6dRNb6YSeaNXTezNnpRNyaWJFq5OKu+Juchb3gtc8aWE2B94CfgucHue38bQJOa5XPVHcCxLFXXFv0SZWtDqpuCvuprPF/QjgOUKNmg1sVOPr9o6XdW2CZUkOrA9cAzwBPABcBayS4n3GAzcALxN+8btUPL8AcA7wCvAf4DfAmBreN+0HsITiXpAo7m0s7sf3RKue1ps5Kx2f2+hTvepkM1DcFXeTt7jnV/P2At4GDiCcm3MF8DrV7/5czgjgJeBWFPeOYQfgXGBXqov7ycDfgZ2BtYDrgGeARQd4z7QfwHIU94JEcW9jcT+uJ1r11N7MWem4wu82HmyAY2Y8vTw3JVyG4l5gca8leQttpySTuOdX82YT7vRcYggwFzhlgNcsSLgT9EGEGqS4F4SFgA9WJC2VDWkBwkj7CWXTPkIYRd97gPdJ8wGsRHEvSBR3xb0A4p61Tg42wDET+AWwTFn+J+EyFHfF3bSfuA+N37uURaose2HgXeavK1cSBkv7YzLw0/j/M1HcW5rFCCddvQy8VyVpqWxIo+Jp61TMdwvwpX7eI+0HcBH6friHorgXIop7+4r7ysf2RKud0ps5Kx+bi7g3q05CfZqm4q64m5zFvQE1rzLdVZa9XPzc2Irp0wgDodXYjHCIzMfixzNR3Fuai4E5hENS3gQOJnwY5gL7ZXjfyoY0Lp62bMV8PyAcN1qNNB9ACOs/34dccW/9KO5tLO7H9ESrndybOSsfk4u4N6tOQmiafwdeJRxPfxnw0YTvq7gr7iZvca9/zatlxD2pN30YeJawJ7DETBT3luYFYIv4//8EVoz/vz/Z7giYp7g74l7QKO6Ke4uKe7PqJIRDB3cC1oyfewy4i3AMan9Ur3mKu+Ju2knca1mHpEcqrBO/97tleT/Ou8DoBNstTeLfwLD4/3OBDeP/jwT+leF98zxUphKPcS9IFPc2Fveje6LVTurNnJWPzu0GTM2ok9Uo1c6tBpinm2p7GRV3xd3kJ+751bzZhL2EJYYQDoWpdm7gooQbyZXnWuDm+P8LJ9huaRIPE65RDOEXNS3+/xGEBpWW/k5OPb5sWhe1nZxa6wewPxT3gkRxb19xX+Wonmj1E3szZ5WjchH3ZtXJ/vgLcOgAzzvirribFhP3HGveXgS32h9YFbiccDW+pePnrwKmDvD6mXioTEtzAnB0/P9tCb/sNwknXB2X8L0WJ4yol3a9HBv/vzRSdTLhw1PaDXwt818O8mbgyLLHg30Aa0FxL0gUd8W9RcW9nnWynFrEfXnCbuudEryvx7gr7qZzxR2CRz1PuJz2bGDjsudmEeS8P2aiuBeKUcCewHopXjuR6mc+z4yfL92A6U+ExvcbYKWK93iO+c+UHugDWAuKe0GiuLexuH+hJ1r9hN7MWeULuV8OErLVyYEGOBYHLgQ2IdwMZSvgXuBJqp+I1h+Ku+Ju8hb39qp50iIsBPyS2u5eWmQU94JEcW9fcV/1yJ5ojeN7M2fVI5vexOpdJyfS/wDHYvGyXgXeIQxkXEGyPYyguCvuJndxL3DNkxbnr8y7QkK7ori3cdLI/sStp6bKsK9dkDinPrhr4nz3yQ0T5wd/XC9VfvzUOjXn2/ev2alNrGh1UnFvc3FX7hV36Vy+BJyX90o0GMW9jaO4F0Tcj+iJ1jiuN3NWPSKXJla0Oqm4K+65S287JJO4F7vmSQtzEfAG4djxSwlXSyhPO6C4t3EU94KI++E90RrH9mbOqofn0sSKVicVd8U9d+lth2QS92LXPGlhbhsgt+a4XvVEcW/jKO6KexMoWp1U3BX33KW3HaK4i+SD4t7GUdyLIe6rHd4TrXlsb+asZhOrBcVdcc9detshWcTdmif1ZhTh8oydgOLexlHcCyLuh/VEax7TmzmrHdbUJlbUOqm4K+65S287JJO4F7PmSQvzHrBU2eNrSH7JsaKguLdxFHfFvYEUtU4q7op77tLbDlHcpZV4n74N6Z+E0aV2RHFv4yjuxRD31T/fE611dG/mrP75VE3sCMI10d8inGC6UY2vez9eVukugkWpk4q74p679LZDsoh7zjVP2hDF3bRFFPeCiPuhPdFaR/VmzuqHJm5iexHuunwAsBrhhkav07f+VWNEvJzfo7i3bRR30zBxz6/mSZvyHrBk2eN/AiNzWpdGo7i3cRR3xX0QZgOXlD0eAswFThngNQsCdxAGOL5PX3EvQp1U3BV35V5xlzbjfeBG4Cdx/ku41fZPKtIOKO5tHMW9QOL+hd7MKWtiQ+P3LmWRKsteGHgX2KVi+pXAdQOs82Tgp4Q6+SLwMsWqk4q74q64t4K417fmKe4dzrdqTDuguLdxFPdiiPsah/REax/ZmzlrHPL/Tawy3VWWvVz83NiK6dMII/HV2Ax4CfgYoQb+EXieYtVJxV1xV9xzFvcG1DzFXToGxb2No7h3rLjXMuKeVNw/DDwL7FA2bSbzDpUpCoq74q64K+4ihUVxN32y3UonpUoq2b98WuIcd/8eiXPRY1ulymVzxtec3nvGpW9in+uJ1j6iN3PW+FyiJpb0UJl14vd+tyzvx3kXGJ1gu/NEcVfcFfe8xT2fmifSFijupk8U9+aL+5oH90TrHN6bOWsenOrk1IvLHg8hHApT7eTURYE1KnItcHP8/4UTbHeeKO6Ku+Kes7jnWPNECo/ibvpEce8ocd+LcP32/YFVgcsJl4Ms3UjpKmDqAK+fiYfKtG0U9/zluJWjuIvkg+Ju+kRxz0HcD+qJ1jmsN3PWPChVEzuScILp24QR+I3LnptFkPP+mIni3rZR3POX41ZOJnHPt+aJFBrF3fSJ4t58cV/rwJ5o3c/3Zs5aB9rEakBxV9wV95zF3Zonkh7F3fSJ4q64tzmKu+Ku3CvuIoVFcTd9orjnIO4H9ETrHtqbOWsdYBOrAcVdcVfc8xZ3a55IahR30yeKew7iPqknWveQ3sxZa5JNrAYUd8Vdcc9b3K15IqlR3E2fKO6Ke5ujuCvuirviLjIo44EbgJcJH7Lym54sBFwAPAz8O57nKsKdDQeim/lvbT4n4Xop7qZPFPfmi/va+/dE632uN3PW3t8mVgOKu+KuuOcs7tY8KQI7AOcCuzK/uH8E+DWwJ7AysAnhsmz3DPKe3cAjwDJl+VjC9VLcTZ8o7jmI+2d7ovUO7s2ctT9rE6sBxV1xV9zzFndrnhSMSnGvxobxfMMGmKcbeCDjuijupk8Ud8W9zVHcFXfFXXEXSUQt4r418D4Dfxi7mXdozTPAdxlY9AEWid+zlKEo7qYstLo7DwAAHIBJREFUinvzxX2d/c6L1j9oRuass995NrHBUdwVd8U9Z3G35knRGEzcFwXuJYj4QOwA7AGsBWwH3Em4A+KHB3hNN/MfF6+4m8xJJftbTU2cNLJ/wgO7p0rTxH3f86L1D5yROevsaxOrAcVdcVfc8xZ3a54UjIHEfSHgeuA+kn8QlwDeAA4aYB5H3E1Dorgr7gVBcVfcFXzFXSQR/Yn7QsBPgQeBj6Z877uBqQnm9xh3U5co7unFfd3PnBdtcMCMzFn3MzaxGlDcFXfFPWdxt+ZJ0agm7iVpfwRYMuX7Lg78DTgqwWsUd1OXKO4ZxH2f86INJs3InHX3sYnVgOKuuCvueYu7NU8KwOLAOnEi4Nj4/8MI0n4d8CKwNn0v77hw2XvcDBxZ9ng6MAEYAYwjXFLyLyQTf8Xd1CWKu+JeEBR3xV1xV9xFBmUiVU4KBWYSxLvac1H8uhLPEU4uLXE14YoybwMvxY9HJ1wvxd3UJYp7enFf79PnRRvuPyNz1vu0TawGFHfFXXHPWdyteSLpUdxNXaK4ZxD3vc+LNvzsjMxZb2+bWA0o7oq74p63uFvzRFKjuJu6RHFX3AuC4q64K+6Ku0hhUdxNXaK4pxf39fc6L9povxmZs/5eNrEaUNwVd8U9Z3G35omkR3E3dYninkHc9zwv2mjfGZmz/p42sRpQ3BV3xT1vcbfmiaRGcTd1ieKuuNeBUwn3ovgn8CpwLbByxTyLApcCrwH/An4MLJ1gGYq74q64K+4ihUVxN3WJ4p5N3Df+zIzMaYMmdhMwCVidcGncG4HngQ+VzXMZ8AKwJbA+8HvgjgTLUNwV95ZMLbL8zsujBs1bL49sSl59Yrg1TyQHFHdTlyjuGcR9j3OjjfeZnjnr73FuuzWxJQnbMz5+/BHgHWD3snlWiefZpMb3VNwV95ZMR4m7NU8kNYq7qUsUd8W9AaxI2J414sdbxo+XqJjvecJN7WpBcVfcWzKKuzVPpBYUd5NbWln2h10+LTr0nn1rzgGz9kzdxDbY/dxok09Pz5wNdm+rJjYE+Blwe9m0fQg3nKvkLuCCft5nEcLPo5ShKO6Kewumk8TdmieSHsXd5BbFPW5iu50bbbL39MzZYLe2amKXEe4WvXzZtDTi3k2Vu1Ir7op7q6WjxN2aJ5Iaxd3kFsXdJtYPlwAvAiMrpqc5VMYRd8W9EFHcO7rmidSM4m5yi+Ie/v423PXcaOye0zNnw10L38QWIEj7XGBMledLJ6fuVjZtZTw5VXFvg3SSuFvzRNKjuJvcorjHTWyXc6Oxe0zPnA13KXwT+wrwd2ACsExZFiub5zLCCPsWhMtB3hmnVhR3xb0l01Hibs0TSY3ibnKL4m4Tq2C+Y9HjTCqbp3QDpr8B/wZ+QpD7WlHcFfeWjOLekTVPJDGKu8ktinv4+9to53OjcbtPz5yNdraJ1YDirrgXNrUI9Rtzl68pf35p2UHz/IvL9JtHHlvKmieSA4q7yS2Ke9zEdpoSjdvtwszZaKcpNrHBUdwV98KmbcTdmieSGsXd5BbF3SaWA4q74l7YKO7WPBHF3eQWxT38/W38ySnRpp+6MHM2/qRNrAYUd8W9sGkXcbfmiaRHcTe5RXGPm9gnpkSb7nph5mz8CZtYDSjuinth0zbibs0TSY3ibnKL4m4TywHFXXEvbBR3a55IFxBttsAno4lDPmVMU7PV6KMTZ7MJZyfO8hedkyoHzNqz5nzmxl1TN7FNPj4l2myXCzNnk4/bxGpAcVfcC5t2EXdrnkh6htL/tZONMckzlNoJTWzHc6LNdp6WOZvseE5pHWxi/aO4K+6FTduIuzVPJDULEESjqyIloa/2XCfFn4M/hyQ/i6GEv6la6SL/JnYE8BzwFjAb2GiAeT8F3EO4u+m/gQeA/RIsqxXoAqLN2DGayM5mgGzxgd0Gz0J7DJqtljp40Gy9/GGmhrz6xPBB8+Kc5WrKU48vPWgeeWypfvOHuz6muIu0EF34xwD+HEr4c5hHPX8WXUA0dodzos13mpY5Y3dI3MT2At4GDgBWA64AXgeW6mf+icCuwKrAaOBo4F1gu7Q/gBxwL6Mx9U3ivYw51jyRtkVRC/hzCPhzmEf9xX37c6LNPzktc8Zun7iJzQYuKXs8BJgLnJJgG+4DpiSYP28q9zIWdW+S6+16t8J6p9rLmGPNE2lbuvCPAfw5lPDnMI96/izybGILE0bLd6mYfiVwXQ2vXwDYinDIzDa1b3LLUdTPtuvdXFzv+qC4izSIRYDu+N9Oxp9DwJ/DPOr5s+gConHbnRON/8S0zBm33Tnlu67LR8Wqrety8bxjK6ZPI4zE98dHgH8B/yUcF39gpp9A/rSa2NSK691cXO/60Kia1yrbJyIibUxoYttMjsbveEHmjNtmcn/HoHZXWXZacR8CrAisAxxPOFF1YvofQe60mtjUiuvdXFzv+tComtcq2yciIm1Mo5pYLSPuWQ+VKfF14JdJN7yFKOreJNe7ubje9UFxFxGRwhKa2NaTo/E7XJA547ZO3MRmAxeXPR4CvESyk1O/CcxKML+IdC551zwREZHUdAHRpltPjiZsf0HmbJq8ie1FOE59f8IlHi8nXA5y6fj5q4CpZfOfSjgRdVQ8//GEY90PzvhzEJHOIO+aJ9K2JLkpSzvSzfzHCc/Jc4WaxHjgBuBlwjZXHkaxAHAO8ArwH+A3wJhmrmCTGOznMJP5Px83pVhOKzSxI4HnCddznw1sXPbcrHhbS5wL/JHwu/8bcCdB/kVEaqEVap5I25H0piztSDfwCLBMWT6W5wo1iR0IcrYr1YX1ZMLJiDsDaxGOhX4GWLSJ69gMBvs5zAR+Qd/Px/+kWE5oYlt1RxO2Oz9zNt2q2yYmIq1MK9S8JAOTnwNuIzjQ64TBqk4byJQCUI+bshSdbsLt3DuZSmFdgDDSfkLZtI8Qit/eTVyvZtOfuF9bh/fuAqLNtuyOJm57fuZstqXinoKi7V3sphh7A4u6965Ze9vqyanA3cA/gVcJtWnlinkWBS4FXiNczvXHzDskrpnkXfOSDkx+FziccBWtVYBvEQavktwtVqSh1OtKE0Wnm3BjmZcJI8rfBYbluUI5UNm0RsXT1qmY7xbgS81aqRzor3n/ndAknwAuAz6a4r3zbmKdThH3LnZTjL2BRd1716y9bfXkJmASsDqwNnAj4fC3D5XNcxnwArAlsD7we+COpq5lIO+al3VgckHgH8Bna5xfpOGkvbZzu7EDsAehoWxHOJb3eeDDea5Uk6lsWuPiactWzPcD4JpmrVQOVGveewM7AWvGzz0G3EUo6kkITWyL7mjiNudnzmZbKO4JKeLexW6KtzewqHvvGrm3rZEsSVj38fHjjwDvALuXzbNKPM8mzV21XGtePQYmP0zYQ/SJGucXaTiKe3WWAN4ADsp7RZqI4h6o1rwrKe2N2Crhe3cB0eYTz4622Hpq5mw+8WzFvXaKunexm+LtDSzq3rtG7m1rJCsS1n2N+PGW8eMlKuZ7Hji2iesFjat5jbxbdDlfAZ4m/z1DIv9PUZtZM7ibvpfGa3eK2mzrTS3iDvAX4NCE762450dRBymKuDewqIMAjdzb1iiGAD8Dbi+btg/hkLBK7gIuaMZKldGomtfIu0WXOIVwNa210m26SOOox01Z2o3FCX+wR+W9Ik2kv93bx5dN66L1dm/Xm1rEfXngfUJDT0JoYuPPjrbYcmrmbD5ecU9AUcW9kiLsDWwnca8k7d62RnEZ4WTr5cumtZ6417/mNfpu0ScQ9rRskGHbRRrGYDdl6QSmAxOAEYQm82vCiOqSOa5TM1icMKK+DqEYHhv/v7Qr/mTCZ6E04nQtrXFCWb0Z6OewOHAh4djQEYSGfS/wJMlvK94FROM3PyvacouezBm/+VmKe+20097FVt8bWNS9d43c29YILgFeBEZWTG+5Q2VyrHlpBiZPInw5bvb5ACKJGOimLJ3A1YRjSN8m/FFfDYzOdY2aw0Sq73acGT9fuoTbnwhf7n4DrNTslWwCE+n/57AY8EvCMa7vEEa3riDdF9u8m1in0w57F4uwN7Coe+8aubetnixAkPa5VL+kZunk1N3Kpq1Mjien5ljzkt4t+mSCB+xG3ysJLZ7x5yAiIgUkNLHNzoq2nNiTOeM3U9wTUsS9i0XZG1jUvXfN2ttWT75COIxjAn3lcrGyeS4jDMZtQbgc5J1xmk0r1Lwkd4t+jtqPoRcRkTYnNLFNz4y2nHBe5ozf9EzFPTlF27tYlL2BEynm3ruJNGdvWz3p7wTNSWXzlG7A9DfCVYl+QpD7ZmPNExGRwmITE5FOwponIiKFpQuIJow7M9pq/HmZM2GcTUxEWhprnoiIFJbQxMaeEW21+bmZM2HsGTYxEWllrHkiIlJYbGIi0klY80REpLB0AdHEjc+Itt703MyZuLFNTERaGmueiIgUltDENjo92nrclMyZuNHpNjERaWWseSIiUlhsYiLSSVjzRGRAJhFuTNEpfJRwjeERTV7uDELx/AmwYIOW0Q08UOf33D5+zyF1ft9aCU1sw9OjrcdOyZyJG9rERKSlseaJtAAzCXe9a0UWA5bKeyUGYASh6KxTp/frBb5WMe3LhLv7vc3A4jsBeDH+/6nA3cA/CV8EriXcIrsap8XzHUK4jXnl8utFN/UXdwjbuV8D3rcWuoBoiw1Oi7bZ5JzM2WKD02xiItLKWPNEWoCZNF/cF27y8pJS6/qNoH7i/kHgDcLtuMv5MnAEcBUDi28v4dbZADcR9lasDqwN3Ei4u+SHKl5zCOFOfKW7TY4BXgCmptmAQeimMeJ+BEHe88AmJvVmEu5lbAbuZUyHNU+kBZjJwOK+BPB14C/AP4DfEmSwxGjgOuDPwL8IErV1xXs8B5xJkM9/xMscQfiD/RTwO+BN4EFgbNnrJtG3iXUTitZ+8Xu+Qbhl+IfL5vkw8F3CbaFfAY4FZgEXDbCNpfc9GHgWeD+evj1we7wOrwE/o+/tyStvUT2r7LmDgccJtwOfAxw+wPIBdic0sMHWsT+eite3GkvG6ze+Ynmv0Pd3CTAM+CNwXPy4C/gPsEPFfLsSRuo/GD++AHiS8Ht8BpgCLDTA+s9i/t/Jtcy7pTrAIsB0YC7h9zmbcEvzyvWNyOe28aGJrX9qtM3GkzNni/VPtYk1h5m4lzEtI3AvY6100657Ga15Irkyk4Gb2K+B64ENCCOy04G/Av8bP782cCiwRvz8FILoDSt7j+cIkn08QbBGM68BPA58nP9r7+6D5arrO46/CSEJkNwQDCEkAgkhgYQECPJQGByTe5d7d6APCkJ4KCQxaCCEEExIIqBexmkd849TpRawjutTO1Rr60On06njxLE63k6n1WqrTrWNDCj9w6FV22LEOf3je+7k7Lm7e+/dk3vvOez7NfMb2HPOnj1n9+b7+e3vt3sW1gCfTrednd5vG2M77j8H/jx9vNcThff3Mtt8ON3HQLrNZ4k3C+N13H8B/DWwEbg0XX4L8cbiQiKoPg/8M8dHO65Kz2EAWJp5Tu4Cfpzed2X6358CWzscwx+kj9/pGNuFwCXEObabKbgwPc71HfbfyaeBT+SWfSa37HHgOuJ1/S3gReBAZv0wk++4fxj4GvE6rwL2E2+EVufu9yLxtzLdIsSueEdyw1VPFG6brzDEpkkDZxnznGU88YZ5tc4yWvOkGdWgfYhdTxTWubnlPyAKYDvfAXZnbh8F/iK3zQriH+yOzLJ16bKL09vbGNtx/x+aR9gPA99I/38BcIwYTR61ML3PeB33Y8TIdCeLae4Aj55DPsR+ANyRW/Y48PUO+/5L4CPjHGO7EHiU6Fy3MouYKfi7DvsezxtpHl0fHYVvN8IP0cn+h8ztYSbXcT8PeAVYltvmS8Dv55b9I/DuDscyVQyxamrgLOPofp1ldJZxMqx5Ugk0aB9iDwC/JsIp235NFC2A+USh+S5R7EfXH87s5yjwWG7fK4h/sFdlli2iudhuY2yI/UtuPw8TRROiICc0j/ZDdOzGC7F/a7F8NfCn6f5/RpxbAtyYO4dsx/30dNn/0vycvUwEfTt/A/zhOMfYLsRGaD91+kfE8//aDvsezxxilOr29PZ24lxmZ7bZQoyOv8jx882G8jCTC7GbiOcx/7f3K+DZ3P2+xvG/x+nUByT9Gw8lg1cOF279Gw8ZYtOjgbOMwzjL2ImzjK1Z86QSaNA+xA4CzxNFMN8Wp9s8BfyQGI3YkK77Js0F6iiwN7fvFYzt9J6RLtuU3t5G69GnrL3p/qFYx71VQHyP6FAPAGuJsEiIEeh253B2uuwuxj5nKzscw6eAP+niGM8hZgvObLHuSeIzoJ0ed6KeIUIcomPzgcy6a4nR8cc43tl5J51fuy8TwZ31VxwPsS3pPi9i7PO4NHe/7xIBN90ixC4/lAy+7t2FW//lhtg0aeAs4zDOMnbiLGNr1jypBBq0D7EbiEKyosP9v0100kbNJ4JnJjruoyF2S2b9QmJkabId99ekx/L6zLLrae64L0tvvy533xdofk4mYn+LYxjvGCE6E0dyy04iOu0vMHakpltvIJ7bS4gZlWsy6/YRb96y/pjOr92zwJ9lbp9MfCa1kd5ew9jnv5V56XENjLPdVDDEqqmBs4zDOMvYibOMrVnzpBJoEJ+3vDzXziU6gF8lis8gUbSvI6Zpr0zv/1ngn9L7XEaMyuanaY8yPR13iKnGfwc2E53Mz6TH8/78iY+z31nE9PgniFHefuDvae64zybC6jFipH1huvzedPkeogO6gSj8o5+hbGUDUaAX5ZaPTlk/BXyf46/P6BTxF1vs90PE8/YGYnR6tJ3a4fHHcxLxJa5vEqNrWb+dHvvtxPTuHmKavNNrt5MYFbyJGG18hhjpbGS2+STx+dvR6feriatH3JTZZhPNI2PTKULssoPJ4BXvKtz6LztoiE2PBs4yttovOMuY5SzjWNY8qQQajP3CUUKMmEKMYn+AGL09RnTePkl07CEK+ZeJjupzxIjVEWau497qi1ojdL5qQKv9Qnzh7F+JkZRvER3hbIhBdNKfI0bcjmSW30m8ofklMXLzFSLoOxkhOrRZR2j9+qwgRrr+jyjsWa22Tyj+mcj3pft5osW6w8QbnZ8TX57bS+fX7hTiDcZPiZGsQ4z9vOcp6WP9B/G392PijeKGzDZPEx2pmRAhtuFgMnj5uwq3/g2G2DRp4Cxjq/06y9jMWcaxrHmSptzpRDHdMd6GJXAT8UZhoj+ucXO6fa9aTHT8T8ToWjcMsWpq4Cxjq/06y9jMWcaxrHmSTriNxJekVgFXECNr/8Xxae6y28vx2YzxDBJXM+hVVxLTyzOlD0gG1h9Ihi57Z+E2sP6AITY9GjjL2Gq/4CxjnrOMzax5kk64jcSv7v2CCI+/pbnoSSdKhNgljyRDlz5euA1c8oghphPBWcZXr1LMMlrzJElVZIipDJxl7B3lmGW05kmSKihCbN3+ZGjDY4XbwLr9hpi64Syjpos1T5JUWRFia/clQ+sfLdwG1u4zxCSVmTVPklRZhpikXmLNkyRVVoTYRfuSoXWPFm4DFxlikkrNmidJqqw+IKmteXtSX/uOwq225u2GmKQys+ZJkiqrDCH2AHFN75eJa2Jf3WHbtxI/LvRS2r40zvaSlFWGmidJUlcixFY/nNQvPlS41VY/PNkQ20L82M12YB3xS4wvAUvabP8pYBfxA0AXAx8lLhu4vOtnQFIvmemaJ0lS1yLEVu1N6msOFm61VXsnG2IjwJOZ27OIX/s8NMH7nwz8DLhngttL6m0zXfMkSeraVIXY8nTfo21ui8eeA7xC80/RA3wM+NwEj38B8fPxv9nFuUvqPXbcJUmVFSF2wUNJffWBwq12wUOjIZZvwy0ee1m67trc8sPESPxEfAj4ITBv8qcuqQdNVc2z4y5JmnJpiO1J6hc+UrjVLtgzmRH3oh33Q8SvbF7a5blL6j1TVfPsuEuSptxMhliRj8rsJ76UeuUkz1dSb7PjLkmqrAixlQ8m9VX7C7fayge7+XLqBzO3ZwHP0/nLqQeA/wZ+o4vzldTbZrrmSZLUtQix83cn9ZX7Crfa+bu7uRzky8BWYC3wNHE5yLPT9R8H3pvZ/iBx+chbgKWZNr/g8yCpN8x0zZMkqWtlCLHdwI+IDvkIcE1m3RGgkbl9lIl/+VWS8spQ8yRJ6kqE2Hm7kvqKhwu32nm7DDFJZWbNkyRVVoTYufcn9fP3Fm61c+83xCSVmTVPklRZhpikXmLNkyRVVoTY8vuS+rkPFW615fcZYpLKzJonSaqsCLFlO5P6a/cUbrVlOw0xSWVmzZMkVZYhJqmXWPMkSZUVIXbOzqS+/MHCrXaOISap1Kx5kqTKihBb+rakvmx34VZb+jZDTFKZWfMkSZVliEnqJdY8SVJlRYgtuTepL91VuNWW3GuISSoza54kqbIixM7akdTPvr9wq521wxCTVGbWPElSZRliknqJNU+SVFkRYovfktSX3Fe41Ra/xRCTVGbWPElSZUWInbk9qS/eWbjVztxuiEkqM2ueJKmyDDFJvcSaJ0mqrD4gGVi0NRl6zVsLt4FFWw0xSWVmzZMkVVaE2Bn3JEOL7i3cBs64xxCTVGbWPElSZRliknqJNU+SVFkRYgvvTobO2FG4DSy82xCTVGbWPElSZUWILbgrGerbXrgNLLjLEJNUZtY8SVJlGWKSeok1T5JUWRFi8+9MhhZsK9wG5t9piEkqM2ueJKmy+oCk/7Tbk8HT7ync+k+73RCTVGbWPElSZRliknqJNU+SVFkRYqduSQZPu7tw6z91iyEmqcyseZKkyooQm3tbMjjvdwu3/rm3GWKSysyaJ0mqLENMUi+x5kmSKitCbM6tyeDcOwu3/jm3GmKSysyaJ0mqrD4g2Tz7zckNp9xRuG2e/WZDTFKZWfMkSZVliEnqJWWoeQ8AR4GXgRHg6nG2vxX4Xrr9t4EbJ3vSkqRXhwixk29Obpi9pXDbfPLNdtwlldlM17wtwC+B7cA64BngJWBJm+2vA14BHgHWAu8BjgHru34GJEmV1Qckm056U1KbdVvhtumkN9lxl1RmM13zRoAnM7dnAS8Ah9ps/yzwxdyybwBPTfyUJUmvFjMdYpI0naaq5i1P9z3a5rZ47DnE6Pkbc8s/BnyuzfE+B+zNLXsC+FZ3py9JqrI+ILmeG5NN/E7hdj032nGXVGZTVfPybbjFYy9L112bW36YGIlv5RhwR27ZLuA/uzp7SVKlzQN+Quvg6bb9JN2vJJXNVNW8sxh/xN2OuySpsHk0B07RZqddUpnNVM3zozKSJElSRYwAH8zcngU8T+cvp34ht+zr+OVUSZIkaUptIa7HvpW4vOPTxOUgz07Xfxx4b2b764BfAfuAi4nPzns5SEmSJGka7AZ+RFzPfQS4JrPuCNDIbX8r8P10++/gDzBJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkkrv/wHOfL7hvaY5fwAAAABJRU5ErkJggg==\" width=\"750\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fixed_bw = load_object(\"mnist_accuracies.pickle\")\n",
    "float_bw = load_object(\"mnist_bw_float_accuracies.pickle\")\n",
    "\n",
    "fig = plt.figure(figsize=(7.5, 5))\n",
    "\n",
    "# left figure\n",
    "ax1 = fig.add_subplot(121)\n",
    "surface = ax1.imshow(fixed_bw[0])\n",
    "ax1.set_xlabel(\"Learning rate (1/2^value)\")\n",
    "ax1.set_ylabel(\"Fraction bitwidth\")\n",
    "#plt.xticks(np.arange(np.min(fixed_bw[1]), np.max(fixed_bw[1])+1, 1.0))\n",
    "ax1.set_title(\"Acc. when training with ints\")\n",
    "cbar = fig.colorbar(surface)\n",
    "\n",
    "# right figure\n",
    "ax2 = fig.add_subplot(122)\n",
    "surface = ax2.imshow(float_bw[0])\n",
    "ax2.set_xlabel(\"Learning rate (1/2^value)\")\n",
    "ax2.set_ylabel(\"Fraction bitwidth\")\n",
    "#plt.xticks(np.arange(np.min(fixed_bw[1]), np.max(fixed_bw[1])+1, 1.0))\n",
    "ax2.set_title(\"Acc. when training with floats\")\n",
    "cbar = fig.colorbar(surface)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.],\n",
       "       [  8.,   9.,  10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,\n",
       "         19.,  20.,  21.,  22.,  23.]])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "float_bw[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}