analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Estimando Aluguel com Redes Neurais"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introdução\n",
    "\n",
    "Apesar da população, Florianópolis é uma cidade vasta e esparsa, com vários bairros que funcionam quase independentemente uns dos outros. Além disso, aluguel na cidade pode variar muito dependendo da região. Esse projeto pretende criar um modelo para que, dado as características de um apartamento, possamos estimar o valor do aluguel.\n",
    "\n",
    "O projeto tem como segunda finalidade estudar e exemplificar o uso de redes neurais usando Python.\n",
    "\n",
    "\n",
    "## Coleta de Dados\n",
    "\n",
    "Os dados foram coletados a partir do site **VivaReal.com.br**, usando os scripts encontrados em `downloaddata.py`. Para garantir reproducibilidade, os dados usados nessa análise estão salvos no arquivo `rentals.json`. Ao total, foram coletadas informações de 3102 vagas de aluguel na cidade de Florianópolis.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Limpandos os dados\n",
    "\n",
    "Ao longo do projeto, usaremos Pandas para visualizar e limpar dados, combinado à biblioteca `GeoPandas` para visualização de mapas. O primeiro passo do processo é carregar o json com os dados:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Colunas disponíveis: account, accountLicenseNumber, accountLogo, accountName, accountPagePath, accountRole, additionalFeatures, address, area, areaUnit, backgroundImage, baseRentPrice, baseSalePrice, bathrooms, businessId, businessName, cityName, cityUrl, condominiumPrice, constructionStatus, contactAddress, contactCellPhoneNumber, contactLogo, contactName, contactPhoneNumber, countryName, countryUrl, creationDate, currency, currencySymbol, developmentInformation, email, externalId, garages, geoDistance, image, images, iptu, isDevelopmentUnit, latitude, leadEmails, legend, links, listingType, locationId, longitude, neighborhoodName, neighborhoodUrl, numImages, positioning, price, priceValue, promotions, propertyId, propertyTypeId, propertyTypeName, publicationType, rentPeriod, rentPeriodId, rentPrice, rooms, salePrice, saved, showAddress, siteUrl, stateName, stateUrl, suites, thumbnail, thumbnails, title, usageId, usageName, video, zipCode, zoneName, zoneUrl\n"
     ]
    }
   ],
   "source": [
    "# Setup\n",
    "import json \n",
    "import matplotlib.pyplot as plt\n",
    "import geopandas as geo\n",
    "import numpy as np\n",
    "from shapely.geometry import Point\n",
    "import pandas as pd\n",
    "\n",
    "\n",
    "%matplotlib notebook\n",
    "\n",
    "data = pd.read_json(\"rentals.json\")\n",
    "\n",
    "print(\"Colunas disponíveis: \"+ \", \".join(data.columns))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ao todo possuímos 77 colunas com diversas informações (inclusive imagens das vagas), porém apenas algumas são necessárias. Além disso, queremos tratar latitude e longitude como pontos no plano cartesiano."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/bonet/.virtualenvs/data-python3.5/lib/python3.5/site-packages/ipykernel/__main__.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    }
   ],
   "source": [
    "data = data[['area','additionalFeatures','priceValue','bathrooms','condominiumPrice',\n",
    "             'garages','iptu','latitude', 'longitude', 'neighborhoodName', 'rentPeriodId', \n",
    "             'rooms', 'suites', 'usageId']]\n",
    "data['coord'] = data.apply(lambda x: Point(float(x['longitude']), float(x['latitude'])), 1)\n",
    "data = geo.GeoDataFrame(data).set_geometry('coord')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Queremos analisar apenas alugueis mensais. Ao verificarmos os dados, podemos ver que existem entradas marcadas como mensais porém com alugueis muito baixos, indicando possível erro no cadastro. Alguns valores de latitude e longitude também estão incorretos. Por isso, precisamos fazer uma triagem do nosso conjunto de dados. De 3100 entradas passamos agora a ter apenas 1228."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"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",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "count    1228.000000\n",
       "mean     1631.399023\n",
       "std       770.010345\n",
       "min       600.000000\n",
       "25%      1100.000000\n",
       "50%      1450.000000\n",
       "75%      1900.000000\n",
       "max      5000.000000\n",
       "Name: priceValue, dtype: float64"
      ]
     },
     "execution_count": 254,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = data[data.priceValue.between(600, 5000) ]\n",
    "data = data[data.bathrooms > 0 ]\n",
    "data = data[data.rooms > 0 ]\n",
    "data = data[data.rentPeriodId == 'MON' ]\n",
    "data = data[data.usageId == 'RESIDENCIAL' ]\n",
    "\n",
    "data['priceValue'].hist(bins=50)\n",
    "plt.title(\"Distribuição de valores de aluguel\")\n",
    "plt.xlabel(\"Valor (R$)\")\n",
    "\n",
    "\n",
    "data.priceValue.describe()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Para visualizarmos melhor, podemos plotar as vagas em cima do mapa de Florianópolis. As cores representam o preço listado do aluguel (veremlho representa valores altos). Os dados cartográficos foram obtidos no site do Inep seguindo atrabés desse [link](ftp://geoftp.ibge.gov.br/cartas_e_mapas/bases_cartograficas_continuas/bc250/versao2015/Shapefile/Limites_v2015.zip)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"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",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x109a27198>"
      ]
     },
     "execution_count": 255,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "geoFloripa = geo.GeoDataFrame.from_file(\"geodata/floripa/floripa.shp\")\n",
    "\n",
    "floripaHull = geoFloripa.convex_hull\n",
    "\n",
    "data =  data[data.coord.map(lambda x: floripaHull.contains(x).bool())]\n",
    "ax1 = geoFloripa.plot(color=\"white\")\n",
    "\n",
    "data.plot(ax=ax1, column='priceValue', markersize=5, cmap='coolwarm')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Redes Neurais Simples\n",
    "\n",
    "Nossa primeira tentativa será usando de redes neurais simples, com funções disponibilizadas pela biblioteca [Scikit-learn](scikit-learn.org). Apesar de termos diversas informações disponíveis, usaremos apenas 5 delas: número de quartos, número de suítes, número de garagens, area e número de banheiros.\n",
    "\n",
    "Como alguns imóveis foram cadastrados erroneamente, alguns dados estão em branco. Trocaremos esses valores faltantes por 0. Usaremos 80% dos dados para treino e o restante para testes.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/bonet/.virtualenvs/data-python3.5/lib/python3.5/site-packages/sklearn/utils/validation.py:428: DataConversionWarning: Data with input dtype int64 was converted to float64 by the scale function.\n",
      "  warnings.warn(msg, _DataConversionWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neural_network import MLPRegressor as mlp\n",
    "from sklearn.preprocessing import scale\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "model_features = ['rooms','suites','garages', 'area','bathrooms'] \n",
    "\n",
    "x = scale(data[model_features].fillna(0.0))\n",
    "\n",
    "y = scale(data['priceValue'])\n",
    "\n",
    "x_train, x_test , y_train, y_test = train_test_split(x, y,\n",
    "                                                     train_size=0.8,\n",
    "                                                     random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nossa rede terá duas camadas de 10 neurônios cada uma."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MLPRegressor(activation='relu', algorithm='adam', alpha=0.0001,\n",
       "       batch_size='auto', beta_1=0.9, beta_2=0.999, early_stopping=False,\n",
       "       epsilon=1e-08, hidden_layer_sizes=(10, 10),\n",
       "       learning_rate='constant', learning_rate_init=0.001, max_iter=2000,\n",
       "       momentum=0.9, nesterovs_momentum=True, power_t=0.5,\n",
       "       random_state=None, shuffle=True, tol=0.0001,\n",
       "       validation_fraction=0.1, verbose=False, warm_start=False)"
      ]
     },
     "execution_count": 258,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nn = mlp(hidden_layer_sizes=(10,10), max_iter=2000)\n",
    "\n",
    "nn.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Obtemos os seguintes resultados para a rede treinada:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R2 para treino =0.5444\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"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",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"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",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from  scipy import stats\n",
    "\n",
    "pred_train = nn.predict(x_train)\n",
    "print(\"R2 para treino =%.4f\" % stats.linregress(pred_train,y_train).rvalue **2)\n",
    "\n",
    "plt.figure()\n",
    "plt.hist(abs(pred_train-y_train), bins=50)\n",
    "plt.title(\"Erro absoluto no preço \")\n",
    "plt.xlabel(\"Erro absoluto (dados escalados com média 0 e desvio padrão 1)\")\n",
    "plt.show()\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(y_train,train[:,5], c=abs(y_train-train[:,5]))\n",
    "plt.plot([-1.5,3], [-1.5,3], 'k-')\n",
    "plt.title(\"Resultado da rede neural simples\")\n",
    "plt.xlabel(\"Valor estimado\")\n",
    "plt.xlabel(\"Valor real\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Redes Neurais Clusterizadas\n",
    "\n",
    "Como vimos, usar apenas atributos do apartamento como entradas da rede neural não rendeu bons resultados. Isso por que um mesmo apartamento com a mesma área e mesmo número de quartos podem ter uma diferença considerável no valor dependendo de sua localização.\n",
    "\n",
    "Lidaremos com através do uso de múltiplas redes neurais. Dividiremos os dados de teste em 10 clusters, baseado na latitude, longitude e valor do aluguel. Treinaremos uma ANN para cada um destes clusters.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Criando os clusters\n",
    "\n",
    "O primeiro passo será agrupar os dados, onde usaremos KMeans. Novamente, a biblioteca sklearn nos fornece uma implementação simples do algoritmo. Usaremos como propriedas para agrupamento o preço e as coordenadas, a fim de identificar bairros normalmente mais caros. Essa é apenas uma das heurísticas possíveis. Porém, para o caso de testes, removeremos o preço do vetor que representa a coordenada do imóvel.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn.preprocessing import StandardScaler, MinMaxScaler\n",
    "\n",
    "# Reproduscibilidade\n",
    "random_state = 42\n",
    "\n",
    "# Divide em conjuntos de teste/treino\n",
    "\n",
    "km_train, km_test = train_test_split(\n",
    "                       data,\n",
    "                       train_size = 0.7,\n",
    "                       random_state=random_state)\n",
    "\n",
    "# Propriedads das coordendadas\n",
    "coord_properties = ['longitude', 'latitude']\n",
    "\n",
    "# Normaliza\n",
    "coord_scaler = MinMaxScaler()\n",
    "coords = coord_scaler.fit_transform(km_train[coord_properties])\n",
    "\n",
    "\n",
    "price_scaler = MinMaxScaler()\n",
    "prices = price_scaler.fit_transform(km_train[['priceValue']])\n",
    "\n",
    "# Calcula os centros\n",
    "centers = KMeans(n_clusters=10, random_state=42).fit(np.c_[coords, prices])\n",
    "\n",
    "\n",
    "# Remove o preço das coordenadas dos centros\n",
    "coord_centers = coord_scaler.inverse_transform(centers.cluster_centers_[:,[0,1]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "* Identificação dos clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "scaler.get_params()\n",
    "\n",
    "nn_train_coords = km_train[coord_properties]\n",
    "\n",
    "def euclidean_distance(A,b):\n",
    "    return np.sum((A-b)**2, axis=1)**0.5\n",
    "\n",
    "def predict_kmean(centers, X):\n",
    "    return np.apply_along_axis(lambda x:np.argmin(euclidean_distance(centers,x)), 1, X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos visualizar agora os clusters de dados:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"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",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "geoFloripa.plot(color=\"white\")\n",
    "\n",
    "plt.scatter(coord_centers[:, 0], coord_centers[:, 1],\n",
    "            marker='x', s=169, linewidths=3,\n",
    "            zorder=10,\n",
    "            c = np.arange(10))\n",
    "\n",
    "plt.scatter(np.array(km_train.longitude[:]), \n",
    "            np.array(km_train.latitude[:]),\n",
    "            c = clusters)\n",
    "\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O último passo será treinar nossas redes neurais (uma para cada cluster). A função `train_clustered_nn` faz exatamente isso, retornando um array de redes neurais"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Treinando rede neural para o cluster 0\n",
      "Treinando rede neural para o cluster 1\n",
      "Treinando rede neural para o cluster 2\n",
      "Treinando rede neural para o cluster 3\n",
      "Treinando rede neural para o cluster 4\n",
      "Treinando rede neural para o cluster 5\n",
      "Treinando rede neural para o cluster 6\n",
      "Treinando rede neural para o cluster 7\n",
      "Treinando rede neural para o cluster 8\n",
      "Treinando rede neural para o cluster 9\n",
      "Treinamento Finalizado! \n"
     ]
    }
   ],
   "source": [
    "def train_clustered_nn(X, y, Xcoord, centers):\n",
    "    \"\"\"\n",
    "    Treina uma rede neural clusterizada\n",
    "    \"\"\"\n",
    "    \n",
    "    n_clusters = len(centers)\n",
    "    \n",
    "    cluster_indexes = predict_kmean(centers, Xcoord)\n",
    "    \n",
    "    nnetworks = [None]*n_clusters\n",
    "        \n",
    "    for i in range(n_clusters):\n",
    "        print(\"Treinando rede neural para o cluster %d\" % (i,))\n",
    "        nn = mlp(hidden_layer_sizes=(10,10), max_iter=2000)\n",
    "        idx = cluster_indexes == i\n",
    "        nn.fit(X[idx], y[idx])\n",
    "        nnetworks[i]= nn\n",
    "    \n",
    "    print(\"Treinamento Finalizado! \")\n",
    "    \n",
    "    return nnetworks\n",
    "\n",
    "x_scaler = StandardScaler()\n",
    "y_scaler = StandardScaler()\n",
    "\n",
    "nn_train_x = x_scaler.fit_transform(km_train[['rooms','suites','garages', 'area','bathrooms']].fillna(0.0))\n",
    "\n",
    "nn_train_coords = km_train[['longitude','latitude']]\n",
    "\n",
    "nn_train_y = y_scaler.fit_transform(km_train[['priceValue']]).ravel()\n",
    "\n",
    "trained_nnets = train_clustered_nn(nn_train_x, nn_train_y, nn_train_coords, coord_centers)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tendo a rede treinada, a função `predict_clustered_nn` roda a rede neural correta com base nas coordenadas:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def predict_clustered_nn(X, Xcoord, centers, nnets):\n",
    "    n_clusters = len(centers)\n",
    "    cluster_indexes = predict_kmean(centers, Xcoord)\n",
    "    return np.array([nnets[cluster_indexes[i]].predict(X[i].reshape(1, -1)) for i in range(len(X))]).ravel()\n",
    "    \n",
    "nn_pred_y = predict_clustered_nn(nn_train_x, nn_train_coords, coord_centers, trained_nnets)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para o qual teremos os seguintes resultados no conjunto de treino:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R2 para treino =0.7867\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"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",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"R2 para treino =%.4f\" % stats.linregress(nn_pred_y,nn_train_y).rvalue **2)\n",
    "plt.figure()\n",
    "plt.scatter(nn_pred_y,nn_train_y, c=abs(nn_pred_y-nn_train_y))\n",
    "\n",
    "plt.plot([-1.5,3], [-1.5,3], 'k-')\n",
    "plt.title(\"Resultado do conjunto de treino para NN clusterizadas\")\n",
    "plt.xlabel(\"Estimado\")\n",
    "plt.xlabel(\"Real\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Observamos um resultado bem melhor que o obtido anteriormente, com $Rˆ2=0.78$. Agora rodando para o conjunto de de testes:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R2 para treino =0.5430\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        this.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width);\n",
       "        canvas.attr('height', height);\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'];\n",
       "    var y0 = fig.canvas.height - msg['y0'];\n",
       "    var x1 = msg['x1'];\n",
       "    var y1 = fig.canvas.height - msg['y1'];\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x;\n",
       "    var y = canvas_pos.y;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"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",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nn_test_x = x_scaler.transform(km_test[['rooms','suites','garages', 'area','bathrooms']].fillna(0.0))\n",
    "\n",
    "nn_test_coords = km_test[['longitude','latitude']]\n",
    "\n",
    "nn_test_y = y_scaler.fit_transform(km_test[['priceValue']]).ravel()\n",
    "\n",
    "nn_test_y_pred = predict_clustered_nn(nn_test_x, nn_test_coords, coord_centers, trained_nnets)\n",
    "\n",
    "print(\"R2 para treino =%.4f\" % stats.linregress(nn_test_y_pred,nn_test_y).rvalue **2)\n",
    "plt.figure()\n",
    "plt.scatter(nn_test_y_pred,nn_test_y, c=abs(nn_test_y_pred-nn_test_y))\n",
    "\n",
    "plt.plot([-1.5,3], [-1.5,3], 'k-')\n",
    "plt.title(\"Resultado do conjunto de teste para NN clusterizadas\")\n",
    "plt.xlabel(\"Estimado\")\n",
    "plt.xlabel(\"Real\")\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusão\n",
    "\n",
    "Estudamos aqui a possibilidade de estimar valores de aluguel na cidade de Florianópolis usando dados reais. Para tal, focamos que approaches que fazem o uso de redes neurais. Ao usar redes comuns, tivemos um resultado insatisfatório com $R^2$ de 0.5 para o conjunto de treino.\n",
    "\n",
    "Em sequência, propusemos uma solução que segmenta o problema através da clusterização dos dados, e em seguida treina uma rede neural para cada um desses subproblemas. Esse método é muito parecido com redes neurais competitivas, por exemplo. Conseguimos melhorar consideralmente nosso resultado para casos de treino ($R^2$ de 0.75), porém para casos de teste temos apenas 0.55.\n",
    "\n",
    "Podemos alcançar resultados ainda melhores. Apesar de predição de aluguel ser um problema comum para exemplificar algoritmos, o cálculo real involve uma série de parâmetros (como posição em relação ao sol). Por isso, o aumento marginal da troca de algoritmos é ínfimo comparado ao possível ganho obtivo através da melhoria da qualidade dos dados com coleta ou melhor limpeza dos já possuímos. Poderíamos, por exemplo, verificar a distância dos pontos para pontos de interesse como hospitais e escolas, ou então cruzar com bancos de dados de criminalidade. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Raw Cell Format",
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}