Simple data sonification and tests with prepared data.ipynb

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "heading",
     "level": 1,
     "metadata": {},
     "source": [
      "Simple data sonification and tests with prepared data"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "0. Intro"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This tentative <a href=\"http://ipython.org/notebook.html\" target=\"_blank\">notebook</a> is about the sonification of data values with short audible frequencies processed with granular synthesis done in an early phase of \"Experimental Data Aesthetics. Multi-sensory Exploration of High-dimensional Data Sets as an Issue of Design Research\" a project funded by the Swiss National Science Foundation. IPython code lines are executable in case the environment is <a href=\"http://ipython.org/install.html\" target=\"_blank\">installed.</a> For the synthesis of audio <a href=\"https://supercollider.github.io/\" target=\"_blank\"> Supercollider (SC3)</a> is running in parallel to the IPython-server (See <a href=\"http://doc.sccode.org/Tutorials/Getting-Started/00-Getting-Started-With-SC.html\" traget=\"_blank\">this</a> for tutorials). Following code lines will create a synthdef, which is receiving <a href=\"http://opensoundcontrol.org/introduction-osc\" target=\"_blank\">Open Sound Control (OSC)</a> messages from IPython.  "
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<code>(\n",
      "SynthDef(\"grain\", { |out, amp=0.1, freq=440, sustain=0.01, pan|\n",
      "\tvar snd = FSinOsc.ar(freq);\n",
      "\tvar amp2 = amp * AmpComp.ir(freq.max(50)) * 0.5;\n",
      "\tvar env = EnvGen.ar(Env.sine(sustain, amp2), doneAction: 2);\n",
      "\tOffsetOut.ar(out, Pan2.ar(snd * env, pan));\n",
      "}, \\ir ! 5).add;\n",
      ")\n",
      "</code>"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "In order to send OSC messages <a href=\"https://trac.v2.nl/wiki/pyOSC\" target=\"_blank\">pyOSC</a> is required. Using the terminal or other command line tools and doing \"sudo ipython setup.py install\" inside the pyOSC folder will install the package. Following lines will load all required packages (below the versions numbers)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import numpy as np\n",
      "from numpy import random\n",
      "import pandas as pd\n",
      "from StringIO import StringIO\n",
      "from IPython.display import Image\n",
      "from IPython import display\n",
      "\n",
      "import matplotlib.pyplot as plt\n",
      "import matplotlib as mpl\n",
      "import matplotlib.gridspec as gridspec\n",
      "\n",
      "import OSC\n",
      "import time, sys\n",
      "import time, random\n",
      "client = OSC.OSCClient()\n",
      "client.connect( ( '127.0.0.1', 57110 ) )"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The synthdefs in the Supercollider-server on localhost is receving OSC messages via port 57110 in the following way:\n",
      "\n",
      "<code>\"s_new\", \\nameofsynthdef, -1, 0, 1, \\argument_1, argumenvalue_1 (as int or float), \\argument_2, argumenvalue_2, ... \\argument_n, argumenvalue_n</code>\n",
      "\n",
      "The values (-1, 0, 1) between \\nameofsynthdef and \\argument_1 are explained <a href=\"http://doc.sccode.org/Reference/Server-Command-Reference.html#Synth%20Commands\" target=\"_blank\">here</a> in detail. The procedure to send was named \"oscgrain\" and defined as a function."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def oscgrain( frequency, vol, sustain ):\n",
      "    msg = OSC.OSCMessage()\n",
      "    msg.setAddress(\"s_new\")\n",
      "    msg.append(\"grain\")\n",
      "    msg.append(-1)\n",
      "    msg.append(0)\n",
      "    msg.append(1)\n",
      "    msg.append(\"amp\")\n",
      "    msg.append(vol)\n",
      "    msg.append(\"freq\")\n",
      "    msg.append(frequency)     #read in data points\n",
      "    msg.append(\"sustain\")\n",
      "    msg.append(sustain)\n",
      "    client.send(msg)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Sending is now easier. Running following line"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "oscgrain(5000, 0.8, 0.1)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "will create this short sonic grain: "
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/148147735&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_artwork=true\"></iframe>"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "1. Reading data from csv-files"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Loading data is done by running the lines below. The second line is scaling all values of an attribute/dimension between 0.0 and 1.0. Data is processed with <a href=\"http://www.numpy.org/\" traget=\"_blank\">numpy.</a> As an example the famous <a href=\"http://en.wikipedia.org/wiki/Iris_flower_data_set\" target=\"_blank\">iris-flower-data set</a> is loaded.  "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = np.genfromtxt(\"iris.data\", dtype=float, delimiter=',', skip_header=1) \n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0)) \n",
      "#axis=0 is max/min in each column and axis=1 in each row"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Its SPLOM and parallel coordinate visualization is featured in nature methods:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.display import HTML\n",
      "HTML('<iframe src=http://www.nature.com/nmeth/journal/v9/n10/images/nmeth.2186-F1.jpg width=900 height=800></iframe>')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<iframe src=http://www.nature.com/nmeth/journal/v9/n10/images/nmeth.2186-F1.jpg width=900 height=800></iframe>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 5,
       "text": [
        "<IPython.core.display.HTML at 0x102472090>"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "2. Examples of iris-flower-data set sonification"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Random selection of one data point with 4 attributes/dimensions mapped to frequencies between 100 and 1400 Hz with a sustain of 1 sec. The amplitude/volume is a reciprocal of the number of attributes/dimensions, which is specified by <code>np.asarray(data.shape)[1]</code>. With <code>data.shape</code> the shape of the numpy array named data is determined. It consists of 178 rows and 5 columns, the last column for the classes (flower names)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "for i in range(np.asarray(data.shape)[1]-1):\n",
      "    oscgrain(100+(1400*data_all[random.randint(0,(np.asarray(data.shape)[0])-1),i]), 1.0/np.asarray(data.shape)[1], 1.00)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Some examples played randomly in one sequence:"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/148150049&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_artwork=true\"></iframe>"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Other data sets. Notice the small adjustments required to read in the data properly."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = np.genfromtxt(\"ppg2008.csv\", dtype=float, delimiter=',', skip_header=1) \n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "for i in range(np.asarray(data.shape)[1]-1):\n",
      "    oscgrain(100+(1400*data_all[random.randint(0,(np.asarray(data.shape)[0])-1),i+1]), 1.0/np.asarray(data.shape)[1], 1.00)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/170708990&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_comments=true&amp;show_user=true&amp;show_reposts=false\"></iframe>"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = np.genfromtxt(\"wine_names.csv\", dtype=float, delimiter=',', skip_header=1) \n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "for i in range(np.asarray(data.shape)[1]-1):\n",
      "    oscgrain(100+(1400*data_all[random.randint(0,(np.asarray(data.shape)[0])-1),i+1]), 1.0/np.asarray(data.shape)[1], 1.00)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 9
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Running the code below will create a sequence of shorter grains of 30ms, which are all together sonifications of the whole iris-flower-data set. It is a direct sonification of the data sets as it is stored in rows and columns in the provided file and is similar to a heatmap display, althought the information display in heatmaps is more efficient. \n",
      "\n",
      "First load the iris-flower-data set again:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = np.genfromtxt(\"iris.data\", dtype=float, delimiter=',', skip_header=1) \n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "for i in range(np.asarray(data.shape)[0]):\n",
      "    for ii in range (np.asarray(data.shape)[1]-1): #last dimension is not a float = class information\n",
      "        oscgrain(2000+(1500*data_all[i,ii]),1.0/np.asarray(data.shape)[1],0.03)\n",
      "        time.sleep(0.1) # this adds a short break of 100ms between the data attributes\n",
      "    time.sleep(0.2) # this adds a break of 200ms between the data points"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 10
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "It takes about 1 min 30 seconds and sounds like this."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/148175531&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_artwork=true\"></iframe>"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Heatmap of the same data set"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def heatmapper ( filename, width, height ):\n",
      "\n",
      "    # Plot it out\n",
      "    fig, ax = plt.subplots()\n",
      "    heatmap = ax.pcolor(data_all, cmap=plt.cm.Reds, alpha=0.8)\n",
      "\n",
      "    # Format\n",
      "    fig = plt.gcf()\n",
      "    fig.set_size_inches(width, height)\n",
      "\n",
      "    # turn off the frame\n",
      "    ax.set_frame_on(False)\n",
      "\n",
      "    # put the major ticks at the middle of each cell\n",
      "    ax.set_yticks(np.arange(data_all.shape[0]) + 0.5, minor=False)\n",
      "    ax.set_xticks(np.arange(data_all.shape[1]) + 0.5, minor=False)\n",
      "\n",
      "    # want a more natural, table-like display\n",
      "    ax.invert_yaxis()\n",
      "    ax.xaxis.tick_top()\n",
      "\n",
      "    #Labels\n",
      "    ax.set_xticklabels(data_all.columns, minor=False)\n",
      "    ax.set_yticklabels(data_all.index, minor=False)\n",
      "\n",
      "    # rotate the\n",
      "    plt.xticks(rotation=90)\n",
      "\n",
      "    ax.grid(False)\n",
      "\n",
      "    # Turn off all the ticks\n",
      "    ax = plt.gca()\n",
      "    fig.savefig(filename)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = pd.read_csv(\"iris.csv\", index_col=4)\n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "heatmapper ('heatmap_iris_current.pdf', 8, 15)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 13
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Faster sonification"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = np.genfromtxt(\"iris.data\", dtype=float, delimiter=',', skip_header=1) \n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "for i in range(np.asarray(data.shape)[0]):\n",
      "    for ii in range (np.asarray(data.shape)[1]-1): #last dimension is not a float = class information\n",
      "        oscgrain(2000+(1500*data_all[i,ii]),1.0/np.asarray(data.shape)[1],0.03)\n",
      "        time.sleep(0.01) # this adds a short break of 10ms between the data attributes\n",
      "    time.sleep(0.06) # this adds a break of 60ms between the data points"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 61
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/148177539&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_artwork=true\"></iframe>"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Heatmap and spectrogramms are comparable. Although in the spectrogramm the attribues/values/frequencies of a data point are not stacked, but lined. \n",
      "\n",
      "<p><img src=\"https://raw.githubusercontent.com/experimentalDataAesthetics/ipython_notes/master/heatmapvsspectrogram.jpg\" alt=\"heatmapvsspectrogram.jpg\"></p>"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "3. Creating prepared data sets"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "After exploring a bit the simple sonification possibilities of high-dimensional data sets. An evaluation and testing of the sound design possibilities with the current set-up (granular synthesis with sinusoidal grains) is done in the following by sonification of prepare data sets. "
     ]
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "Circle"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def CreateCircleArray(circle, numpoints):\n",
      "    theta = np.linspace(0, circle*np.pi, numpoints)\n",
      "    x = 1 * np.cos(theta)\n",
      "    y = 1 * np.sin(theta)\n",
      "    return np.array([x, y]).T\n",
      "\n",
      "def PlotArray(array):\n",
      "    ax = plt.subplot(111, aspect=\"equal\")\n",
      "    ax.scatter(array[:, 0], array[:, 1], color=\"k\")\n",
      "    plt.show()\n",
      "%pylab inline"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Populating the interactive namespace from numpy and matplotlib\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "WARNING: pylab import has clobbered these variables: ['random', 'display']\n",
        "`%pylab --no-import-all` prevents importing * from pylab and numpy\n"
       ]
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "http://stackoverflow.com/questions/23499751/scatter-plot-2d-which-shows-a-dotted-circle-and-other-2d-shapes-made-by-geome"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "PlotArray(CreateCircleArray(2,50))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAQ8AAAEACAYAAACtefPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107afd810>"
       ]
      }
     ],
     "prompt_number": 104
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = CreateCircleArray(2, 100)\n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 16
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "for i in range(np.asarray(data.shape)[0]):\n",
      "    for ii in range (np.asarray(data.shape)[1]): \n",
      "        oscgrain(200+(2400*data_all[i,ii]),1.0/np.asarray(data.shape)[1],0.03) #plays all attributeds datapoints at once\n",
      "    time.sleep(0.1) # this adds a break of 100ms between the data points"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 17
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/148585192&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_artwork=true\"></iframe>"
     ]
    },
    {
     "cell_type": "heading",
     "level": 4,
     "metadata": {},
     "source": [
      "Sorted by columns (dimensions)"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Played sequentially incrementing the x axis. Creates an ping-pong-effect."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = CreateCircleArray(2, 100)\n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "order = data_all[:, 0].argsort()\n",
      "data_sorted = np.take(data_all, order, 0)\n",
      "for i in range(np.asarray(data.shape)[0]): #read rows \n",
      "    for ii in range (np.asarray(data.shape)[1]): #read attribute in one row\n",
      "        oscgrain(200+(2400*data_sorted[i,ii]),1.5/np.asarray(data.shape)[1],0.03)\n",
      "    time.sleep(0.1) # this adds a break of 100ms between the data points"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 18
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "http://stackoverflow.com/questions/5047407/python-numpy-array-sorting"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/148681371&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_artwork=true\"></iframe>"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Psychoacoustics and Sound design"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Upper limit of frequencies needs to be around 5 kHz.[1] A mapping of 130.813 Hz (C3) and 4186.01 Hz (C8), which are 5 octaves seems to make sense."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Running the following line will create a \"sweep\" with 1000 steps."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "line = np.linspace(0.0, 1.0, num=200)\n",
      "for i in range(200): \n",
      "    oscgrain((130.813+(4186.01*line[i])),1.0,0.029) \n",
      "    time.sleep(0.035) # this adds a break of 100ms between "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 29
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "A semi-circle, slow"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = CreateCircleArray(1, 50)\n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "order = data_all[:, 0].argsort()\n",
      "data_sorted = np.take(data_all, order, 0)\n",
      "for i in range(np.asarray(data.shape)[0]): #read rows \n",
      "    for ii in range (np.asarray(data.shape)[1]): #read attribute in one row\n",
      "        oscgrain(130.813+(4186.01*data_sorted[i,ii]),1.5/np.asarray(data.shape)[1],0.03)\n",
      "    time.sleep(1.0) # this adds a break of 1000ms between the data points"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "PlotArray(data_sorted)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAQ4AAAEACAYAAABCu5jVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107b3da50>"
       ]
      }
     ],
     "prompt_number": 118
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "slope down is not really audible?.. "
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/148683640&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_artwork=true\"></iframe>"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Frequency mapping is somehow wrong. Octaves increase exponentially when measured on a linear frequency scale (Hz). <br>\n",
      "http://en.wikipedia.org/wiki/Cent_%28music%29"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.display import HTML\n",
      "HTML('<iframe src=http://upload.wikimedia.org/wikipedia/commons/thumb/1/13/4Octaves.and.Frequencies.svg/500px-4Octaves.and.Frequencies.svg.png width=512 height=100></iframe>')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<iframe src=http://upload.wikimedia.org/wikipedia/commons/thumb/1/13/4Octaves.and.Frequencies.svg/500px-4Octaves.and.Frequencies.svg.png width=512 height=100></iframe>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 151,
       "text": [
        "<IPython.core.display.HTML at 0x107b3df50>"
       ]
      }
     ],
     "prompt_number": 151
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.display import HTML\n",
      "HTML('<iframe src=http://upload.wikimedia.org/wikipedia/commons/thumb/3/31/4Octaves.and.Frequencies.Ears.svg/500px-4Octaves.and.Frequencies.Ears.svg.png width=512 height=100></iframe>')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<iframe src=http://upload.wikimedia.org/wikipedia/commons/thumb/3/31/4Octaves.and.Frequencies.Ears.svg/500px-4Octaves.and.Frequencies.Ears.svg.png width=512 height=100></iframe>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 150,
       "text": [
        "<IPython.core.display.HTML at 0x107b3d0d0>"
       ]
      }
     ],
     "prompt_number": 150
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "With some scaling function distance between the octaves is linear:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def freqscaler(source, basicfreq, octaves):\n",
      "    out = basicfreq * np.power(np.power(2.0, octaves), source)\n",
      "    return out"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 19
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = CreateCircleArray(1, 50)\n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "order = data_all[:, 0].argsort()\n",
      "data_sorted = np.take(data_all, order, 0)\n",
      "for i in range(np.asarray(data.shape)[0]): #read rows \n",
      "    for ii in range (np.asarray(data.shape)[1]): #read attribute in one row\n",
      "        oscgrain(freqscaler(data_sorted[i,ii], 440, 2),1.0/np.asarray(data.shape)[1],0.03)\n",
      "    time.sleep(0.1) "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 22
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/170710132&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_comments=true&amp;show_user=true&amp;show_reposts=false\"></iframe>"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "PlotArray(data_sorted)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAQ4AAAEACAYAAABCu5jVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x102472e90>"
       ]
      }
     ],
     "prompt_number": 24
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "line = np.linspace(0.0, 1.0, num=100)\n",
      "for i in range(100): \n",
      "    oscgrain(freqscaler(line[i], 220, 4),1.0/np.asarray(data.shape)[1],0.03)\n",
      "    time.sleep(0.1) # this adds a break of 35ms between "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 28
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = np.genfromtxt(\"iris.data\", dtype=float, delimiter=',', skip_header=1) \n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "\n",
      "for i in range(np.asarray(data.shape)[0]):\n",
      "    for ii in range (np.asarray(data.shape)[1]-1): #last dimension is not a float = class information\n",
      "        oscgrain(freqscaler(data_all[i,ii], 880, 2),1.0/np.asarray(data.shape)[1],0.03)\n",
      "        time.sleep(0.01) # this adds a short break of 10ms between the data attributes\n",
      "    time.sleep(0.06) # this adds a break of 60ms between the data points"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 30
    },
    {
     "cell_type": "heading",
     "level": 4,
     "metadata": {},
     "source": [
      "Pink noise grain to double map/enforce X-axis"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<code>(\n",
      "SynthDef(\"noiseburst\", { |out, amp=0.1, sustain=0.01, pan|\n",
      "\tvar snd = PinkNoise.ar(1.0);\n",
      "\tvar amp2 = amp * AmpComp.ir(freq.max(50)) * 0.5;\n",
      "\tvar env = EnvGen.ar(Env.sine(sustain, amp2), doneAction: 2);\n",
      "\tOffsetOut.ar(out, Pan2.ar(snd * env, pan));\n",
      "}, \\ir ! 5).add;\n",
      ")</code>"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def noiseburst( pan ):\n",
      "    msg = OSC.OSCMessage()\n",
      "    msg.setAddress(\"s_new\")\n",
      "    msg.append(\"noiseburst\")\n",
      "    msg.append(-1)\n",
      "    msg.append(0)\n",
      "    msg.append(1)\n",
      "    msg.append(\"amp\")\n",
      "    msg.append(0.2)\n",
      "    msg.append(\"sustain\")\n",
      "    msg.append(0.08)\n",
      "    msg.append(\"pan\")\n",
      "    msg.append(-1+(2*pan))    \n",
      "    client.send(msg)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 26
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = CreateCircleArray(1, 50)\n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "order = data_all[:, 0].argsort()\n",
      "data_sorted = np.take(data_all, order, 0)\n",
      "for i in range(np.asarray(data.shape)[0]): #read rows \n",
      "    for ii in range (np.asarray(data.shape)[1]): #read attribute in one row\n",
      "        oscgrain(freqscaler(data_sorted[i,ii], 880, 2),1.5/np.asarray(data.shape)[1],0.03)\n",
      "        noiseburst(data_sorted[i,:])\n",
      "    time.sleep(0.1) "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 32
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/170723465&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_comments=true&amp;show_user=true&amp;show_reposts=false\"></iframe>"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Panning the grains directly is not so audible"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def oscgrain2( frequency, vol, pan ):\n",
      "    msg = OSC.OSCMessage()\n",
      "    msg.setAddress(\"s_new\")\n",
      "    msg.append(\"grain\")\n",
      "    msg.append(-1)\n",
      "    msg.append(0)\n",
      "    msg.append(1)\n",
      "    msg.append(\"amp\")\n",
      "    msg.append(vol)\n",
      "    msg.append(\"freq\")\n",
      "    msg.append(frequency)     #read in data points\n",
      "    msg.append(\"sustain\")\n",
      "    msg.append(0.03)\n",
      "    msg.append(\"pan\")\n",
      "    msg.append(-1+(2*pan))    \n",
      "    client.send(msg)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 33
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = CreateCircleArray(1, 50)\n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "order = data_all[:, 0].argsort()\n",
      "data_sorted = np.take(data_all, order, 0)\n",
      "for i in range(np.asarray(data.shape)[0]): #read rows \n",
      "    for ii in range (np.asarray(data.shape)[1]): #read attribute in one row\n",
      "        oscgrain2(freqscaler(data_sorted[i,ii], (1*880),2),1.0/np.asarray(data.shape)[1],data_sorted[i,:])\n",
      "    time.sleep(0.1) "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 38
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "combined and scatter plot a full circle (still reading from the x-axis)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = CreateCircleArray(2, 50)\n",
      "data_all = (data - np.min(data, axis=0))/(np.max(data, axis=0) - np.min(data, axis=0))\n",
      "order = data_all[:, 0].argsort()\n",
      "data_sorted = np.take(data_all, order, 0)\n",
      "for i in range(np.asarray(data.shape)[0]): #read rows \n",
      "    for ii in range (np.asarray(data.shape)[1]): #read attribute in one row\n",
      "        oscgrain2(freqscaler(data_sorted[i,ii], (1*880),2),1.0/np.asarray(data.shape)[1],data_sorted[i,:])\n",
      "        noiseburst(data_sorted[i,:])\n",
      "    time.sleep(0.1) "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 40
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<iframe width=\"100%\" height=\"166\" scrolling=\"no\" frameborder=\"no\" src=\"https://w.soundcloud.com/player/?url=https%3A//api.soundcloud.com/tracks/170723872&amp;color=ff5500&amp;auto_play=false&amp;hide_related=false&amp;show_comments=true&amp;show_user=true&amp;show_reposts=false\"></iframe>"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Brushing is now ping-ponging... How to brush with a \"bar\" >> how to read more than one datapoint/row at once?"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Applicable to SPLOM with more than 2D?"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Remaing issues"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "http://en.wikipedia.org/wiki/Equal-loudness_contour and http://en.wikipedia.org/wiki/A-weighting"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "\"One of the most salient features of biologically interesting sounds is the rapid variation in spectral content over time.\"[2]<br>\n",
      "Rhythm?<br>"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Next project: Scatterplot as a GUI for interaction"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Include processing sketches?\n",
      "\n",
      "Brushing = reading = data point triggering<br>\n",
      "Hand-controlled Brushing<br>\n",
      "Hand-Looped brushing<br>\n",
      "Automated brushing/triggering<br>\n",
      "Echolocation ===> Data resonance (shortest paths waveguide?), mouse/ ant in a maze?\n",
      "http://en.wikipedia.org/wiki/Minimum_spanning_tree\n",
      "\n",
      "Check: https://github.com/experimentalDataAesthetics/processing-prototyping/tree/master/SPLOM_with_GUI"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Endnotes"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "[1] Carlile, Simon. \u201cPsychoacoustics.\u201d In The Sonification Handbook, edited by Thomas Hermann, Andy Hunt, and John G. Neuhoff, 41\u201361. Berlin, Germany: Logos Publishing House, 2011. http://sonification.de/handbook/chapters/chapter3/, p. 48.<br>\n",
      "[2] Ibid., p. 49."
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Versioning"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%load_ext version_information\n",
      "%version_information numpy, scipy, matplotlib, sympy, pyosc, pandas, conda"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<table><tr><th>Software</th><th>Version</th></tr><tr><td>Python</td><td>2.7.6 |Anaconda 1.9.2 (x86_64)| (default, Jan 10 2014, 11:23:15) [GCC 4.0.1 (Apple Inc. build 5493)]</td></tr><tr><td>IPython</td><td>1.1.0</td></tr><tr><td>OS</td><td>posix [darwin]</td></tr><tr><td>numpy</td><td>1.8.1</td></tr><tr><td>scipy</td><td>0.13.3</td></tr><tr><td>matplotlib</td><td>1.3.1</td></tr><tr><td>sympy</td><td>0.7.4.1</td></tr><tr><td>pyosc</td><td>0.3.5b-5294</td></tr><tr><td>pandas</td><td>0.13.1</td></tr><tr><td>conda</td><td>3.4.1</td></tr><tr><td colspan='2'>Sun Oct 05 14:54:55 2014 CEST</td></tr></table>"
       ],
       "json": [
        "{\"Software versions\": [{\"version\": \"2.7.6 |Anaconda 1.9.2 (x86_64)| (default, Jan 10 2014, 11:23:15) [GCC 4.0.1 (Apple Inc. build 5493)]\", \"module\": \"Python\"}, {\"version\": \"1.1.0\", \"module\": \"IPython\"}, {\"version\": \"posix [darwin]\", \"module\": \"OS\"}, {\"version\": \"1.8.1\", \"module\": \"numpy\"}, {\"version\": \"0.13.3\", \"module\": \"scipy\"}, {\"version\": \"1.3.1\", \"module\": \"matplotlib\"}, {\"version\": \"0.7.4.1\", \"module\": \"sympy\"}, {\"version\": \"0.3.5b-5294\", \"module\": \"pyosc\"}, {\"version\": \"0.13.1\", \"module\": \"pandas\"}, {\"version\": \"3.4.1\", \"module\": \"conda\"}]}"
       ],
       "latex": [
        "\\begin{tabular}{|l|l|}\\hline\n",
        "{\\bf Software} & {\\bf Version} \\\\ \\hline\\hline\n",
        "Python & 2.7.6 |Anaconda 1.9.2 (x86\\letterunderscore{}64)| (default, Jan 10 2014, 11:23:15) [GCC 4.0.1 (Apple Inc. build 5493)] \\\\ \\hline\n",
        "IPython & 1.1.0 \\\\ \\hline\n",
        "OS & posix [darwin] \\\\ \\hline\n",
        "numpy & 1.8.1 \\\\ \\hline\n",
        "scipy & 0.13.3 \\\\ \\hline\n",
        "matplotlib & 1.3.1 \\\\ \\hline\n",
        "sympy & 0.7.4.1 \\\\ \\hline\n",
        "pyosc & 0.3.5b-5294 \\\\ \\hline\n",
        "pandas & 0.13.1 \\\\ \\hline\n",
        "conda & 3.4.1 \\\\ \\hline\n",
        "\\hline \\multicolumn{2}{|l|}{Sun Oct 05 14:54:55 2014 CEST} \\\\ \\hline\n",
        "\\end{tabular}\n"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 41,
       "text": [
        "Software versions\n",
        "Python 2.7.6 |Anaconda 1.9.2 (x86_64)| (default, Jan 10 2014, 11:23:15) [GCC 4.0.1 (Apple Inc. build 5493)]\n",
        "IPython 1.1.0\n",
        "OS posix [darwin]\n",
        "numpy 1.8.1\n",
        "scipy 0.13.3\n",
        "matplotlib 1.3.1\n",
        "sympy 0.7.4.1\n",
        "pyosc 0.3.5b-5294\n",
        "pandas 0.13.1\n",
        "conda 3.4.1\n",
        "<tr><td colspan='2'>Sun Oct 05 14:54:55 2014 CEST</td></tr>"
       ]
      }
     ],
     "prompt_number": 41
    }
   ],
   "metadata": {}
  }
 ]
}