Grasp_and_Lift.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#Grasp and Lift EEG analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://www.kaggle.com/c/grasp-and-lift-eeg-detection/data?train.zip"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "http://www.bci2000.org/wiki/index.php/User_Tutorial:Introduction_to_the_Mu_Rhythm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"http://www.bci2000.org/wiki/images/thumb/7/70/MuRhythmModulation.PNG/471px-MuRhythmModulation.PNG\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image, HTML \n",
    "Image(url='http://www.bci2000.org/wiki/images/thumb/7/70/MuRhythmModulation.PNG/471px-MuRhythmModulation.PNG')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\"\n",
       "  \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\">\n",
       "\n",
       "\n",
       "<html xmlns=\"http://www.w3.org/1999/xhtml\">\n",
       "  <head>\n",
       "    <meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n",
       "    \n",
       "    <title>Motor imagery decoding from EEG data using the Common Spatial Pattern (CSP) &mdash; MNE 0.10.dev0 documentation</title>\n",
       "    \n",
       "    <link rel=\"stylesheet\" href=\"../../_static/navy.css\" type=\"text/css\" />\n",
       "    <link rel=\"stylesheet\" href=\"../../_static/pygments.css\" type=\"text/css\" />\n",
       "    \n",
       "    <script type=\"text/javascript\">\n",
       "      var DOCUMENTATION_OPTIONS = {\n",
       "        URL_ROOT:    '../../',\n",
       "        VERSION:     '0.10.dev0',\n",
       "        COLLAPSE_INDEX: false,\n",
       "        FILE_SUFFIX: '.html',\n",
       "        HAS_SOURCE:  true\n",
       "      };\n",
       "    </script>\n",
       "    <script type=\"text/javascript\" src=\"../../_static/jquery.js\"></script>\n",
       "    <script type=\"text/javascript\" src=\"../../_static/underscore.js\"></script>\n",
       "    <script type=\"text/javascript\" src=\"../../_static/doctools.js\"></script>\n",
       "    <link rel=\"top\" title=\"MNE 0.10.dev0 documentation\" href=\"../../index.html\" />\n",
       "    <link rel=\"up\" title=\"Examples\" href=\"../index.html\" />\n",
       "    <link rel=\"next\" title=\"Decoding source space data\" href=\"plot_decoding_spatio_temporal_source.html\" />\n",
       "    <link rel=\"prev\" title=\"Decoding in sensor space data using the Common Spatial Pattern (CSP)\" href=\"plot_decoding_csp_space.html\" />\n",
       "\n",
       "<link href='http://fonts.googleapis.com/css?family=Open+Sans:400italic,700italic,400,700' rel='stylesheet' type='text/css'>\n",
       "\n",
       "\n",
       "    <script type=\"text/javascript\">\n",
       "    var _gaq = _gaq || [];\n",
       "    _gaq.push(['_setAccount', 'UA-37225609-1']);\n",
       "    _gaq.push(['_trackPageview']);\n",
       "\n",
       "    (function() {\n",
       "        var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;\n",
       "        ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';\n",
       "        var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);\n",
       "    })();\n",
       "    </script>\n",
       "\n",
       "\n",
       "\n",
       "    <script type=\"text/javascript\">\n",
       "    !function(d,s,id){var js,fjs=d.getElementsByTagName(s)[0];if(!d.getElementById(id)){js=d.createElement(s);\n",
       "    js.id=id;js.src=\"http://platform.twitter.com/widgets.js\";\n",
       "    fjs.parentNode.insertBefore(js,fjs);}}(document,\"script\",\"twitter-wjs\");\n",
       "    </script>\n",
       "\n",
       "\n",
       "\n",
       "    <script type=\"text/javascript\">\n",
       "    (function() {\n",
       "    var po = document.createElement('script'); po.type = 'text/javascript'; po.async = true;\n",
       "    po.src = 'https://apis.google.com/js/plusone.js';\n",
       "    var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(po, s);\n",
       "    })();\n",
       "    </script>\n",
       "\n",
       "\n",
       "  </head>\n",
       "  <body role=\"document\">\n",
       "\n",
       "<div style=\"background-color: red; color: white; font-weight:bold; text-align: center; padding: 10px; min-width: 910px\">\n",
       "This documentation is for the development version (0.10.dev0) - <a href=\"http://martinos.org/mne/stable\">Stable version</a>\n",
       "</div>\n",
       "\n",
       "<div style=\"background-color: white; text-align: left; padding: 10px 7px 15px 15px; min-width: 910px\">\n",
       "<div style=\"float: left\">\n",
       "<a href=\"../../index.html\"><img src=\"../../_static/mne_logo.png\" border=\"0\" alt=\"py4sci\"/></a>\n",
       "</div>\n",
       "\n",
       "<div style=\"float: right\">\n",
       "<a href=\"../../index.html\"><img src=\"../../_static/institutions.png\" border=\"0\" alt=\"py4sci\"/></a>\n",
       "</div>\n",
       "<br style=\"clear:both\"/>\n",
       "</div>\n",
       "\n",
       "    <div class=\"related\" role=\"navigation\" aria-label=\"related navigation\">\n",
       "      <h3>Navigation</h3>\n",
       "      <ul>\n",
       "        <li class=\"right\" style=\"margin-right: 10px\">\n",
       "          <a href=\"../../np-modindex.html\" title=\"Python Module Index\"\n",
       "             >modules</a></li>\n",
       "        <li class=\"right\" >\n",
       "          <a href=\"plot_decoding_spatio_temporal_source.html\" title=\"Decoding source space data\"\n",
       "             accesskey=\"N\">next</a> |</li>\n",
       "        <li class=\"right\" >\n",
       "          <a href=\"plot_decoding_csp_space.html\" title=\"Decoding in sensor space data using the Common Spatial Pattern (CSP)\"\n",
       "             accesskey=\"P\">previous</a> |</li>\n",
       "        <li><a href=\"../../index.html\">Home</a>&nbsp;|&nbsp;</li>\n",
       "        <li><a href=\"../../manual.html\">Manual</a>&nbsp;|&nbsp;</li>\n",
       "        <li><a href=\"../../mne-python.html\">Python</a>&nbsp;|&nbsp;</li>\n",
       "        <li><a href=\"../../cite.html\">Cite MNE</a>&nbsp;|&nbsp;</li>\n",
       "        <!-- <li><a href=\"../../search.html\">Search</a></li> -->\n",
       "\n",
       "          <li class=\"nav-item nav-item-1\"><a href=\"../../mne-python.html\" >MNE with Python</a> &raquo;</li>\n",
       "          <li class=\"nav-item nav-item-2\"><a href=\"../index.html\" accesskey=\"U\">Examples</a> &raquo;</li> \n",
       "      </ul>\n",
       "    </div>\n",
       "\n",
       "      <div class=\"sphinxsidebar\" role=\"navigation\" aria-label=\"main navigation\">\n",
       "        <div class=\"sphinxsidebarwrapper\">\n",
       "<h3><a href=\"../../index.html\">Table Of Contents</a></h3>\n",
       "<ul>\n",
       "<li class=\"toctree-l1\"><a class=\"reference internal\" href=\"../../manual.html\">Manual</a></li>\n",
       "<li class=\"toctree-l1\"><a class=\"reference internal\" href=\"../../mne-python.html\">MNE with Python</a></li>\n",
       "<li class=\"toctree-l1\"><a class=\"reference internal\" href=\"../../mne-cpp.html\">MNE with CPP</a></li>\n",
       "<li class=\"toctree-l1\"><a class=\"reference internal\" href=\"../../cite.html\">Cite MNE and MNE-Python</a></li>\n",
       "</ul>\n",
       "\n",
       "<div id=\"searchbox\" style=\"display: none\" role=\"search\">\n",
       "  <h3>Quick search</h3>\n",
       "    <form class=\"search\" action=\"../../search.html\" method=\"get\">\n",
       "      <input type=\"text\" name=\"q\" />\n",
       "      <input type=\"submit\" value=\"Go\" />\n",
       "      <input type=\"hidden\" name=\"check_keywords\" value=\"yes\" />\n",
       "      <input type=\"hidden\" name=\"area\" value=\"default\" />\n",
       "    </form>\n",
       "    <p class=\"searchtip\" style=\"font-size: 90%\">\n",
       "    Enter search terms or a module, class or function name.\n",
       "    </p>\n",
       "</div>\n",
       "<script type=\"text/javascript\">$('#searchbox').show(0);</script><h3>Versions</h3>\n",
       "<ul class=\"current\">\n",
       "    <li class=\"toctree-l1\"><a href=http://martinos.org/mne/stable>Stable</a></li>\n",
       "    <li class=\"toctree-l1\"><a href=http://martinos.org/mne/dev>Development</a></li>\n",
       "</ul>\n",
       "        </div>\n",
       "      </div>\n",
       "\n",
       "    <div class=\"document\">\n",
       "      <div class=\"documentwrapper\">\n",
       "        <div class=\"bodywrapper\">\n",
       "          <div class=\"body\" role=\"main\">\n",
       "            \n",
       "  <div class=\"section\" id=\"motor-imagery-decoding-from-eeg-data-using-the-common-spatial-pattern-csp\">\n",
       "<span id=\"example-decoding-plot-decoding-csp-eeg-py\"></span><h1>Motor imagery decoding from EEG data using the Common Spatial Pattern (CSP)<a class=\"headerlink\" href=\"#motor-imagery-decoding-from-eeg-data-using-the-common-spatial-pattern-csp\" title=\"Permalink to this headline\">¶</a></h1>\n",
       "<p>Decoding of motor imagery applied to EEG data decomposed using CSP.\n",
       "Here the classifier is applied to features extracted on CSP filtered signals.</p>\n",
       "<p>See <a class=\"reference external\" href=\"http://en.wikipedia.org/wiki/Common_spatial_pattern\">http://en.wikipedia.org/wiki/Common_spatial_pattern</a> and [1]</p>\n",
       "<p>The EEGBCI dataset is documented in [2]\n",
       "The data set is available at PhysioNet [3]</p>\n",
       "<dl class=\"docutils\">\n",
       "<dt>[1] Zoltan J. Koles. The quantitative extraction and topographic mapping</dt>\n",
       "<dd>of the abnormal components in the clinical EEG. Electroencephalography\n",
       "and Clinical Neurophysiology, 79(6):440&#8211;447, December 1991.</dd>\n",
       "<dt>[2] Schalk, G., McFarland, D.J., Hinterberger, T., Birbaumer, N.,</dt>\n",
       "<dd>Wolpaw, J.R. (2004) BCI2000: A General-Purpose Brain-Computer Interface\n",
       "(BCI) System. IEEE TBME 51(6):1034-1043</dd>\n",
       "<dt>[3] Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG,</dt>\n",
       "<dd>Mietus JE, Moody GB, Peng C-K, Stanley HE. (2000) PhysioBank,\n",
       "PhysioToolkit, and PhysioNet: Components of a New Research Resource for\n",
       "Complex Physiologic Signals. Circulation 101(23):e215-e220</dd>\n",
       "</dl>\n",
       "<ul class=\"horizontal\">\n",
       "<li><a class=\"first reference internal image-reference\" href=\"../../_images/plot_decoding_csp_eeg_001.png\"><img alt=\"../../_images/plot_decoding_csp_eeg_001.png\" src=\"../../_images/plot_decoding_csp_eeg_001.png\" style=\"width: 399.5px; height: 66.27px;\" /></a>\n",
       "</li>\n",
       "<li><a class=\"first reference internal image-reference\" href=\"../../_images/plot_decoding_csp_eeg_002.png\"><img alt=\"../../_images/plot_decoding_csp_eeg_002.png\" src=\"../../_images/plot_decoding_csp_eeg_002.png\" style=\"width: 376.0px; height: 282.0px;\" /></a>\n",
       "</li>\n",
       "</ul>\n",
       "<p><strong>Script output</strong>:</p>\n",
       "<div class=\"highlight-python\"><div class=\"highlight\"><pre>Extracting edf Parameters from /data1/agramfort/work/src/mne-python/examples/MNE-eegbci-data/physiobank/database/eegmmidb/S001/S001R06.edf...\n",
       "Setting channel info structure...\n",
       "Creating Raw.info structure...\n",
       "Reading 0 ... 19999  =      0.000 ...   124.994 secs...\n",
       "[done]\n",
       "Ready.\n",
       "Extracting edf Parameters from /data1/agramfort/work/src/mne-python/examples/MNE-eegbci-data/physiobank/database/eegmmidb/S001/S001R10.edf...\n",
       "Setting channel info structure...\n",
       "Creating Raw.info structure...\n",
       "Reading 0 ... 19999  =      0.000 ...   124.994 secs...\n",
       "[done]\n",
       "Ready.\n",
       "Extracting edf Parameters from /data1/agramfort/work/src/mne-python/examples/MNE-eegbci-data/physiobank/database/eegmmidb/S001/S001R14.edf...\n",
       "Setting channel info structure...\n",
       "Creating Raw.info structure...\n",
       "Reading 0 ... 19999  =      0.000 ...   124.994 secs...\n",
       "[done]\n",
       "Ready.\n",
       "Band-pass filtering from 7 - 30 Hz\n",
       "Removing orphaned offset at the beginning of the file.\n",
       "...\n",
       "</pre></div>\n",
       "</div>\n",
       "<p><strong>Python source code:</strong> <a class=\"reference download internal\" href=\"../../_downloads/plot_decoding_csp_eeg.py\"><code class=\"xref download docutils literal\"><span class=\"pre\">plot_decoding_csp_eeg.py</span></code></a></p>\n",
       "<div class=\"highlight-python\"><div class=\"highlight\"><pre><span class=\"c\"># Authors: Martin Billinger &lt;martin.billinger@tugraz.at&gt;</span>\n",
       "<span class=\"c\">#</span>\n",
       "<span class=\"c\"># License: BSD (3-clause)</span>\n",
       "\n",
       "<span class=\"kn\">import</span> <span class=\"nn\">numpy</span> <span class=\"kn\">as</span> <span class=\"nn\">np</span>\n",
       "<span class=\"kn\">import</span> <span class=\"nn\">matplotlib.pyplot</span> <span class=\"kn\">as</span> <span class=\"nn\">plt</span>\n",
       "\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">mne</span> <span class=\"kn\">import</span> <a href=\"../../generated/mne.Epochs.html#mne.Epochs\"><span class=\"n\">Epochs</span></a><span class=\"p\">,</span> <a href=\"../../generated/mne.pick_types.html#mne.pick_types\"><span class=\"n\">pick_types</span></a>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">mne.io</span> <span class=\"kn\">import</span> <a href=\"../../generated/mne.concatenate_raws.html#mne.concatenate_raws\"><span class=\"n\">concatenate_raws</span></a>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">mne.io.edf</span> <span class=\"kn\">import</span> <a href=\"../../generated/mne.io.read_raw_edf.html#mne.io.read_raw_edf\"><span class=\"n\">read_raw_edf</span></a>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">mne.datasets</span> <span class=\"kn\">import</span> <span class=\"n\">eegbci</span>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">mne.event</span> <span class=\"kn\">import</span> <a href=\"../../generated/mne.find_events.html#mne.find_events\"><span class=\"n\">find_events</span></a>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">mne.decoding</span> <span class=\"kn\">import</span> <a href=\"../../generated/mne.decoding.CSP.html#mne.decoding.CSP\"><span class=\"n\">CSP</span></a>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">mne.channels</span> <span class=\"kn\">import</span> <a href=\"../../generated/mne.channels.read_layout.html#mne.channels.read_layout\"><span class=\"n\">read_layout</span></a>\n",
       "\n",
       "<span class=\"k\">print</span><span class=\"p\">(</span><span class=\"n\">__doc__</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"c\"># #############################################################################</span>\n",
       "<span class=\"c\"># # Set parameters and read data</span>\n",
       "\n",
       "<span class=\"c\"># avoid classification of evoked responses by using epochs that start 1s after</span>\n",
       "<span class=\"c\"># cue onset.</span>\n",
       "<span class=\"n\">tmin</span><span class=\"p\">,</span> <span class=\"n\">tmax</span> <span class=\"o\">=</span> <span class=\"o\">-</span><span class=\"mf\">1.</span><span class=\"p\">,</span> <span class=\"mf\">4.</span>\n",
       "<span class=\"n\">event_id</span> <span class=\"o\">=</span> <span class=\"nb\">dict</span><span class=\"p\">(</span><span class=\"n\">hands</span><span class=\"o\">=</span><span class=\"mi\">2</span><span class=\"p\">,</span> <span class=\"n\">feet</span><span class=\"o\">=</span><span class=\"mi\">3</span><span class=\"p\">)</span>\n",
       "<span class=\"n\">subject</span> <span class=\"o\">=</span> <span class=\"mi\">1</span>\n",
       "<span class=\"n\">runs</span> <span class=\"o\">=</span> <span class=\"p\">[</span><span class=\"mi\">6</span><span class=\"p\">,</span> <span class=\"mi\">10</span><span class=\"p\">,</span> <span class=\"mi\">14</span><span class=\"p\">]</span>  <span class=\"c\"># motor imagery: hands vs feet</span>\n",
       "\n",
       "<span class=\"n\">raw_fnames</span> <span class=\"o\">=</span> <span class=\"n\">eegbci</span><span class=\"o\">.</span><span class=\"n\">load_data</span><span class=\"p\">(</span><span class=\"n\">subject</span><span class=\"p\">,</span> <span class=\"n\">runs</span><span class=\"p\">)</span>\n",
       "<span class=\"n\">raw_files</span> <span class=\"o\">=</span> <span class=\"p\">[</span><a href=\"../../generated/mne.io.read_raw_edf.html#mne.io.read_raw_edf\"><span class=\"n\">read_raw_edf</span></a><span class=\"p\">(</span><span class=\"n\">f</span><span class=\"p\">,</span> <span class=\"n\">preload</span><span class=\"o\">=</span><span class=\"bp\">True</span><span class=\"p\">)</span> <span class=\"k\">for</span> <span class=\"n\">f</span> <span class=\"ow\">in</span> <span class=\"n\">raw_fnames</span><span class=\"p\">]</span>\n",
       "<span class=\"n\">raw</span> <span class=\"o\">=</span> <a href=\"../../generated/mne.concatenate_raws.html#mne.concatenate_raws\"><span class=\"n\">concatenate_raws</span></a><span class=\"p\">(</span><span class=\"n\">raw_files</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"c\"># strip channel names</span>\n",
       "<span class=\"n\">raw</span><span class=\"o\">.</span><span class=\"n\">info</span><span class=\"p\">[</span><span class=\"s\">&#39;ch_names&#39;</span><span class=\"p\">]</span> <span class=\"o\">=</span> <span class=\"p\">[</span><span class=\"n\">chn</span><span class=\"o\">.</span><span class=\"n\">strip</span><span class=\"p\">(</span><span class=\"s\">&#39;.&#39;</span><span class=\"p\">)</span> <span class=\"k\">for</span> <span class=\"n\">chn</span> <span class=\"ow\">in</span> <span class=\"n\">raw</span><span class=\"o\">.</span><span class=\"n\">info</span><span class=\"p\">[</span><span class=\"s\">&#39;ch_names&#39;</span><span class=\"p\">]]</span>\n",
       "\n",
       "<span class=\"c\"># Apply band-pass filter</span>\n",
       "<span class=\"n\">raw</span><span class=\"o\">.</span><span class=\"n\">filter</span><span class=\"p\">(</span><span class=\"mf\">7.</span><span class=\"p\">,</span> <span class=\"mf\">30.</span><span class=\"p\">,</span> <span class=\"n\">method</span><span class=\"o\">=</span><span class=\"s\">&#39;iir&#39;</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"n\">events</span> <span class=\"o\">=</span> <a href=\"../../generated/mne.find_events.html#mne.find_events\"><span class=\"n\">find_events</span></a><span class=\"p\">(</span><span class=\"n\">raw</span><span class=\"p\">,</span> <span class=\"n\">shortest_event</span><span class=\"o\">=</span><span class=\"mi\">0</span><span class=\"p\">,</span> <span class=\"n\">stim_channel</span><span class=\"o\">=</span><span class=\"s\">&#39;STI 014&#39;</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"n\">picks</span> <span class=\"o\">=</span> <a href=\"../../generated/mne.pick_types.html#mne.pick_types\"><span class=\"n\">pick_types</span></a><span class=\"p\">(</span><span class=\"n\">raw</span><span class=\"o\">.</span><span class=\"n\">info</span><span class=\"p\">,</span> <span class=\"n\">meg</span><span class=\"o\">=</span><span class=\"bp\">False</span><span class=\"p\">,</span> <span class=\"n\">eeg</span><span class=\"o\">=</span><span class=\"bp\">True</span><span class=\"p\">,</span> <span class=\"n\">stim</span><span class=\"o\">=</span><span class=\"bp\">False</span><span class=\"p\">,</span> <span class=\"n\">eog</span><span class=\"o\">=</span><span class=\"bp\">False</span><span class=\"p\">,</span>\n",
       "                   <span class=\"n\">exclude</span><span class=\"o\">=</span><span class=\"s\">&#39;bads&#39;</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"c\"># Read epochs (train will be done only between 1 and 2s)</span>\n",
       "<span class=\"c\"># Testing will be done with a running classifier</span>\n",
       "<span class=\"n\">epochs</span> <span class=\"o\">=</span> <a href=\"../../generated/mne.Epochs.html#mne.Epochs\"><span class=\"n\">Epochs</span></a><span class=\"p\">(</span><span class=\"n\">raw</span><span class=\"p\">,</span> <span class=\"n\">events</span><span class=\"p\">,</span> <span class=\"n\">event_id</span><span class=\"p\">,</span> <span class=\"n\">tmin</span><span class=\"p\">,</span> <span class=\"n\">tmax</span><span class=\"p\">,</span> <span class=\"n\">proj</span><span class=\"o\">=</span><span class=\"bp\">True</span><span class=\"p\">,</span> <span class=\"n\">picks</span><span class=\"o\">=</span><span class=\"n\">picks</span><span class=\"p\">,</span>\n",
       "                <span class=\"n\">baseline</span><span class=\"o\">=</span><span class=\"bp\">None</span><span class=\"p\">,</span> <span class=\"n\">preload</span><span class=\"o\">=</span><span class=\"bp\">True</span><span class=\"p\">,</span> <span class=\"n\">add_eeg_ref</span><span class=\"o\">=</span><span class=\"bp\">False</span><span class=\"p\">)</span>\n",
       "<span class=\"n\">epochs_train</span> <span class=\"o\">=</span> <span class=\"n\">epochs</span><span class=\"o\">.</span><span class=\"n\">crop</span><span class=\"p\">(</span><span class=\"n\">tmin</span><span class=\"o\">=</span><span class=\"mf\">1.</span><span class=\"p\">,</span> <span class=\"n\">tmax</span><span class=\"o\">=</span><span class=\"mf\">2.</span><span class=\"p\">,</span> <span class=\"n\">copy</span><span class=\"o\">=</span><span class=\"bp\">True</span><span class=\"p\">)</span>\n",
       "<span class=\"n\">labels</span> <span class=\"o\">=</span> <span class=\"n\">epochs</span><span class=\"o\">.</span><span class=\"n\">events</span><span class=\"p\">[:,</span> <span class=\"o\">-</span><span class=\"mi\">1</span><span class=\"p\">]</span> <span class=\"o\">-</span> <span class=\"mi\">2</span>\n",
       "\n",
       "<span class=\"c\">###############################################################################</span>\n",
       "<span class=\"c\"># Classification with linear discrimant analysis</span>\n",
       "\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">sklearn.lda</span> <span class=\"kn\">import</span> <span class=\"n\">LDA</span>  <span class=\"c\"># noqa</span>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">sklearn.cross_validation</span> <span class=\"kn\">import</span> <span class=\"n\">ShuffleSplit</span>  <span class=\"c\"># noqa</span>\n",
       "\n",
       "<span class=\"c\"># Assemble a classifier</span>\n",
       "<span class=\"n\">svc</span> <span class=\"o\">=</span> <span class=\"n\">LDA</span><span class=\"p\">()</span>\n",
       "<span class=\"n\">csp</span> <span class=\"o\">=</span> <a href=\"../../generated/mne.decoding.CSP.html#mne.decoding.CSP\"><span class=\"n\">CSP</span></a><span class=\"p\">(</span><span class=\"n\">n_components</span><span class=\"o\">=</span><span class=\"mi\">4</span><span class=\"p\">,</span> <span class=\"n\">reg</span><span class=\"o\">=</span><span class=\"bp\">None</span><span class=\"p\">,</span> <span class=\"n\">log</span><span class=\"o\">=</span><span class=\"bp\">True</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"c\"># Define a monte-carlo cross-validation generator (reduce variance):</span>\n",
       "<span class=\"n\">cv</span> <span class=\"o\">=</span> <span class=\"n\">ShuffleSplit</span><span class=\"p\">(</span><span class=\"nb\">len</span><span class=\"p\">(</span><span class=\"n\">labels</span><span class=\"p\">),</span> <span class=\"mi\">10</span><span class=\"p\">,</span> <span class=\"n\">test_size</span><span class=\"o\">=</span><span class=\"mf\">0.2</span><span class=\"p\">,</span> <span class=\"n\">random_state</span><span class=\"o\">=</span><span class=\"mi\">42</span><span class=\"p\">)</span>\n",
       "<span class=\"n\">scores</span> <span class=\"o\">=</span> <span class=\"p\">[]</span>\n",
       "<span class=\"n\">epochs_data</span> <span class=\"o\">=</span> <span class=\"n\">epochs</span><span class=\"o\">.</span><span class=\"n\">get_data</span><span class=\"p\">()</span>\n",
       "<span class=\"n\">epochs_data_train</span> <span class=\"o\">=</span> <span class=\"n\">epochs_train</span><span class=\"o\">.</span><span class=\"n\">get_data</span><span class=\"p\">()</span>\n",
       "\n",
       "<span class=\"c\"># Use scikit-learn Pipeline with cross_val_score function</span>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">sklearn.pipeline</span> <span class=\"kn\">import</span> <span class=\"n\">Pipeline</span>  <span class=\"c\"># noqa</span>\n",
       "<span class=\"kn\">from</span> <span class=\"nn\">sklearn.cross_validation</span> <span class=\"kn\">import</span> <span class=\"n\">cross_val_score</span>  <span class=\"c\"># noqa</span>\n",
       "<span class=\"n\">clf</span> <span class=\"o\">=</span> <span class=\"n\">Pipeline</span><span class=\"p\">([(</span><span class=\"s\">&#39;CSP&#39;</span><span class=\"p\">,</span> <span class=\"n\">csp</span><span class=\"p\">),</span> <span class=\"p\">(</span><span class=\"s\">&#39;SVC&#39;</span><span class=\"p\">,</span> <span class=\"n\">svc</span><span class=\"p\">)])</span>\n",
       "<span class=\"n\">scores</span> <span class=\"o\">=</span> <span class=\"n\">cross_val_score</span><span class=\"p\">(</span><span class=\"n\">clf</span><span class=\"p\">,</span> <span class=\"n\">epochs_data_train</span><span class=\"p\">,</span> <span class=\"n\">labels</span><span class=\"p\">,</span> <span class=\"n\">cv</span><span class=\"o\">=</span><span class=\"n\">cv</span><span class=\"p\">,</span> <span class=\"n\">n_jobs</span><span class=\"o\">=</span><span class=\"mi\">1</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"c\"># Printing the results</span>\n",
       "<span class=\"n\">class_balance</span> <span class=\"o\">=</span> <a href=\"http://docs.scipy.org/doc/numpy-1.6.0/reference/generated/numpy.mean.html#numpy.mean\"><span class=\"n\">np</span><span class=\"o\">.</span><span class=\"n\">mean</span></a><span class=\"p\">(</span><span class=\"n\">labels</span> <span class=\"o\">==</span> <span class=\"n\">labels</span><span class=\"p\">[</span><span class=\"mi\">0</span><span class=\"p\">])</span>\n",
       "<span class=\"n\">class_balance</span> <span class=\"o\">=</span> <span class=\"nb\">max</span><span class=\"p\">(</span><span class=\"n\">class_balance</span><span class=\"p\">,</span> <span class=\"mf\">1.</span> <span class=\"o\">-</span> <span class=\"n\">class_balance</span><span class=\"p\">)</span>\n",
       "<span class=\"k\">print</span><span class=\"p\">(</span><span class=\"s\">&quot;Classification accuracy: </span><span class=\"si\">%f</span><span class=\"s\"> / Chance level: </span><span class=\"si\">%f</span><span class=\"s\">&quot;</span> <span class=\"o\">%</span> <span class=\"p\">(</span><a href=\"http://docs.scipy.org/doc/numpy-1.6.0/reference/generated/numpy.mean.html#numpy.mean\"><span class=\"n\">np</span><span class=\"o\">.</span><span class=\"n\">mean</span></a><span class=\"p\">(</span><span class=\"n\">scores</span><span class=\"p\">),</span>\n",
       "                                                          <span class=\"n\">class_balance</span><span class=\"p\">))</span>\n",
       "\n",
       "<span class=\"c\"># plot CSP patterns estimated on full data for visualization</span>\n",
       "<span class=\"n\">csp</span><span class=\"o\">.</span><span class=\"n\">fit_transform</span><span class=\"p\">(</span><span class=\"n\">epochs_data</span><span class=\"p\">,</span> <span class=\"n\">labels</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"n\">evoked</span> <span class=\"o\">=</span> <span class=\"n\">epochs</span><span class=\"o\">.</span><span class=\"n\">average</span><span class=\"p\">()</span>\n",
       "<span class=\"n\">evoked</span><span class=\"o\">.</span><span class=\"n\">data</span> <span class=\"o\">=</span> <span class=\"n\">csp</span><span class=\"o\">.</span><span class=\"n\">patterns_</span><span class=\"o\">.</span><span class=\"n\">T</span>\n",
       "<span class=\"n\">evoked</span><span class=\"o\">.</span><span class=\"n\">times</span> <span class=\"o\">=</span> <a href=\"http://docs.scipy.org/doc/numpy-1.6.0/reference/generated/numpy.arange.html#numpy.arange\"><span class=\"n\">np</span><span class=\"o\">.</span><span class=\"n\">arange</span></a><span class=\"p\">(</span><span class=\"n\">evoked</span><span class=\"o\">.</span><span class=\"n\">data</span><span class=\"o\">.</span><span class=\"n\">shape</span><span class=\"p\">[</span><span class=\"mi\">0</span><span class=\"p\">])</span>\n",
       "\n",
       "<span class=\"n\">layout</span> <span class=\"o\">=</span> <a href=\"../../generated/mne.channels.read_layout.html#mne.channels.read_layout\"><span class=\"n\">read_layout</span></a><span class=\"p\">(</span><span class=\"s\">&#39;EEG1005&#39;</span><span class=\"p\">)</span>\n",
       "<span class=\"n\">evoked</span><span class=\"o\">.</span><span class=\"n\">plot_topomap</span><span class=\"p\">(</span><span class=\"n\">times</span><span class=\"o\">=</span><span class=\"p\">[</span><span class=\"mi\">0</span><span class=\"p\">,</span> <span class=\"mi\">1</span><span class=\"p\">,</span> <span class=\"mi\">2</span><span class=\"p\">,</span> <span class=\"mi\">61</span><span class=\"p\">,</span> <span class=\"mi\">62</span><span class=\"p\">,</span> <span class=\"mi\">63</span><span class=\"p\">],</span> <span class=\"n\">ch_type</span><span class=\"o\">=</span><span class=\"s\">&#39;eeg&#39;</span><span class=\"p\">,</span> <span class=\"n\">layout</span><span class=\"o\">=</span><span class=\"n\">layout</span><span class=\"p\">,</span>\n",
       "                    <span class=\"n\">scale_time</span><span class=\"o\">=</span><span class=\"mi\">1</span><span class=\"p\">,</span> <span class=\"n\">time_format</span><span class=\"o\">=</span><span class=\"s\">&#39;</span><span class=\"si\">%i</span><span class=\"s\">&#39;</span><span class=\"p\">,</span> <span class=\"n\">scale</span><span class=\"o\">=</span><span class=\"mi\">1</span><span class=\"p\">,</span>\n",
       "                    <span class=\"n\">unit</span><span class=\"o\">=</span><span class=\"s\">&#39;Patterns (AU)&#39;</span><span class=\"p\">,</span> <span class=\"n\">size</span><span class=\"o\">=</span><span class=\"mf\">1.5</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"c\">###############################################################################</span>\n",
       "<span class=\"c\"># Look at performance over time</span>\n",
       "\n",
       "<span class=\"n\">sfreq</span> <span class=\"o\">=</span> <span class=\"n\">raw</span><span class=\"o\">.</span><span class=\"n\">info</span><span class=\"p\">[</span><span class=\"s\">&#39;sfreq&#39;</span><span class=\"p\">]</span>\n",
       "<span class=\"n\">w_length</span> <span class=\"o\">=</span> <span class=\"nb\">int</span><span class=\"p\">(</span><span class=\"n\">sfreq</span> <span class=\"o\">*</span> <span class=\"mf\">0.5</span><span class=\"p\">)</span>   <span class=\"c\"># running classifier: window length</span>\n",
       "<span class=\"n\">w_step</span> <span class=\"o\">=</span> <span class=\"nb\">int</span><span class=\"p\">(</span><span class=\"n\">sfreq</span> <span class=\"o\">*</span> <span class=\"mf\">0.1</span><span class=\"p\">)</span>  <span class=\"c\"># running classifier: window step size</span>\n",
       "<span class=\"n\">w_start</span> <span class=\"o\">=</span> <a href=\"http://docs.scipy.org/doc/numpy-1.6.0/reference/generated/numpy.arange.html#numpy.arange\"><span class=\"n\">np</span><span class=\"o\">.</span><span class=\"n\">arange</span></a><span class=\"p\">(</span><span class=\"mi\">0</span><span class=\"p\">,</span> <span class=\"n\">epochs_data</span><span class=\"o\">.</span><span class=\"n\">shape</span><span class=\"p\">[</span><span class=\"mi\">2</span><span class=\"p\">]</span> <span class=\"o\">-</span> <span class=\"n\">w_length</span><span class=\"p\">,</span> <span class=\"n\">w_step</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"n\">scores_windows</span> <span class=\"o\">=</span> <span class=\"p\">[]</span>\n",
       "\n",
       "<span class=\"k\">for</span> <span class=\"n\">train_idx</span><span class=\"p\">,</span> <span class=\"n\">test_idx</span> <span class=\"ow\">in</span> <span class=\"n\">cv</span><span class=\"p\">:</span>\n",
       "    <span class=\"n\">y_train</span><span class=\"p\">,</span> <span class=\"n\">y_test</span> <span class=\"o\">=</span> <span class=\"n\">labels</span><span class=\"p\">[</span><span class=\"n\">train_idx</span><span class=\"p\">],</span> <span class=\"n\">labels</span><span class=\"p\">[</span><span class=\"n\">test_idx</span><span class=\"p\">]</span>\n",
       "\n",
       "    <span class=\"n\">X_train</span> <span class=\"o\">=</span> <span class=\"n\">csp</span><span class=\"o\">.</span><span class=\"n\">fit_transform</span><span class=\"p\">(</span><span class=\"n\">epochs_data_train</span><span class=\"p\">[</span><span class=\"n\">train_idx</span><span class=\"p\">],</span> <span class=\"n\">y_train</span><span class=\"p\">)</span>\n",
       "    <span class=\"n\">X_test</span> <span class=\"o\">=</span> <span class=\"n\">csp</span><span class=\"o\">.</span><span class=\"n\">transform</span><span class=\"p\">(</span><span class=\"n\">epochs_data_train</span><span class=\"p\">[</span><span class=\"n\">test_idx</span><span class=\"p\">])</span>\n",
       "\n",
       "    <span class=\"c\"># fit classifier</span>\n",
       "    <span class=\"n\">svc</span><span class=\"o\">.</span><span class=\"n\">fit</span><span class=\"p\">(</span><span class=\"n\">X_train</span><span class=\"p\">,</span> <span class=\"n\">y_train</span><span class=\"p\">)</span>\n",
       "\n",
       "    <span class=\"c\"># running classifier: test classifier on sliding window</span>\n",
       "    <span class=\"n\">score_this_window</span> <span class=\"o\">=</span> <span class=\"p\">[]</span>\n",
       "    <span class=\"k\">for</span> <span class=\"n\">n</span> <span class=\"ow\">in</span> <span class=\"n\">w_start</span><span class=\"p\">:</span>\n",
       "        <span class=\"n\">X_test</span> <span class=\"o\">=</span> <span class=\"n\">csp</span><span class=\"o\">.</span><span class=\"n\">transform</span><span class=\"p\">(</span><span class=\"n\">epochs_data</span><span class=\"p\">[</span><span class=\"n\">test_idx</span><span class=\"p\">][:,</span> <span class=\"p\">:,</span> <span class=\"n\">n</span><span class=\"p\">:(</span><span class=\"n\">n</span> <span class=\"o\">+</span> <span class=\"n\">w_length</span><span class=\"p\">)])</span>\n",
       "        <span class=\"n\">score_this_window</span><span class=\"o\">.</span><span class=\"n\">append</span><span class=\"p\">(</span><span class=\"n\">svc</span><span class=\"o\">.</span><span class=\"n\">score</span><span class=\"p\">(</span><span class=\"n\">X_test</span><span class=\"p\">,</span> <span class=\"n\">y_test</span><span class=\"p\">))</span>\n",
       "    <span class=\"n\">scores_windows</span><span class=\"o\">.</span><span class=\"n\">append</span><span class=\"p\">(</span><span class=\"n\">score_this_window</span><span class=\"p\">)</span>\n",
       "\n",
       "<span class=\"c\"># Plot scores over time</span>\n",
       "<span class=\"n\">w_times</span> <span class=\"o\">=</span> <span class=\"p\">(</span><span class=\"n\">w_start</span> <span class=\"o\">+</span> <span class=\"n\">w_length</span> <span class=\"o\">/</span> <span class=\"mf\">2.</span><span class=\"p\">)</span> <span class=\"o\">/</span> <span class=\"n\">sfreq</span> <span class=\"o\">+</span> <span class=\"n\">epochs</span><span class=\"o\">.</span><span class=\"n\">tmin</span>\n",
       "\n",
       "<a href=\"http://matplotlib.org/api/figure_api.html#matplotlib.figure\"><span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">figure</span></a><span class=\"p\">()</span>\n",
       "<a href=\"http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.plot\"><span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">plot</span></a><span class=\"p\">(</span><span class=\"n\">w_times</span><span class=\"p\">,</span> <a href=\"http://docs.scipy.org/doc/numpy-1.6.0/reference/generated/numpy.mean.html#numpy.mean\"><span class=\"n\">np</span><span class=\"o\">.</span><span class=\"n\">mean</span></a><span class=\"p\">(</span><span class=\"n\">scores_windows</span><span class=\"p\">,</span> <span class=\"mi\">0</span><span class=\"p\">),</span> <span class=\"n\">label</span><span class=\"o\">=</span><span class=\"s\">&#39;Score&#39;</span><span class=\"p\">)</span>\n",
       "<a href=\"http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.axvline\"><span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">axvline</span></a><span class=\"p\">(</span><span class=\"mi\">0</span><span class=\"p\">,</span> <span class=\"n\">linestyle</span><span class=\"o\">=</span><span class=\"s\">&#39;--&#39;</span><span class=\"p\">,</span> <span class=\"n\">color</span><span class=\"o\">=</span><span class=\"s\">&#39;k&#39;</span><span class=\"p\">,</span> <span class=\"n\">label</span><span class=\"o\">=</span><span class=\"s\">&#39;Onset&#39;</span><span class=\"p\">)</span>\n",
       "<span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">axhline</span><span class=\"p\">(</span><span class=\"mf\">0.5</span><span class=\"p\">,</span> <span class=\"n\">linestyle</span><span class=\"o\">=</span><span class=\"s\">&#39;-&#39;</span><span class=\"p\">,</span> <span class=\"n\">color</span><span class=\"o\">=</span><span class=\"s\">&#39;k&#39;</span><span class=\"p\">,</span> <span class=\"n\">label</span><span class=\"o\">=</span><span class=\"s\">&#39;Chance&#39;</span><span class=\"p\">)</span>\n",
       "<a href=\"http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.xlabel\"><span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">xlabel</span></a><span class=\"p\">(</span><span class=\"s\">&#39;time (s)&#39;</span><span class=\"p\">)</span>\n",
       "<a href=\"http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.ylabel\"><span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">ylabel</span></a><span class=\"p\">(</span><span class=\"s\">&#39;classification accuracy&#39;</span><span class=\"p\">)</span>\n",
       "<a href=\"http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.title\"><span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">title</span></a><span class=\"p\">(</span><span class=\"s\">&#39;Classification score over time&#39;</span><span class=\"p\">)</span>\n",
       "<a href=\"http://matplotlib.org/api/legend_api.html#matplotlib.legend\"><span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">legend</span></a><span class=\"p\">(</span><span class=\"n\">loc</span><span class=\"o\">=</span><span class=\"s\">&#39;lower right&#39;</span><span class=\"p\">)</span>\n",
       "<a href=\"http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.show\"><span class=\"n\">plt</span><span class=\"o\">.</span><span class=\"n\">show</span></a><span class=\"p\">()</span>\n",
       "</pre></div>\n",
       "</div>\n",
       "<p><strong>Total running time of the example:</strong>    3 seconds</p>\n",
       "<div class=\"social-button-container\">\n",
       "    <div class=\"social-button\">\n",
       "        <a href=\"https://twitter.com/share\" class=\"twitter-share-button\">Tweet</a>\n",
       "    </div>\n",
       "    <div class=\"social-button\">\n",
       "        <g:plusone annotation=\"inline\" width=\"120\" size=\"medium\"></g:plusone>\n",
       "    </div>\n",
       "    <div class=\"social-button\">\n",
       "        <div id=\"fb-root\"></div>\n",
       "        <script>(function(d, s, id) {\n",
       "            var js, fjs = d.getElementsByTagName(s)[0];\n",
       "            if (d.getElementById(id)) return;\n",
       "            js = d.createElement(s); js.id = id;\n",
       "            js.src = \"//connect.facebook.net/en_US/all.js#xfbml=1\";\n",
       "            fjs.parentNode.insertBefore(js, fjs);\n",
       "            }(document, 'script', 'facebook-jssdk'));\n",
       "        </script>\n",
       "        <div class=\"fb-like\" data-send=\"false\" data-width=\"450\" data-show-faces=\"false\"></div>\n",
       "    </div>\n",
       "</div></div>\n",
       "\n",
       "\n",
       "          </div>\n",
       "        </div>\n",
       "      </div>\n",
       "      <div class=\"clearer\"></div>\n",
       "    </div>\n",
       "    <div class=\"footer\" role=\"contentinfo\">\n",
       "        &copy; Copyright 2012-2015, MNE Developers.\n",
       "      Last updated on May 22, 2015.\n",
       "      Created using <a href=\"http://sphinx-doc.org/\">Sphinx</a> 1.4a0+.\n",
       "    </div>\n",
       "  </body>\n",
       "</html>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "HTML('http://martinos.org/mne/dev/auto_examples/decoding/plot_decoding_csp_eeg.html')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied (use --upgrade to upgrade): mne in /Users/Neuromancer/anaconda/envs/mne/lib/python2.7/site-packages\r\n"
     ]
    }
   ],
   "source": [
    "!pip install mne "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'/Users/Neuromancer/Documents/neuralcode/kaggle/Grasp-and-Lift_EEG_Detection'"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# -*- coding: utf-8 -*-\n",
    "\"\"\"\n",
    "Created on Mon Jun 29 14:00:37 2015\n",
    "\n",
    "@author: alexandrebarachant\n",
    "\n",
    "During a hand movement, the mu (~10Hz) and beta (~20Hz) oscillations are suppressed \n",
    "over the contralateral motor cortex, i.e. we can observe a reduction of the \n",
    "signal power in the corresponding frequency band. This effect is know as \n",
    "Event Related Desynchronization.\n",
    "\n",
    "I used MNE python to epoch signal corresponding to the hand movement, by assuming that \n",
    "the hand movement occur before the 'Replace' event.\n",
    "\n",
    "Using Common spatial patterns algorithm, i extract spatial filters that maximize \n",
    "the difference of variance during and after the movement, and then visualize the \n",
    "corresponding spectrum. \n",
    "\n",
    "For each subject, we should see a spot over the electrode C3 (Left motor cortex,\n",
    "corresponding to a right hand movement), and a decrease of the signal power in \n",
    "10 and 20 Hz during the movement (by reference to after the movement).\n",
    "\n",
    "Each subject has a different cortex organization, and a different apha and beta \n",
    "peak. The CSP algorithm is also sensitive to artefacts, so it could give eronous \n",
    "maps (for example subject 5 seems to trig on eye movements)\n",
    "\n",
    "\"\"\"\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from mne.io import RawArray\n",
    "from mne.channels import read_montage\n",
    "from mne.epochs import concatenate_epochs\n",
    "from mne import create_info, find_events, Epochs\n",
    "from mne.viz.topomap import _prepare_topo_plot, plot_topomap\n",
    "from mne.decoding import CSP\n",
    "\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn.cross_validation import cross_val_score, LeaveOneLabelOut\n",
    "from glob import glob\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.axes_grid1 import make_axes_locatable\n",
    "\n",
    "from scipy.signal import welch\n",
    "from mne import pick_types\n",
    "\n",
    "def creat_mne_raw_object(fname):\n",
    "    \"\"\"Create a mne raw instance from csv file\"\"\"\n",
    "    # Read EEG file\n",
    "    data = pd.read_csv(fname)\n",
    "    \n",
    "    # get chanel names\n",
    "    ch_names = list(data.columns[1:])\n",
    "    \n",
    "    # read EEG standard montage from mne\n",
    "    montage = read_montage('standard_1005',ch_names)\n",
    "\n",
    "    # events file\n",
    "    ev_fname = fname.replace('_data','_events')\n",
    "    # read event file\n",
    "    events = pd.read_csv(ev_fname)\n",
    "    events_names = events.columns[1:]\n",
    "    events_data = np.array(events[events_names]).T\n",
    "    \n",
    "    # concatenate event file and data\n",
    "    data = np.concatenate((1e-6*np.array(data[ch_names]).T,events_data))        \n",
    "    \n",
    "    # define channel type, the first is EEG, the last 6 are stimulations\n",
    "    ch_type = ['eeg']*len(ch_names) + ['stim']*6\n",
    "    \n",
    "    # create and populate MNE info structure\n",
    "    ch_names.extend(events_names)\n",
    "    info = create_info(ch_names,sfreq=500.0, ch_types=ch_type, montage=montage)\n",
    "    info['filename'] = fname\n",
    "    \n",
    "    # create raw object \n",
    "    raw = RawArray(data,info,verbose=False)\n",
    "    return raw\n",
    "\n",
    "subjects = range(1,12)\n",
    "auc = []\n",
    "for subject in subjects:\n",
    "    epochs_tot = []\n",
    "    \n",
    "    #eid = 'HandStart'\n",
    "    #fnames =  glob('../input/train/subj%d_series*_data.csv' % (subject))\n",
    "    fnames =  glob('train/subj%d_series*_data.csv' % (subject))\n",
    "    \n",
    "    session = []\n",
    "    y = []\n",
    "    for i,fname in enumerate(fnames):\n",
    "      \n",
    "        # read data \n",
    "        raw = creat_mne_raw_object(fname)\n",
    "        \n",
    "        # pick eeg signal\n",
    "        picks = pick_types(raw.info,eeg=True)\n",
    "        \n",
    "        # Filter data for alpha frequency and beta band\n",
    "        # Note that MNE implement a zero phase (filtfilt) filtering not compatible\n",
    "        # with the rule of future data.\n",
    "        raw.filter(7,35, picks=picks, method='iir', n_jobs=-1, verbose=False)\n",
    "        \n",
    "        # get event posision corresponding to Replace\n",
    "        events = find_events(raw,stim_channel='Replace', verbose=False)\n",
    "        # epochs signal for 1.5 second before the movement\n",
    "        epochs = Epochs(raw, events, {'during' : 1}, -2, -0.5, proj=False,\n",
    "                        picks=picks, baseline=None, preload=True,\n",
    "                        add_eeg_ref=False, verbose=False)\n",
    "        \n",
    "        epochs_tot.append(epochs)\n",
    "        session.extend([i]*len(epochs))\n",
    "        y.extend([1]*len(epochs))\n",
    "        \n",
    "        # epochs signal for 1.5 second after the movement, this correspond to the \n",
    "        # rest period.\n",
    "        epochs_rest = Epochs(raw, events, {'after' : 1}, 0.5, 2, proj=False,\n",
    "                        picks=picks, baseline=None, preload=True,\n",
    "                        add_eeg_ref=False, verbose=False)\n",
    "        \n",
    "        # Workaround to be able to concatenate epochs\n",
    "        epochs_rest.times = epochs.times\n",
    "        \n",
    "        epochs_tot.append(epochs_rest)\n",
    "        session.extend([i]*len(epochs_rest))\n",
    "        y.extend([-1]*len(epochs_rest))\n",
    "        \n",
    "    #concatenate all epochs\n",
    "    epochs = concatenate_epochs(epochs_tot)\n",
    "    \n",
    "    # get data \n",
    "    X = epochs.get_data()\n",
    "    y = np.array(y)\n",
    "    \n",
    "    # run CSP\n",
    "    csp = CSP(reg='lws')\n",
    "    csp.fit(X,y)\n",
    "    \n",
    "    # compute spatial filtered spectrum\n",
    "    po = []\n",
    "    for x in X:\n",
    "        f,p = welch(np.dot(csp.filters_[0,:].T,x), 500, nperseg=512)\n",
    "        po.append(p)\n",
    "    po = np.array(po)\n",
    "    \n",
    "    # prepare topoplot\n",
    "    _,epos,_,_,_ = _prepare_topo_plot(epochs,'eeg',None)\n",
    "    \n",
    "    # plot first pattern\n",
    "    pattern = csp.patterns_[0,:]\n",
    "    pattern -= pattern.mean()\n",
    "    ix = np.argmax(abs(pattern))\n",
    "    # the parttern is sign invariant.\n",
    "    # invert it for display purpose\n",
    "    if pattern[ix]>0:\n",
    "        sign = 1.0\n",
    "    else:\n",
    "        sign = -1.0\n",
    "    \n",
    "    fig, ax_topo = plt.subplots(1, 1, figsize=(12, 4))\n",
    "    title = 'Spatial Pattern'\n",
    "    fig.suptitle(title, fontsize=14)\n",
    "    img, _ = plot_topomap(sign*pattern,epos,axis=ax_topo,show=False)\n",
    "    divider = make_axes_locatable(ax_topo)\n",
    "    # add axes for colorbar\n",
    "    ax_colorbar = divider.append_axes('right', size='5%', pad=0.05)\n",
    "    plt.colorbar(img, cax=ax_colorbar)\n",
    "    \n",
    "    # plot spectrum\n",
    "    fix = (f>7) & (f<35)\n",
    "    ax_spectrum = divider.append_axes('right', size='300%', pad=1.2)\n",
    "    ax_spectrum.plot(f[fix],np.log(po[y==1][:,fix].mean(axis=0).T),'-r',lw=2)\n",
    "    ax_spectrum.plot(f[fix],np.log(po[y==-1][:,fix].mean(axis=0).T),'-b',lw=2)\n",
    "    ax_spectrum.set_xlabel('Frequency (Hz)')\n",
    "    ax_spectrum.set_ylabel('Power (dB)')\n",
    "    plt.grid()\n",
    "    plt.legend(['during','after'])\n",
    "    plt.title('Subject %d' % subject)\n",
    "    plt.show()\n",
    "    plt.savefig('spatial_pattern_subject_%d.png' % subject ,bbox_inches='tight')\n",
    "    \n",
    "    # run cross validation\n",
    "    clf = make_pipeline(CSP(),LogisticRegression())\n",
    "    cv = LeaveOneLabelOut(session)\n",
    "    auc.append(cross_val_score(clf,X,y,cv=cv,scoring='roc_auc').mean())\n",
    "    print(\"Subject %d : AUC cross val score : %.3f\" % (subject,auc[-1]))\n",
    "\n",
    "auc = pd.DataFrame(data=auc,index=subjects,columns=['auc'])\n",
    "auc.to_csv('cross_val_auc.csv')\n",
    "plt.figure(figsize=(4,4))\n",
    "auc.plot(kind='bar',y='auc')\n",
    "plt.xlabel('Subject')\n",
    "plt.ylabel('AUC')\n",
    "plt.title('During Vs. After classification')\n",
    "plt.savefig('cross_val_auc.png' ,bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['train/subj1_series1_data.csv',\n",
       " 'train/subj1_series2_data.csv',\n",
       " 'train/subj1_series3_data.csv',\n",
       " 'train/subj1_series4_data.csv',\n",
       " 'train/subj1_series5_data.csv',\n",
       " 'train/subj1_series6_data.csv',\n",
       " 'train/subj1_series7_data.csv',\n",
       " 'train/subj1_series8_data.csv']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "glob('train/subj%d_series*_data.csv' % (subject))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}