{
  "nbformat_minor": 0, 
  "nbformat": 4, 
  "cells": [
    {
      "execution_count": null, 
      "cell_type": "code", 
      "source": [
        "%matplotlib inline"
      ], 
      "outputs": [], 
      "metadata": {
        "collapsed": false
      }
    }, 
    {
      "source": [
        "\n#  XDAWN Decoding From EEG data\n\n\nERP decoding with Xdawn. For each event type, a set of spatial Xdawn filters\nare trained and applied on the signal. Channels are concatenated and rescaled\nto create features vectors that will be fed into a Logistic Regression.\n\nReferences\n----------\n\n.. [1] Rivet, B., Souloumiac, A., Attina, V., & Gibert, G. (2009). xDAWN\n       algorithm to enhance evoked potentials: application to brain-computer\n       interface. Biomedical Engineering, IEEE Transactions on, 56(8),\n       2035-2043.\n.. [2] Rivet, B., Cecotti, H., Souloumiac, A., Maby, E., & Mattout, J. (2011,\n       August). Theoretical analysis of xDAWN algorithm: application to an\n       efficient sensor selection in a P300 BCI. In Signal Processing\n       Conference, 2011 19th European (pp. 1382-1386). IEEE.\n\n"
      ], 
      "cell_type": "markdown", 
      "metadata": {}
    }, 
    {
      "execution_count": null, 
      "cell_type": "code", 
      "source": [
        "# Authors: Alexandre Barachant <alexandre.barachant@gmail.com>\n#\n# License: BSD (3-clause)\n\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nfrom sklearn.cross_validation import StratifiedKFold\nfrom sklearn.pipeline import make_pipeline\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.metrics import classification_report, confusion_matrix\nfrom sklearn.preprocessing import MinMaxScaler\n\nfrom mne import io, pick_types, read_events, Epochs\nfrom mne.datasets import sample\nfrom mne.preprocessing import Xdawn\nfrom mne.decoding import Vectorizer\nfrom mne.viz import tight_layout\n\n\nprint(__doc__)\n\ndata_path = sample.data_path()"
      ], 
      "outputs": [], 
      "metadata": {
        "collapsed": false
      }
    }, 
    {
      "source": [
        "Set parameters and read data\n\n"
      ], 
      "cell_type": "markdown", 
      "metadata": {}
    }, 
    {
      "execution_count": null, 
      "cell_type": "code", 
      "source": [
        "raw_fname = data_path + '/MEG/sample/sample_audvis_filt-0-40_raw.fif'\nevent_fname = data_path + '/MEG/sample/sample_audvis_filt-0-40_raw-eve.fif'\ntmin, tmax = -0.1, 0.3\nevent_id = dict(aud_l=1, aud_r=2, vis_l=3, vis_r=4)\n\n# Setup for reading the raw data\nraw = io.read_raw_fif(raw_fname, preload=True)\nraw.filter(1, 20)\nevents = read_events(event_fname)\n\npicks = pick_types(raw.info, meg=False, eeg=True, stim=False, eog=False,\n                   exclude='bads')\n\nepochs = Epochs(raw, events, event_id, tmin, tmax, proj=False,\n                picks=picks, baseline=None, preload=True,\n                verbose=False)\n\n# Create classification pipeline\nclf = make_pipeline(Xdawn(n_components=3),\n                    Vectorizer(),\n                    MinMaxScaler(),\n                    LogisticRegression(penalty='l1'))\n\n# Get the labels\nlabels = epochs.events[:, -1]\n\n# Cross validator\ncv = StratifiedKFold(y=labels, n_folds=10, shuffle=True, random_state=42)\n\n# Do cross-validation\npreds = np.empty(len(labels))\nfor train, test in cv:\n    clf.fit(epochs[train], labels[train])\n    preds[test] = clf.predict(epochs[test])\n\n# Classification report\ntarget_names = ['aud_l', 'aud_r', 'vis_l', 'vis_r']\nreport = classification_report(labels, preds, target_names=target_names)\nprint(report)\n\n# Normalized confusion matrix\ncm = confusion_matrix(labels, preds)\ncm_normalized = cm.astype(float) / cm.sum(axis=1)[:, np.newaxis]\n\n# Plot confusion matrix\nplt.imshow(cm_normalized, interpolation='nearest', cmap=plt.cm.Blues)\nplt.title('Normalized Confusion matrix')\nplt.colorbar()\ntick_marks = np.arange(len(target_names))\nplt.xticks(tick_marks, target_names, rotation=45)\nplt.yticks(tick_marks, target_names)\ntight_layout()\nplt.ylabel('True label')\nplt.xlabel('Predicted label')\nplt.show()"
      ], 
      "outputs": [], 
      "metadata": {
        "collapsed": false
      }
    }
  ], 
  "metadata": {
    "kernelspec": {
      "display_name": "Python 2", 
      "name": "python2", 
      "language": "python"
    }, 
    "language_info": {
      "mimetype": "text/x-python", 
      "nbconvert_exporter": "python", 
      "name": "python", 
      "file_extension": ".py", 
      "version": "2.7.13", 
      "pygments_lexer": "ipython2", 
      "codemirror_mode": {
        "version": 2, 
        "name": "ipython"
      }
    }
  }
}