Linear classifier on sensor data with plot patterns and filters

Decoding, a.k.a MVPA or supervised machine learning applied to MEG and EEG data in sensor space. Fit a linear classifier with the LinearModel object providing topographical patterns which are more neurophysiologically interpretable [1] than the classifier filters (weight vectors). The patterns explain how the MEG and EEG data were generated from the discriminant neural sources which are extracted by the filters. Note patterns/filters in MEG data are more similar than EEG data because the noise is less spatially correlated in MEG than EEG.

[1] Haufe, S., Meinecke, F., Görgen, K., Dähne, S., Haynes, J.-D., Blankertz, B., & Bießmann, F. (2014). On the interpretation of weight vectors of linear models in multivariate neuroimaging. NeuroImage, 87, 96–110. doi:10.1016/j.neuroimage.2013.10.067

# Authors: Alexandre Gramfort <>
#          Romain Trachel <>
# License: BSD (3-clause)

import mne
from mne import io
from mne.datasets import sample

from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression

# import a linear classifier from mne.decoding
from mne.decoding import LinearModel


data_path = sample.data_path()

Set parameters

raw_fname = data_path + '/MEG/sample/sample_audvis_filt-0-40_raw.fif'
event_fname = data_path + '/MEG/sample/sample_audvis_filt-0-40_raw-eve.fif'
tmin, tmax = -0.2, 0.5
event_id = dict(aud_l=1, vis_l=3)

# Setup for reading the raw data
raw = io.read_raw_fif(raw_fname, preload=True)
raw.filter(2, None, method='iir')  # replace baselining with high-pass
events = mne.read_events(event_fname)

# Read epochs
epochs = mne.Epochs(raw, events, event_id, tmin, tmax, proj=True,
                    decim=4, baseline=None, preload=True)

labels =[:, -1]

# get MEG and EEG data
meg_epochs = epochs.copy().pick_types(meg=True, eeg=False)
meg_data = meg_epochs.get_data().reshape(len(labels), -1)
eeg_epochs = epochs.copy().pick_types(meg=False, eeg=True)
eeg_data = eeg_epochs.get_data().reshape(len(labels), -1)


Opening raw data file /home/ubuntu/mne_data/MNE-sample-data/MEG/sample/sample_audvis_filt-0-40_raw.fif...
    Read a total of 4 projection items:
        PCA-v1 (1 x 102)  idle
        PCA-v2 (1 x 102)  idle
        PCA-v3 (1 x 102)  idle
        Average EEG reference (1 x 60)  idle
    Range : 6450 ... 48149 =     42.956 ...   320.665 secs
Current compensation grade : 0
add_eeg_ref defaults to True in 0.13, will default to False in 0.14, and will be removed in 0.15. We recommend to use add_eeg_ref=False and set_eeg_reference() instead.
Reading 0 ... 41699  =      0.000 ...   277.709 secs...
High-pass filtering at 2 Hz
145 matching events found
No baseline correction applied
The measurement information indicates a low-pass frequency of 40 Hz. The decim=4 parameter will result in a sampling frequency of 37.5384 Hz, which can cause aliasing artifacts.
add_eeg_ref defaults to True in 0.13, will default to False in 0.14, and will be removed in 0.15. We recommend to use add_eeg_ref=False and set_eeg_reference() instead.
Created an SSP operator (subspace dimension = 4)
4 projection items activated
Loading data for 145 events and 106 original time points ...
0 bad epochs dropped

Decoding in sensor space using a LogisticRegression classifier

clf = LogisticRegression()
sc = StandardScaler()

# create a linear model with LogisticRegression
model = LinearModel(clf)

# fit the classifier on MEG data
X = sc.fit_transform(meg_data), labels)
# plot patterns and filters
model.plot_patterns(, title='MEG Patterns')
model.plot_filters(, title='MEG Filters')

# fit the classifier on EEG data
X = sc.fit_transform(eeg_data), labels)
# plot patterns and filters
model.plot_patterns(, title='EEG Patterns')
model.plot_filters(, title='EEG Filters')
  • ../../_images/sphx_glr_plot_linear_model_patterns_001.png
  • ../../_images/sphx_glr_plot_linear_model_patterns_002.png
  • ../../_images/sphx_glr_plot_linear_model_patterns_003.png
  • ../../_images/sphx_glr_plot_linear_model_patterns_004.png

Total running time of the script: ( 0 minutes 14.507 seconds)

Generated by Sphinx-Gallery