Note
Go to the end to download the full example code.
Plotting whitened data#
This tutorial demonstrates how to plot whitened evoked data.
Data are whitened for many processes, including dipole fitting, source localization and some decoding algorithms. Viewing whitened data thus gives a different perspective on the data that these algorithms operate on.
Let’s start by loading some data and computing a signal (spatial) covariance that we’ll consider to be noise.
# Authors: The MNE-Python contributors.
# License: BSD-3-Clause
# Copyright the MNE-Python contributors.
import mne
from mne.datasets import sample
Raw data with whitening#
Note
In the mne.io.Raw.plot()
with noise_cov
supplied,
you can press they “w” key to turn whitening on and off.
data_path = sample.data_path()
raw_fname = data_path / "MEG" / "sample" / "sample_audvis_filt-0-40_raw.fif"
raw = mne.io.read_raw_fif(raw_fname, preload=True)
events = mne.find_events(raw, stim_channel="STI 014")
event_id = {
"auditory/left": 1,
"auditory/right": 2,
"visual/left": 3,
"visual/right": 4,
"smiley": 5,
"button": 32,
}
reject = dict(grad=4000e-13, mag=4e-12, eog=150e-6)
epochs = mne.Epochs(raw, events, event_id=event_id, reject=reject)
# baseline noise cov, not a lot of samples
noise_cov = mne.compute_covariance(
epochs, tmax=0.0, method="shrunk", rank=None, verbose="error"
)
# butterfly mode shows the differences most clearly
raw.plot(events=events, butterfly=True)
raw.plot(noise_cov=noise_cov, events=events, butterfly=True)
Opening raw data file /home/circleci/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
Ready.
Reading 0 ... 41699 = 0.000 ... 277.709 secs...
319 events found on stim channel STI 014
Event IDs: [ 1 2 3 4 5 32]
Not setting metadata
319 matching events found
Setting baseline interval to [-0.19979521315838786, 0.0] s
Applying baseline correction (mode: mean)
Created an SSP operator (subspace dimension = 4)
4 projection items activated
Using qt as 2D backend.
Computing rank from covariance with rank=None
Using tolerance 8.3e-14 (2.2e-16 eps * 59 dim * 6.3 max singular value)
Estimated rank (eeg): 58
EEG: rank 58 computed from 59 data channels with 1 projector
Computing rank from covariance with rank=None
Using tolerance 1.9e-13 (2.2e-16 eps * 203 dim * 4.3 max singular value)
Estimated rank (grad): 203
GRAD: rank 203 computed from 203 data channels with 0 projectors
Computing rank from covariance with rank=None
Using tolerance 2.7e-14 (2.2e-16 eps * 102 dim * 1.2 max singular value)
Estimated rank (mag): 99
MAG: rank 99 computed from 102 data channels with 3 projectors
Epochs with whitening#
epochs.plot(events=True)
epochs.plot(noise_cov=noise_cov, events=True)
Using data from preloaded Raw for 319 events and 106 original time points ...
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on MAG : ['MEG 1711']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on MAG : ['MEG 1711']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
Rejecting epoch based on EOG : ['EOG 061']
51 bad epochs dropped
You seem to have overlapping epochs. Some event lines may be duplicated in the plot.
Using data from preloaded Raw for 20 events and 106 original time points ...
Using data from preloaded Raw for 20 events and 106 original time points ...
Using data from preloaded Raw for 20 events and 106 original time points ...
You seem to have overlapping epochs. Some event lines may be duplicated in the plot.
Computing rank from covariance with rank=None
Using tolerance 8.3e-14 (2.2e-16 eps * 59 dim * 6.3 max singular value)
Estimated rank (eeg): 58
EEG: rank 58 computed from 59 data channels with 1 projector
Computing rank from covariance with rank=None
Using tolerance 1.9e-13 (2.2e-16 eps * 203 dim * 4.3 max singular value)
Estimated rank (grad): 203
GRAD: rank 203 computed from 203 data channels with 0 projectors
Computing rank from covariance with rank=None
Using tolerance 2.7e-14 (2.2e-16 eps * 102 dim * 1.2 max singular value)
Estimated rank (mag): 99
MAG: rank 99 computed from 102 data channels with 3 projectors
Using data from preloaded Raw for 20 events and 106 original time points ...
Using data from preloaded Raw for 20 events and 106 original time points ...
Using data from preloaded Raw for 20 events and 106 original time points ...
Evoked data with whitening#
evoked = epochs.average()
evoked.plot(time_unit="s")
evoked.plot(noise_cov=noise_cov, time_unit="s")
Computing rank from covariance with rank=None
Using tolerance 8.3e-14 (2.2e-16 eps * 59 dim * 6.3 max singular value)
Estimated rank (eeg): 58
EEG: rank 58 computed from 59 data channels with 1 projector
Computing rank from covariance with rank=None
Using tolerance 1.9e-13 (2.2e-16 eps * 203 dim * 4.3 max singular value)
Estimated rank (grad): 203
GRAD: rank 203 computed from 203 data channels with 0 projectors
Computing rank from covariance with rank=None
Using tolerance 2.7e-14 (2.2e-16 eps * 102 dim * 1.2 max singular value)
Estimated rank (mag): 99
MAG: rank 99 computed from 102 data channels with 3 projectors
Evoked data with scaled whitening#
The mne.Evoked.plot_white()
function takes an additional step of
scaling the whitened plots to show how well the assumption of Gaussian
noise is satisfied by the data:
evoked.plot_white(noise_cov=noise_cov, time_unit="s")
NOTE: pick_types() is a legacy function. New code should use inst.pick(...).
Computing rank from covariance with rank=None
Using tolerance 8.3e-14 (2.2e-16 eps * 59 dim * 6.3 max singular value)
Estimated rank (eeg): 58
EEG: rank 58 computed from 59 data channels with 1 projector
Computing rank from covariance with rank=None
Using tolerance 1.9e-13 (2.2e-16 eps * 203 dim * 4.3 max singular value)
Estimated rank (grad): 203
GRAD: rank 203 computed from 203 data channels with 0 projectors
Computing rank from covariance with rank=None
Using tolerance 2.7e-14 (2.2e-16 eps * 102 dim * 1.2 max singular value)
Estimated rank (mag): 99
MAG: rank 99 computed from 102 data channels with 3 projectors
Created an SSP operator (subspace dimension = 4)
Computing rank from covariance with rank={'eeg': 58, 'grad': 203, 'mag': 99, 'meg': 302}
Setting small MEG eigenvalues to zero (without PCA)
Setting small EEG eigenvalues to zero (without PCA)
Created the whitener using a noise covariance matrix with rank 360 (4 small eigenvalues omitted)
Topographic plot with whitening#
evoked.comment = "All trials"
evoked.plot_topo(title="Evoked data")
evoked.plot_topo(noise_cov=noise_cov, title="Whitened evoked data")
Created an SSP operator (subspace dimension = 4)
Computing rank from covariance with rank=None
Using tolerance 3.6e-13 (2.2e-16 eps * 305 dim * 5.3 max singular value)
Estimated rank (mag + grad): 302
MEG: rank 302 computed from 305 data channels with 3 projectors
Using tolerance 8.3e-14 (2.2e-16 eps * 59 dim * 6.3 max singular value)
Estimated rank (eeg): 58
EEG: rank 58 computed from 59 data channels with 1 projector
Setting small MEG eigenvalues to zero (without PCA)
Setting small EEG eigenvalues to zero (without PCA)
Created the whitener using a noise covariance matrix with rank 360 (4 small eigenvalues omitted)
Total running time of the script: (0 minutes 27.687 seconds)