Note
Click here to download the full example code
Decoding sensor space data with generalization across time and conditions#
This example runs the analysis described in [1]. It illustrates how one can fit a linear classifier to identify a discriminatory topography at a given time instant and subsequently assess whether this linear model can accurately predict all of the time samples of a second set of conditions.
# Authors: Jean-Remi King <jeanremi.king@gmail.com>
# Alexandre Gramfort <alexandre.gramfort@inria.fr>
# Denis Engemann <denis.engemann@gmail.com>
#
# License: BSD-3-Clause
import matplotlib.pyplot as plt
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
import mne
from mne.datasets import sample
from mne.decoding import GeneralizingEstimator
print(__doc__)
# Preprocess data
data_path = sample.data_path()
# Load and filter data, set up epochs
meg_path = data_path / 'MEG' / 'sample'
raw_fname = meg_path / 'sample_audvis_filt-0-40_raw.fif'
events_fname = meg_path / 'sample_audvis_filt-0-40_raw-eve.fif'
raw = mne.io.read_raw_fif(raw_fname, preload=True)
picks = mne.pick_types(raw.info, meg=True, exclude='bads') # Pick MEG channels
raw.filter(1., 30., fir_design='firwin') # Band pass filtering signals
events = mne.read_events(events_fname)
event_id = {'Auditory/Left': 1, 'Auditory/Right': 2,
'Visual/Left': 3, 'Visual/Right': 4}
tmin = -0.050
tmax = 0.400
# decimate to make the example faster to run, but then use verbose='error' in
# the Epochs constructor to suppress warning about decimation causing aliasing
decim = 2
epochs = mne.Epochs(raw, events, event_id=event_id, tmin=tmin, tmax=tmax,
proj=True, picks=picks, baseline=None, preload=True,
reject=dict(mag=5e-12), decim=decim, verbose='error')
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...
Filtering raw data in 1 contiguous segment
Setting up band-pass filter from 1 - 30 Hz
FIR filter parameters
---------------------
Designing a one-pass, zero-phase, non-causal bandpass filter:
- Windowed time-domain design (firwin) method
- Hamming window with 0.0194 passband ripple and 53 dB stopband attenuation
- Lower passband edge: 1.00
- Lower transition bandwidth: 1.00 Hz (-6 dB cutoff frequency: 0.50 Hz)
- Upper passband edge: 30.00 Hz
- Upper transition bandwidth: 7.50 Hz (-6 dB cutoff frequency: 33.75 Hz)
- Filter length: 497 samples (3.310 sec)
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done 1 out of 1 | elapsed: 0.0s remaining: 0.0s
[Parallel(n_jobs=1)]: Done 2 out of 2 | elapsed: 0.0s remaining: 0.0s
[Parallel(n_jobs=1)]: Done 3 out of 3 | elapsed: 0.0s remaining: 0.0s
[Parallel(n_jobs=1)]: Done 4 out of 4 | elapsed: 0.0s remaining: 0.0s
[Parallel(n_jobs=1)]: Done 366 out of 366 | elapsed: 0.6s finished
We will train the classifier on all left visual vs auditory trials and test on all right visual vs auditory trials.
clf = make_pipeline(
StandardScaler(),
LogisticRegression(solver='liblinear') # liblinear is faster than lbfgs
)
time_gen = GeneralizingEstimator(clf, scoring='roc_auc', n_jobs=None,
verbose=True)
# Fit classifiers on the epochs where the stimulus was presented to the left.
# Note that the experimental condition y indicates auditory or visual
time_gen.fit(X=epochs['Left'].get_data(),
y=epochs['Left'].events[:, 2] > 2)
0%| | Fitting GeneralizingEstimator : 0/35 [00:00<?, ?it/s]
6%|5 | Fitting GeneralizingEstimator : 2/35 [00:00<00:00, 58.43it/s]
14%|#4 | Fitting GeneralizingEstimator : 5/35 [00:00<00:00, 73.99it/s]
20%|## | Fitting GeneralizingEstimator : 7/35 [00:00<00:00, 68.89it/s]
31%|###1 | Fitting GeneralizingEstimator : 11/35 [00:00<00:00, 82.31it/s]
43%|####2 | Fitting GeneralizingEstimator : 15/35 [00:00<00:00, 90.27it/s]
51%|#####1 | Fitting GeneralizingEstimator : 18/35 [00:00<00:00, 90.02it/s]
60%|###### | Fitting GeneralizingEstimator : 21/35 [00:00<00:00, 89.85it/s]
71%|#######1 | Fitting GeneralizingEstimator : 25/35 [00:00<00:00, 94.00it/s]
77%|#######7 | Fitting GeneralizingEstimator : 27/35 [00:00<00:00, 89.31it/s]
89%|########8 | Fitting GeneralizingEstimator : 31/35 [00:00<00:00, 92.95it/s]
97%|#########7| Fitting GeneralizingEstimator : 34/35 [00:00<00:00, 92.52it/s]
100%|##########| Fitting GeneralizingEstimator : 35/35 [00:00<00:00, 91.98it/s]
Score on the epochs where the stimulus was presented to the right.
scores = time_gen.score(X=epochs['Right'].get_data(),
y=epochs['Right'].events[:, 2] > 2)
0%| | Scoring GeneralizingEstimator : 0/1225 [00:00<?, ?it/s]
1%|1 | Scoring GeneralizingEstimator : 18/1225 [00:00<00:02, 527.49it/s]
3%|3 | Scoring GeneralizingEstimator : 40/1225 [00:00<00:02, 590.55it/s]
5%|5 | Scoring GeneralizingEstimator : 62/1225 [00:00<00:01, 610.09it/s]
7%|6 | Scoring GeneralizingEstimator : 85/1225 [00:00<00:01, 629.49it/s]
9%|8 | Scoring GeneralizingEstimator : 107/1225 [00:00<00:01, 634.47it/s]
11%|# | Scoring GeneralizingEstimator : 130/1225 [00:00<00:01, 643.49it/s]
12%|#2 | Scoring GeneralizingEstimator : 153/1225 [00:00<00:01, 650.00it/s]
14%|#4 | Scoring GeneralizingEstimator : 176/1225 [00:00<00:01, 654.74it/s]
16%|#6 | Scoring GeneralizingEstimator : 198/1225 [00:00<00:01, 654.23it/s]
18%|#8 | Scoring GeneralizingEstimator : 221/1225 [00:00<00:01, 657.86it/s]
20%|#9 | Scoring GeneralizingEstimator : 244/1225 [00:00<00:01, 660.82it/s]
22%|##1 | Scoring GeneralizingEstimator : 267/1225 [00:00<00:01, 663.12it/s]
24%|##3 | Scoring GeneralizingEstimator : 291/1225 [00:00<00:01, 668.01it/s]
26%|##5 | Scoring GeneralizingEstimator : 314/1225 [00:00<00:01, 669.46it/s]
28%|##7 | Scoring GeneralizingEstimator : 337/1225 [00:00<00:01, 670.66it/s]
29%|##9 | Scoring GeneralizingEstimator : 360/1225 [00:00<00:01, 671.64it/s]
31%|###1 | Scoring GeneralizingEstimator : 383/1225 [00:00<00:01, 672.63it/s]
33%|###3 | Scoring GeneralizingEstimator : 405/1225 [00:00<00:01, 671.05it/s]
35%|###5 | Scoring GeneralizingEstimator : 429/1225 [00:00<00:01, 674.43it/s]
37%|###6 | Scoring GeneralizingEstimator : 452/1225 [00:00<00:01, 674.77it/s]
39%|###8 | Scoring GeneralizingEstimator : 475/1225 [00:00<00:01, 675.29it/s]
41%|#### | Scoring GeneralizingEstimator : 498/1225 [00:00<00:01, 675.61it/s]
43%|####2 | Scoring GeneralizingEstimator : 521/1225 [00:00<00:01, 675.83it/s]
44%|####4 | Scoring GeneralizingEstimator : 544/1225 [00:00<00:01, 676.14it/s]
46%|####6 | Scoring GeneralizingEstimator : 568/1225 [00:00<00:00, 678.52it/s]
48%|####8 | Scoring GeneralizingEstimator : 591/1225 [00:00<00:00, 678.77it/s]
50%|##### | Scoring GeneralizingEstimator : 614/1225 [00:00<00:00, 679.03it/s]
52%|#####2 | Scoring GeneralizingEstimator : 637/1225 [00:00<00:00, 679.18it/s]
54%|#####3 | Scoring GeneralizingEstimator : 660/1225 [00:00<00:00, 679.30it/s]
56%|#####5 | Scoring GeneralizingEstimator : 682/1225 [00:01<00:00, 677.35it/s]
58%|#####7 | Scoring GeneralizingEstimator : 705/1225 [00:01<00:00, 677.56it/s]
59%|#####9 | Scoring GeneralizingEstimator : 728/1225 [00:01<00:00, 677.84it/s]
61%|######1 | Scoring GeneralizingEstimator : 751/1225 [00:01<00:00, 677.99it/s]
63%|######3 | Scoring GeneralizingEstimator : 774/1225 [00:01<00:00, 678.25it/s]
65%|######5 | Scoring GeneralizingEstimator : 797/1225 [00:01<00:00, 678.47it/s]
67%|######6 | Scoring GeneralizingEstimator : 820/1225 [00:01<00:00, 678.54it/s]
69%|######8 | Scoring GeneralizingEstimator : 843/1225 [00:01<00:00, 678.76it/s]
71%|####### | Scoring GeneralizingEstimator : 866/1225 [00:01<00:00, 678.87it/s]
73%|#######2 | Scoring GeneralizingEstimator : 889/1225 [00:01<00:00, 678.88it/s]
74%|#######4 | Scoring GeneralizingEstimator : 912/1225 [00:01<00:00, 678.87it/s]
76%|#######6 | Scoring GeneralizingEstimator : 935/1225 [00:01<00:00, 678.96it/s]
78%|#######8 | Scoring GeneralizingEstimator : 957/1225 [00:01<00:00, 677.35it/s]
80%|######## | Scoring GeneralizingEstimator : 981/1225 [00:01<00:00, 679.23it/s]
82%|########1 | Scoring GeneralizingEstimator : 1003/1225 [00:01<00:00, 677.68it/s]
84%|########3 | Scoring GeneralizingEstimator : 1026/1225 [00:01<00:00, 677.75it/s]
86%|########5 | Scoring GeneralizingEstimator : 1049/1225 [00:01<00:00, 677.70it/s]
88%|########7 | Scoring GeneralizingEstimator : 1072/1225 [00:01<00:00, 676.37it/s]
89%|########9 | Scoring GeneralizingEstimator : 1095/1225 [00:01<00:00, 676.35it/s]
91%|#########1| Scoring GeneralizingEstimator : 1118/1225 [00:01<00:00, 676.32it/s]
93%|#########3| Scoring GeneralizingEstimator : 1141/1225 [00:01<00:00, 676.46it/s]
95%|#########5| Scoring GeneralizingEstimator : 1164/1225 [00:01<00:00, 676.57it/s]
97%|#########6| Scoring GeneralizingEstimator : 1187/1225 [00:01<00:00, 676.56it/s]
99%|#########8| Scoring GeneralizingEstimator : 1210/1225 [00:01<00:00, 676.53it/s]
100%|##########| Scoring GeneralizingEstimator : 1225/1225 [00:01<00:00, 676.57it/s]
100%|##########| Scoring GeneralizingEstimator : 1225/1225 [00:01<00:00, 674.74it/s]
Plot
fig, ax = plt.subplots(1)
im = ax.matshow(scores, vmin=0, vmax=1., cmap='RdBu_r', origin='lower',
extent=epochs.times[[0, -1, 0, -1]])
ax.axhline(0., color='k')
ax.axvline(0., color='k')
ax.xaxis.set_ticks_position('bottom')
ax.set_xlabel('Testing Time (s)')
ax.set_ylabel('Training Time (s)')
ax.set_title('Generalization across time and condition')
plt.colorbar(im, ax=ax)
plt.show()
References#
Total running time of the script: ( 0 minutes 6.620 seconds)
Estimated memory usage: 129 MB