Note
Go to the end to download the full example code.
Comparison of embeddings of covariance matrices¶
Comparison of several embeddings of a set of ERP covariance matrices extracted on MEG data: SE, LLE and t-SNE
Spectral Embedding (SE) is based on computing the low-dimensional representation that best preserves locality instead of local linearity in LLE [1].
Locally Linear Embedding (LLE) assumes that the local neighborhood of a matrix on the manifold can be well approximated by the affine subspace spanned by the k-nearest neighbors of the matrix and finds a low-dimensional embedding of the data based on these affine approximations.
t-SNE reduces SPD matrices into lower dimensional SPD matrices by computing conditional probabilities that represent similarities [2]. This fully Riemannian algorithm helps preserve the non-Euclidean structure of the data.
# Authors: Pedro Rodrigues <pedro.rodrigues01@gmail.com>,
# Gabriel Wagner vom Berg <gabriel@bccn-berlin.de>
# Thibault de Surrel <thibault.de-surrel@lamsade.dauphine.fr>
#
# License: BSD (3-clause)
import matplotlib.pyplot as plt
import mne
from mne import io
from mne.datasets import sample
from sklearn.model_selection import train_test_split
from pyriemann.estimation import XdawnCovariances
from pyriemann.utils.viz import plot_embedding
print(__doc__)
Set parameters and read data¶
data_path = str(sample.data_path())
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., 1
event_id = dict(aud_l=1, aud_r=2, vis_l=3, vis_r=4)
# Setup for reading the raw data
raw = io.Raw(raw_fname, preload=True, verbose=False)
raw.filter(2, None, method="iir") # replace baselining with high-pass
events = mne.read_events(event_fname)
raw.info["bads"] = ["MEG 2443"] # set bad channels
picks = mne.pick_types(raw.info, meg=True, eeg=False, stim=False, eog=False,
exclude="bads")
# Read epochs
epochs = mne.Epochs(raw, events, event_id, tmin, tmax, proj=False,
picks=picks, baseline=None, preload=True, verbose=False)
X = epochs.get_data(copy=False)
y = epochs.events[:, -1]
Using default location ~/mne_data for sample...
Fetching 1 file for the sample dataset ...
0%| | 0.00/1.65G [00:00<?, ?B/s]
0%| | 163k/1.65G [00:00<17:47, 1.55MB/s]
0%| | 1.97M/1.65G [00:00<02:29, 11.1MB/s]
0%|▏ | 7.97M/1.65G [00:00<00:49, 33.1MB/s]
1%|▎ | 14.3M/1.65G [00:00<00:36, 44.8MB/s]
1%|▍ | 20.7M/1.65G [00:00<00:31, 51.7MB/s]
2%|▌ | 27.0M/1.65G [00:00<00:29, 55.7MB/s]
2%|▋ | 33.4M/1.65G [00:00<00:27, 58.4MB/s]
2%|▉ | 39.8M/1.65G [00:00<00:26, 60.1MB/s]
3%|█ | 46.1M/1.65G [00:00<00:26, 60.9MB/s]
3%|█▏ | 52.5M/1.65G [00:01<00:25, 61.8MB/s]
4%|█▎ | 58.9M/1.65G [00:01<00:25, 62.6MB/s]
4%|█▍ | 65.2M/1.65G [00:01<00:25, 62.7MB/s]
4%|█▌ | 71.7M/1.65G [00:01<00:24, 63.3MB/s]
5%|█▋ | 78.1M/1.65G [00:01<00:24, 63.6MB/s]
5%|█▉ | 84.5M/1.65G [00:01<00:24, 63.9MB/s]
6%|██ | 91.0M/1.65G [00:01<00:24, 64.0MB/s]
6%|██▏ | 97.4M/1.65G [00:01<00:24, 64.0MB/s]
6%|██▍ | 104M/1.65G [00:01<00:24, 63.8MB/s]
7%|██▌ | 110M/1.65G [00:01<00:24, 63.8MB/s]
7%|██▋ | 117M/1.65G [00:02<00:23, 64.1MB/s]
7%|██▊ | 123M/1.65G [00:02<00:23, 64.4MB/s]
8%|██▉ | 130M/1.65G [00:02<00:23, 64.6MB/s]
8%|███▏ | 136M/1.65G [00:02<00:23, 64.6MB/s]
9%|███▎ | 143M/1.65G [00:02<00:23, 64.5MB/s]
9%|███▍ | 149M/1.65G [00:02<00:23, 64.3MB/s]
9%|███▌ | 155M/1.65G [00:02<00:23, 64.3MB/s]
10%|███▋ | 162M/1.65G [00:02<00:23, 64.2MB/s]
10%|███▊ | 168M/1.65G [00:02<00:23, 64.4MB/s]
11%|████ | 175M/1.65G [00:02<00:23, 64.2MB/s]
11%|████▏ | 181M/1.65G [00:03<00:23, 63.5MB/s]
11%|████▎ | 188M/1.65G [00:03<00:23, 62.6MB/s]
12%|████▍ | 194M/1.65G [00:03<00:23, 61.8MB/s]
12%|████▌ | 200M/1.65G [00:03<00:23, 61.4MB/s]
12%|████▋ | 206M/1.65G [00:03<00:23, 61.0MB/s]
13%|████▉ | 212M/1.65G [00:03<00:23, 60.9MB/s]
13%|█████ | 218M/1.65G [00:03<00:23, 60.9MB/s]
14%|█████▏ | 225M/1.65G [00:03<00:23, 60.8MB/s]
14%|█████▎ | 231M/1.65G [00:03<00:23, 60.5MB/s]
14%|█████▍ | 237M/1.65G [00:03<00:23, 60.6MB/s]
15%|█████▌ | 243M/1.65G [00:04<00:23, 60.7MB/s]
15%|█████▋ | 249M/1.65G [00:04<00:23, 60.6MB/s]
15%|█████▊ | 255M/1.65G [00:04<00:23, 60.6MB/s]
16%|█████▉ | 261M/1.65G [00:04<00:22, 60.6MB/s]
16%|██████▏ | 267M/1.65G [00:04<00:22, 60.3MB/s]
17%|██████▎ | 273M/1.65G [00:04<00:22, 60.3MB/s]
17%|██████▍ | 279M/1.65G [00:04<00:22, 60.4MB/s]
17%|██████▌ | 285M/1.65G [00:04<00:22, 60.1MB/s]
18%|██████▋ | 291M/1.65G [00:04<00:22, 60.4MB/s]
18%|██████▊ | 297M/1.65G [00:04<00:22, 60.1MB/s]
18%|██████▉ | 303M/1.65G [00:05<00:22, 60.3MB/s]
19%|███████ | 309M/1.65G [00:05<00:22, 60.4MB/s]
19%|███████▎ | 316M/1.65G [00:05<00:22, 60.5MB/s]
19%|███████▍ | 322M/1.65G [00:05<00:21, 60.6MB/s]
20%|███████▌ | 328M/1.65G [00:05<00:21, 60.5MB/s]
20%|███████▋ | 334M/1.65G [00:05<00:21, 60.6MB/s]
21%|███████▊ | 340M/1.65G [00:05<00:21, 60.7MB/s]
21%|███████▉ | 346M/1.65G [00:05<00:21, 60.8MB/s]
21%|████████ | 352M/1.65G [00:05<00:21, 60.6MB/s]
22%|████████▏ | 358M/1.65G [00:05<00:21, 60.5MB/s]
22%|████████▎ | 364M/1.65G [00:06<00:21, 60.6MB/s]
22%|████████▌ | 370M/1.65G [00:06<00:21, 60.5MB/s]
23%|████████▋ | 376M/1.65G [00:06<00:21, 60.6MB/s]
23%|████████▊ | 382M/1.65G [00:06<00:21, 60.5MB/s]
24%|████████▉ | 389M/1.65G [00:06<00:20, 61.1MB/s]
24%|█████████ | 395M/1.65G [00:06<00:20, 61.8MB/s]
24%|█████████▏ | 401M/1.65G [00:06<00:20, 62.5MB/s]
25%|█████████▎ | 408M/1.65G [00:06<00:19, 62.7MB/s]
25%|█████████▌ | 414M/1.65G [00:06<00:19, 63.1MB/s]
25%|█████████▋ | 420M/1.65G [00:06<00:19, 63.1MB/s]
26%|█████████▊ | 427M/1.65G [00:07<00:19, 62.7MB/s]
26%|█████████▉ | 433M/1.65G [00:07<00:19, 63.1MB/s]
27%|██████████ | 440M/1.65G [00:07<00:19, 63.4MB/s]
27%|██████████▎ | 446M/1.65G [00:07<00:18, 63.6MB/s]
27%|██████████▍ | 452M/1.65G [00:07<00:18, 63.8MB/s]
28%|██████████▌ | 459M/1.65G [00:07<00:18, 64.1MB/s]
28%|██████████▋ | 465M/1.65G [00:07<00:18, 63.7MB/s]
29%|██████████▊ | 472M/1.65G [00:07<00:18, 63.6MB/s]
29%|██████████▉ | 478M/1.65G [00:07<00:18, 63.5MB/s]
29%|███████████▏ | 484M/1.65G [00:07<00:18, 63.5MB/s]
30%|███████████▎ | 491M/1.65G [00:08<00:18, 63.5MB/s]
30%|███████████▍ | 497M/1.65G [00:08<00:18, 63.6MB/s]
30%|███████████▌ | 503M/1.65G [00:08<00:18, 63.6MB/s]
31%|███████████▋ | 510M/1.65G [00:08<00:17, 63.7MB/s]
31%|███████████▊ | 516M/1.65G [00:08<00:17, 63.6MB/s]
32%|████████████ | 523M/1.65G [00:08<00:17, 63.6MB/s]
32%|████████████▏ | 529M/1.65G [00:08<00:17, 62.8MB/s]
32%|████████████▎ | 535M/1.65G [00:08<00:18, 61.9MB/s]
33%|████████████▍ | 541M/1.65G [00:08<00:18, 61.5MB/s]
33%|████████████▌ | 548M/1.65G [00:08<00:18, 60.9MB/s]
33%|████████████▋ | 554M/1.65G [00:09<00:18, 60.9MB/s]
34%|████████████▊ | 560M/1.65G [00:09<00:17, 60.8MB/s]
34%|█████████████ | 566M/1.65G [00:09<00:17, 60.8MB/s]
35%|█████████████▏ | 572M/1.65G [00:09<00:17, 60.7MB/s]
35%|█████████████▎ | 578M/1.65G [00:09<00:17, 60.8MB/s]
35%|█████████████▍ | 584M/1.65G [00:09<00:17, 60.9MB/s]
36%|█████████████▌ | 590M/1.65G [00:09<00:17, 60.8MB/s]
36%|█████████████▋ | 596M/1.65G [00:09<00:17, 60.8MB/s]
36%|█████████████▊ | 602M/1.65G [00:09<00:17, 60.5MB/s]
37%|█████████████▉ | 608M/1.65G [00:09<00:17, 60.4MB/s]
37%|██████████████▏ | 614M/1.65G [00:10<00:17, 60.2MB/s]
38%|██████████████▎ | 620M/1.65G [00:10<00:17, 59.9MB/s]
38%|██████████████▍ | 627M/1.65G [00:10<00:16, 60.8MB/s]
38%|██████████████▌ | 633M/1.65G [00:10<00:16, 61.6MB/s]
39%|██████████████▋ | 639M/1.65G [00:10<00:16, 61.5MB/s]
39%|██████████████▊ | 646M/1.65G [00:10<00:16, 62.2MB/s]
39%|██████████████▉ | 652M/1.65G [00:10<00:15, 62.6MB/s]
40%|███████████████▏ | 658M/1.65G [00:10<00:17, 56.4MB/s]
40%|███████████████▎ | 664M/1.65G [00:10<00:18, 53.7MB/s]
41%|███████████████▍ | 670M/1.65G [00:11<00:17, 56.1MB/s]
41%|███████████████▌ | 676M/1.65G [00:11<00:16, 57.7MB/s]
41%|███████████████▋ | 683M/1.65G [00:11<00:16, 59.3MB/s]
42%|███████████████▊ | 689M/1.65G [00:11<00:15, 60.5MB/s]
42%|███████████████▉ | 696M/1.65G [00:11<00:15, 61.4MB/s]
42%|████████████████▏ | 702M/1.65G [00:11<00:15, 61.7MB/s]
43%|████████████████▎ | 708M/1.65G [00:11<00:15, 62.4MB/s]
43%|████████████████▍ | 715M/1.65G [00:11<00:14, 63.0MB/s]
44%|████████████████▌ | 721M/1.65G [00:11<00:14, 62.5MB/s]
44%|████████████████▋ | 727M/1.65G [00:11<00:14, 63.0MB/s]
44%|████████████████▊ | 734M/1.65G [00:12<00:14, 63.4MB/s]
45%|█████████████████ | 740M/1.65G [00:12<00:14, 63.3MB/s]
45%|█████████████████▏ | 746M/1.65G [00:12<00:14, 63.4MB/s]
46%|█████████████████▎ | 753M/1.65G [00:12<00:14, 63.7MB/s]
46%|█████████████████▍ | 759M/1.65G [00:12<00:14, 63.7MB/s]
46%|█████████████████▌ | 766M/1.65G [00:12<00:14, 63.2MB/s]
47%|█████████████████▊ | 772M/1.65G [00:12<00:13, 63.4MB/s]
47%|█████████████████▉ | 778M/1.65G [00:12<00:13, 63.4MB/s]
47%|██████████████████ | 785M/1.65G [00:12<00:13, 62.8MB/s]
48%|██████████████████▏ | 791M/1.65G [00:12<00:13, 62.9MB/s]
48%|██████████████████▎ | 798M/1.65G [00:13<00:13, 63.4MB/s]
49%|██████████████████▍ | 804M/1.65G [00:13<00:13, 62.9MB/s]
49%|██████████████████▋ | 810M/1.65G [00:13<00:13, 62.6MB/s]
49%|██████████████████▊ | 817M/1.65G [00:13<00:13, 63.0MB/s]
50%|██████████████████▉ | 823M/1.65G [00:13<00:13, 63.4MB/s]
50%|███████████████████ | 829M/1.65G [00:13<00:13, 63.0MB/s]
51%|███████████████████▏ | 836M/1.65G [00:13<00:12, 63.5MB/s]
51%|███████████████████▎ | 842M/1.65G [00:13<00:12, 63.9MB/s]
51%|███████████████████▌ | 849M/1.65G [00:13<00:12, 63.1MB/s]
52%|███████████████████▋ | 855M/1.65G [00:13<00:12, 63.4MB/s]
52%|███████████████████▊ | 862M/1.65G [00:14<00:12, 63.8MB/s]
53%|███████████████████▉ | 868M/1.65G [00:14<00:12, 63.0MB/s]
53%|████████████████████ | 874M/1.65G [00:14<00:12, 63.4MB/s]
53%|████████████████████▎ | 881M/1.65G [00:14<00:12, 63.6MB/s]
54%|████████████████████▍ | 887M/1.65G [00:14<00:12, 63.8MB/s]
54%|████████████████████▌ | 894M/1.65G [00:14<00:12, 63.1MB/s]
54%|████████████████████▋ | 900M/1.65G [00:14<00:11, 63.7MB/s]
55%|████████████████████▊ | 907M/1.65G [00:14<00:11, 63.8MB/s]
55%|████████████████████▉ | 913M/1.65G [00:14<00:11, 64.1MB/s]
56%|█████████████████████▏ | 919M/1.65G [00:14<00:11, 61.6MB/s]
56%|█████████████████████▎ | 926M/1.65G [00:15<00:11, 62.4MB/s]
56%|█████████████████████▍ | 932M/1.65G [00:15<00:11, 63.1MB/s]
57%|█████████████████████▌ | 939M/1.65G [00:15<00:11, 63.7MB/s]
57%|█████████████████████▋ | 945M/1.65G [00:15<00:11, 63.0MB/s]
58%|█████████████████████▉ | 952M/1.65G [00:15<00:11, 63.5MB/s]
58%|██████████████████████ | 958M/1.65G [00:15<00:10, 63.7MB/s]
58%|██████████████████████▏ | 965M/1.65G [00:15<00:10, 63.9MB/s]
59%|██████████████████████▎ | 971M/1.65G [00:15<00:10, 63.1MB/s]
59%|██████████████████████▍ | 977M/1.65G [00:15<00:10, 63.3MB/s]
60%|██████████████████████▌ | 984M/1.65G [00:15<00:10, 63.5MB/s]
60%|██████████████████████▊ | 990M/1.65G [00:16<00:10, 63.6MB/s]
60%|██████████████████████▉ | 996M/1.65G [00:16<00:10, 63.1MB/s]
61%|██████████████████████▍ | 1.00G/1.65G [00:16<00:10, 63.6MB/s]
61%|██████████████████████▌ | 1.01G/1.65G [00:16<00:10, 63.6MB/s]
61%|██████████████████████▋ | 1.02G/1.65G [00:16<00:10, 63.4MB/s]
62%|██████████████████████▉ | 1.02G/1.65G [00:16<00:09, 63.2MB/s]
62%|███████████████████████ | 1.03G/1.65G [00:16<00:09, 63.6MB/s]
63%|███████████████████████▏ | 1.03G/1.65G [00:16<00:09, 63.7MB/s]
63%|███████████████████████▎ | 1.04G/1.65G [00:16<00:09, 63.3MB/s]
63%|███████████████████████▍ | 1.05G/1.65G [00:16<00:09, 63.1MB/s]
64%|███████████████████████▌ | 1.05G/1.65G [00:17<00:09, 63.4MB/s]
64%|███████████████████████▋ | 1.06G/1.65G [00:17<00:09, 63.8MB/s]
65%|███████████████████████▉ | 1.07G/1.65G [00:17<00:09, 63.2MB/s]
65%|████████████████████████ | 1.07G/1.65G [00:17<00:09, 63.5MB/s]
65%|████████████████████████▏ | 1.08G/1.65G [00:17<00:09, 63.6MB/s]
66%|████████████████████████▎ | 1.09G/1.65G [00:17<00:08, 63.6MB/s]
66%|████████████████████████▍ | 1.09G/1.65G [00:17<00:08, 63.6MB/s]
66%|████████████████████████▌ | 1.10G/1.65G [00:17<00:08, 62.4MB/s]
67%|████████████████████████▋ | 1.10G/1.65G [00:17<00:08, 62.4MB/s]
67%|████████████████████████▉ | 1.11G/1.65G [00:17<00:08, 62.4MB/s]
68%|█████████████████████████ | 1.12G/1.65G [00:18<00:08, 62.7MB/s]
68%|█████████████████████████▏ | 1.12G/1.65G [00:18<00:08, 62.8MB/s]
68%|█████████████████████████▎ | 1.13G/1.65G [00:18<00:08, 62.2MB/s]
69%|█████████████████████████▍ | 1.14G/1.65G [00:18<00:08, 62.5MB/s]
69%|█████████████████████████▌ | 1.14G/1.65G [00:18<00:08, 62.7MB/s]
70%|█████████████████████████▋ | 1.15G/1.65G [00:18<00:07, 63.0MB/s]
70%|█████████████████████████▊ | 1.16G/1.65G [00:18<00:07, 62.3MB/s]
70%|██████████████████████████ | 1.16G/1.65G [00:18<00:07, 62.3MB/s]
71%|██████████████████████████▏ | 1.17G/1.65G [00:18<00:07, 62.5MB/s]
71%|██████████████████████████▎ | 1.17G/1.65G [00:19<00:07, 62.6MB/s]
71%|██████████████████████████▍ | 1.18G/1.65G [00:19<00:07, 61.7MB/s]
72%|██████████████████████████▌ | 1.19G/1.65G [00:19<00:07, 58.7MB/s]
72%|██████████████████████████▋ | 1.19G/1.65G [00:19<00:08, 52.8MB/s]
73%|██████████████████████████▊ | 1.20G/1.65G [00:19<00:08, 54.7MB/s]
73%|██████████████████████████▉ | 1.20G/1.65G [00:19<00:07, 56.3MB/s]
73%|███████████████████████████ | 1.21G/1.65G [00:19<00:07, 57.2MB/s]
74%|███████████████████████████▏ | 1.22G/1.65G [00:19<00:07, 57.6MB/s]
74%|███████████████████████████▎ | 1.22G/1.65G [00:19<00:07, 58.3MB/s]
74%|███████████████████████████▍ | 1.23G/1.65G [00:19<00:07, 58.8MB/s]
75%|███████████████████████████▋ | 1.23G/1.65G [00:20<00:07, 59.3MB/s]
75%|███████████████████████████▊ | 1.24G/1.65G [00:20<00:06, 59.3MB/s]
75%|███████████████████████████▉ | 1.25G/1.65G [00:20<00:06, 59.0MB/s]
76%|████████████████████████████ | 1.25G/1.65G [00:20<00:06, 59.5MB/s]
76%|████████████████████████████▏ | 1.26G/1.65G [00:20<00:06, 59.5MB/s]
77%|████████████████████████████▎ | 1.26G/1.65G [00:20<00:06, 59.9MB/s]
77%|████████████████████████████▍ | 1.27G/1.65G [00:20<00:06, 59.3MB/s]
77%|████████████████████████████▌ | 1.28G/1.65G [00:20<00:06, 59.4MB/s]
78%|████████████████████████████▋ | 1.28G/1.65G [00:20<00:06, 59.7MB/s]
78%|████████████████████████████▊ | 1.29G/1.65G [00:20<00:06, 59.6MB/s]
78%|████████████████████████████▉ | 1.29G/1.65G [00:21<00:05, 59.7MB/s]
79%|█████████████████████████████ | 1.30G/1.65G [00:21<00:05, 59.2MB/s]
79%|█████████████████████████████▏ | 1.31G/1.65G [00:21<00:05, 59.5MB/s]
79%|█████████████████████████████▍ | 1.31G/1.65G [00:21<00:05, 59.7MB/s]
80%|█████████████████████████████▌ | 1.32G/1.65G [00:21<00:05, 59.9MB/s]
80%|█████████████████████████████▋ | 1.32G/1.65G [00:21<00:05, 60.2MB/s]
81%|█████████████████████████████▊ | 1.33G/1.65G [00:21<00:05, 59.2MB/s]
81%|█████████████████████████████▉ | 1.34G/1.65G [00:21<00:05, 59.6MB/s]
81%|██████████████████████████████ | 1.34G/1.65G [00:21<00:05, 59.8MB/s]
82%|██████████████████████████████▏ | 1.35G/1.65G [00:21<00:05, 59.9MB/s]
82%|██████████████████████████████▎ | 1.35G/1.65G [00:22<00:04, 60.2MB/s]
82%|██████████████████████████████▍ | 1.36G/1.65G [00:22<00:04, 59.5MB/s]
83%|██████████████████████████████▌ | 1.37G/1.65G [00:22<00:04, 59.8MB/s]
83%|██████████████████████████████▋ | 1.37G/1.65G [00:22<00:04, 59.9MB/s]
83%|██████████████████████████████▊ | 1.38G/1.65G [00:22<00:04, 60.1MB/s]
84%|███████████████████████████████ | 1.38G/1.65G [00:22<00:04, 60.4MB/s]
84%|███████████████████████████████▏ | 1.39G/1.65G [00:22<00:04, 59.3MB/s]
85%|███████████████████████████████▎ | 1.40G/1.65G [00:22<00:04, 59.3MB/s]
85%|███████████████████████████████▍ | 1.40G/1.65G [00:22<00:04, 59.8MB/s]
85%|███████████████████████████████▌ | 1.41G/1.65G [00:22<00:04, 59.7MB/s]
86%|███████████████████████████████▋ | 1.42G/1.65G [00:23<00:03, 60.1MB/s]
86%|███████████████████████████████▊ | 1.42G/1.65G [00:23<00:03, 59.4MB/s]
86%|███████████████████████████████▉ | 1.43G/1.65G [00:23<00:03, 59.8MB/s]
87%|████████████████████████████████ | 1.43G/1.65G [00:23<00:03, 59.9MB/s]
87%|████████████████████████████████▏ | 1.44G/1.65G [00:23<00:03, 60.2MB/s]
87%|████████████████████████████████▎ | 1.45G/1.65G [00:23<00:03, 60.5MB/s]
88%|████████████████████████████████▍ | 1.45G/1.65G [00:23<00:03, 60.6MB/s]
88%|████████████████████████████████▋ | 1.46G/1.65G [00:23<00:03, 61.4MB/s]
89%|████████████████████████████████▊ | 1.46G/1.65G [00:23<00:03, 62.0MB/s]
89%|████████████████████████████████▉ | 1.47G/1.65G [00:23<00:02, 62.5MB/s]
89%|█████████████████████████████████ | 1.48G/1.65G [00:24<00:02, 61.8MB/s]
90%|█████████████████████████████████▏ | 1.48G/1.65G [00:24<00:02, 62.3MB/s]
90%|█████████████████████████████████▎ | 1.49G/1.65G [00:24<00:02, 62.6MB/s]
91%|█████████████████████████████████▍ | 1.50G/1.65G [00:24<00:02, 62.8MB/s]
91%|█████████████████████████████████▋ | 1.50G/1.65G [00:24<00:02, 63.0MB/s]
91%|█████████████████████████████████▊ | 1.51G/1.65G [00:24<00:02, 62.3MB/s]
92%|█████████████████████████████████▉ | 1.51G/1.65G [00:24<00:02, 62.6MB/s]
92%|██████████████████████████████████ | 1.52G/1.65G [00:24<00:02, 62.8MB/s]
92%|██████████████████████████████████▏ | 1.53G/1.65G [00:24<00:01, 63.1MB/s]
93%|██████████████████████████████████▎ | 1.53G/1.65G [00:25<00:01, 62.4MB/s]
93%|██████████████████████████████████▍ | 1.54G/1.65G [00:25<00:01, 62.4MB/s]
94%|██████████████████████████████████▌ | 1.55G/1.65G [00:25<00:01, 62.8MB/s]
94%|██████████████████████████████████▊ | 1.55G/1.65G [00:25<00:01, 63.3MB/s]
94%|██████████████████████████████████▉ | 1.56G/1.65G [00:25<00:01, 63.5MB/s]
95%|███████████████████████████████████ | 1.57G/1.65G [00:25<00:01, 63.6MB/s]
95%|███████████████████████████████████▏ | 1.57G/1.65G [00:25<00:01, 62.7MB/s]
96%|███████████████████████████████████▎ | 1.58G/1.65G [00:25<00:01, 63.0MB/s]
96%|███████████████████████████████████▍ | 1.58G/1.65G [00:25<00:01, 63.1MB/s]
96%|███████████████████████████████████▌ | 1.59G/1.65G [00:25<00:00, 63.3MB/s]
97%|███████████████████████████████████▊ | 1.60G/1.65G [00:26<00:00, 63.5MB/s]
97%|███████████████████████████████████▉ | 1.60G/1.65G [00:26<00:00, 62.6MB/s]
97%|████████████████████████████████████ | 1.61G/1.65G [00:26<00:00, 63.2MB/s]
98%|████████████████████████████████████▏| 1.62G/1.65G [00:26<00:00, 63.8MB/s]
98%|████████████████████████████████████▎| 1.62G/1.65G [00:26<00:00, 63.6MB/s]
99%|████████████████████████████████████▍| 1.63G/1.65G [00:26<00:00, 63.8MB/s]
99%|████████████████████████████████████▋| 1.64G/1.65G [00:26<00:00, 62.7MB/s]
99%|████████████████████████████████████▊| 1.64G/1.65G [00:26<00:00, 63.1MB/s]
100%|████████████████████████████████████▉| 1.65G/1.65G [00:26<00:00, 63.1MB/s]
0%| | 0.00/1.65G [00:00<?, ?B/s]
100%|█████████████████████████████████████| 1.65G/1.65G [00:00<00:00, 6.13TB/s]
Download complete in 01m03s (1576.2 MB)
Filtering raw data in 1 contiguous segment
Setting up high-pass filter at 2 Hz
IIR filter parameters
---------------------
Butterworth highpass zero-phase (two-pass forward and reverse) non-causal filter:
- Filter order 8 (effective, after forward-backward)
- Cutoff at 2.00 Hz: -6.02 dB
Extract Xdawn covariance matrices¶
nfilter = 4
xdwn = XdawnCovariances(estimator="scm", nfilter=nfilter)
split = train_test_split(X, y, train_size=0.25, random_state=42)
Xtrain, Xtest, ytrain, ytest = split
covs = xdwn.fit(Xtrain, ytrain).transform(Xtest)
Spectral Embedding (SE)¶
plot_embedding(covs, ytest, metric="riemann", embd_type="Spectral",
normalize=True)
plt.show()
Locally Linear Embedding (LLE)¶
plot_embedding(covs, ytest, metric="riemann", embd_type="LocallyLinear",
normalize=False)
plt.show()
TNSE¶
plot_embedding(covs, ytest, metric="riemann", embd_type="TSNE",
normalize=False, max_iter=50)
plt.show()
/home/docs/checkouts/readthedocs.org/user_builds/pyriemann/checkouts/latest/pyriemann/optimization/positive_definite.py:284: UserWarning: Convergence not reached. Try increasing max_iter.
warnings.warn("Convergence not reached. Try increasing max_iter.")
References¶
Total running time of the script: (1 minutes 14.627 seconds)