Within Session SSVEP#

This Example shows how to perform a within-session SSVEP analysis on the MAMEM dataset 3, using a CCA pipeline.

The within-session evaluation assesses the performance of a classification pipeline using a 5-fold cross-validation. The reported metric (here, accuracy) is the average of all fold.

# Authors: Sylvain Chevallier <sylvain.chevallier@uvsq.fr>
#
# License: BSD (3-clause)

import warnings

import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.pipeline import make_pipeline

import moabb
from moabb.datasets import Kalunga2016
from moabb.evaluations import WithinSessionEvaluation
from moabb.paradigms import SSVEP
from moabb.pipelines import SSVEP_CCA


warnings.simplefilter(action="ignore", category=FutureWarning)
warnings.simplefilter(action="ignore", category=RuntimeWarning)
moabb.set_log_level("info")

Loading Dataset#

Load 2 subjects of Kalunga2016 dataset

subj = [1, 3]
dataset = Kalunga2016()
dataset.subject_list = subj

Choose Paradigm#

We select the paradigm SSVEP, applying a bandpass filter (3-15 Hz) on the data and we keep only the first 3 classes, that is stimulation frequency of 13Hz, 17Hz and 21Hz.

paradigm = SSVEP(fmin=10, fmax=40, n_classes=3)

Create Pipelines#

Use a Canonical Correlation Analysis classifier

interval = dataset.interval
freqs = paradigm.used_events(dataset)

pipeline = {}
pipeline["CCA"] = make_pipeline(SSVEP_CCA(interval=interval, freqs=freqs, n_harmonics=3))

Get Data (optional)#

To get access to the EEG signals downloaded from the dataset, you could use dataset.get_data(subjects=[subject_id]) to obtain the EEG under MNE format, stored in a dictionary of sessions and runs. Otherwise, paradigm.get_data(dataset=dataset, subjects=[subject_id]) allows to obtain the EEG data in scikit format, the labels and the meta information. In paradigm.get_data, the EEG are preprocessed according to the paradigm requirement.

# sessions = dataset.get_data(subjects=[3])
# X, labels, meta = paradigm.get_data(dataset=dataset, subjects=[3])

Evaluation#

The evaluation will return a DataFrame containing a single AUC score for each subject and pipeline.

overwrite = True  # set to True if we want to overwrite cached results

evaluation = WithinSessionEvaluation(
    paradigm=paradigm, datasets=dataset, suffix="examples", overwrite=overwrite
)
results = evaluation.process(pipeline)

print(results.head())

Kalunga2016-WithinSession:   0%|          | 0/2 [00:00<?, ?it/s]
Kalunga2016-WithinSession:  50%|█████     | 1/2 [00:00<00:00,  2.48it/s]

  0%|                                              | 0.00/2.27M [00:00<?, ?B/s]

  1%|▏                                    | 12.3k/2.27M [00:00<00:30, 74.5kB/s]

  2%|▋                                     | 41.0k/2.27M [00:00<00:15, 144kB/s]

  4%|█▌                                    | 95.2k/2.27M [00:00<00:08, 248kB/s]

  7%|██▋                                    | 157k/2.27M [00:00<00:06, 319kB/s]

 11%|████▏                                  | 245k/2.27M [00:00<00:04, 426kB/s]

 14%|█████▌                                 | 327k/2.27M [00:00<00:04, 474kB/s]

 20%|███████▊                               | 454k/2.27M [00:01<00:02, 609kB/s]

 24%|█████████▏                             | 536k/2.27M [00:01<00:02, 599kB/s]

 27%|██████████▌                            | 616k/2.27M [00:01<00:02, 590kB/s]

 31%|███████████▉                           | 698k/2.27M [00:01<00:02, 586kB/s]

 35%|█████████████▋                         | 797k/2.27M [00:01<00:02, 618kB/s]

 39%|███████████████                        | 879k/2.27M [00:01<00:02, 605kB/s]

 43%|████████████████▊                      | 977k/2.27M [00:01<00:02, 631kB/s]

 47%|█████████████████▋                    | 1.06M/2.27M [00:02<00:01, 614kB/s]

 53%|████████████████████▏                 | 1.21M/2.27M [00:02<00:01, 739kB/s]

 57%|█████████████████████▌                | 1.29M/2.27M [00:02<00:01, 689kB/s]

 60%|██████████████████████▉               | 1.37M/2.27M [00:02<00:01, 655kB/s]

 67%|█████████████████████████▎            | 1.52M/2.27M [00:02<00:00, 767kB/s]

 71%|██████████████████████████▉           | 1.61M/2.27M [00:02<00:00, 742kB/s]

 75%|████████████████████████████▎         | 1.70M/2.27M [00:02<00:00, 692kB/s]

 78%|█████████████████████████████▋        | 1.78M/2.27M [00:03<00:00, 657kB/s]

 82%|███████████████████████████████       | 1.86M/2.27M [00:03<00:00, 633kB/s]

 88%|█████████████████████████████████▌    | 2.01M/2.27M [00:03<00:00, 754kB/s]

 94%|███████████████████████████████████▋  | 2.14M/2.27M [00:03<00:00, 802kB/s]

 98%|█████████████████████████████████████▍| 2.24M/2.27M [00:03<00:00, 768kB/s]

  0%|                                              | 0.00/2.27M [00:00<?, ?B/s]
100%|█████████████████████████████████████| 2.27M/2.27M [00:00<00:00, 8.54GB/s]

  0%|                                              | 0.00/2.13M [00:00<?, ?B/s]

  1%|▏                                    | 12.3k/2.13M [00:00<00:30, 68.9kB/s]

  2%|▋                                     | 39.9k/2.13M [00:00<00:16, 130kB/s]

  4%|█▋                                    | 91.1k/2.13M [00:00<00:09, 218kB/s]

  7%|██▊                                    | 151k/2.13M [00:00<00:07, 282kB/s]

 11%|████▏                                  | 228k/2.13M [00:00<00:05, 360kB/s]

 16%|██████                                 | 334k/2.13M [00:00<00:03, 469kB/s]

 19%|███████▌                               | 415k/2.13M [00:01<00:03, 486kB/s]

 24%|█████████▍                             | 513k/2.13M [00:01<00:03, 532kB/s]

 30%|███████████▌                           | 628k/2.13M [00:01<00:02, 596kB/s]

 33%|█████████████                          | 710k/2.13M [00:01<00:02, 576kB/s]

 36%|██████████████▏                        | 775k/2.13M [00:01<00:02, 530kB/s]

 41%|███████████████▉                       | 872k/2.13M [00:01<00:02, 559kB/s]

 46%|██████████████████                     | 987k/2.13M [00:02<00:01, 613kB/s]

 51%|███████████████████▎                  | 1.08M/2.13M [00:02<00:01, 610kB/s]

 55%|████████████████████▊                 | 1.16M/2.13M [00:02<00:01, 586kB/s]

 58%|██████████████████████▏               | 1.24M/2.13M [00:02<00:01, 561kB/s]

 63%|████████████████████████              | 1.35M/2.13M [00:02<00:01, 595kB/s]

 69%|██████████████████████████▎           | 1.48M/2.13M [00:02<00:00, 672kB/s]

 76%|████████████████████████████▊         | 1.61M/2.13M [00:02<00:00, 734kB/s]

 82%|██████████████████████████████▉       | 1.74M/2.13M [00:03<00:00, 750kB/s]

 87%|█████████████████████████████████▏    | 1.86M/2.13M [00:03<00:00, 757kB/s]

 92%|██████████████████████████████████▊   | 1.95M/2.13M [00:03<00:00, 708kB/s]

 97%|████████████████████████████████████▉ | 2.07M/2.13M [00:03<00:00, 728kB/s]

  0%|                                              | 0.00/2.13M [00:00<?, ?B/s]
100%|█████████████████████████████████████| 2.13M/2.13M [00:00<00:00, 10.5GB/s]

Kalunga2016-WithinSession: 100%|██████████| 2/2 [00:09<00:00,  5.43s/it]
Kalunga2016-WithinSession: 100%|██████████| 2/2 [00:09<00:00,  4.68s/it]
      score      time  samples  ... n_sessions      dataset  pipeline
0  0.773333  0.038621     48.0  ...          1  Kalunga2016       CCA
1  0.915556  0.037432     48.0  ...          1  Kalunga2016       CCA

[2 rows x 9 columns]

Plot Results#

Here we plot the results, indicating the score for each subject

plt.figure()
sns.barplot(data=results, y="score", x="session", hue="subject", palette="viridis")

<Axes: xlabel='session', ylabel='score'>

And the computation time in seconds

plt.figure()
ax = sns.barplot(data=results, y="time", x="session", hue="subject", palette="Reds")
ax.set_ylabel("Time (s)")
plt.show()

Total running time of the script: (0 minutes 10.993 seconds)

Estimated memory usage: 294 MB

Gallery generated by Sphinx-Gallery