Note
Go to the end to download the full example code.
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.
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]No hdf5_path provided, models will not be saved.
Kalunga2016-WithinSession: 50%|█████ | 1/2 [00:00<00:00, 1.84it/s]
0%| | 0.00/2.27M [00:00<?, ?B/s]
1%|▏ | 12.3k/2.27M [00:00<00:44, 50.6kB/s]
2%|▋ | 41.0k/2.27M [00:00<00:17, 126kB/s]
4%|█▋ | 101k/2.27M [00:00<00:10, 210kB/s]
8%|███ | 177k/2.27M [00:00<00:06, 301kB/s]
14%|█████▎ | 311k/2.27M [00:00<00:04, 476kB/s]
16%|██████▎ | 367k/2.27M [00:01<00:04, 430kB/s]
20%|███████▊ | 458k/2.27M [00:01<00:03, 469kB/s]
24%|█████████▎ | 540k/2.27M [00:01<00:03, 477kB/s]
27%|██████████▋ | 621k/2.27M [00:01<00:03, 481kB/s]
32%|████████████▎ | 719k/2.27M [00:01<00:03, 515kB/s]
35%|█████████████▋ | 801k/2.27M [00:01<00:02, 509kB/s]
39%|███████████████▏ | 883k/2.27M [00:02<00:02, 505kB/s]
42%|████████████████▌ | 965k/2.27M [00:02<00:02, 502kB/s]
48%|██████████████████▎ | 1.09M/2.27M [00:02<00:02, 587kB/s]
55%|████████████████████▊ | 1.24M/2.27M [00:02<00:01, 678kB/s]
58%|█████████████████████▉ | 1.31M/2.27M [00:02<00:01, 600kB/s]
62%|███████████████████████▌ | 1.41M/2.27M [00:02<00:01, 593kB/s]
65%|████████████████████████▌ | 1.47M/2.27M [00:03<00:01, 531kB/s]
71%|███████████████████████████ | 1.62M/2.27M [00:03<00:01, 636kB/s]
76%|████████████████████████████▉ | 1.73M/2.27M [00:03<00:00, 653kB/s]
82%|███████████████████████████████▏ | 1.86M/2.27M [00:03<00:00, 695kB/s]
89%|█████████████████████████████████▋ | 2.01M/2.27M [00:03<00:00, 754kB/s]
94%|███████████████████████████████████▊ | 2.14M/2.27M [00:03<00:00, 763kB/s]
98%|█████████████████████████████████████▏| 2.22M/2.27M [00:04<00:00, 683kB/s]
0%| | 0.00/2.27M [00:00<?, ?B/s]
100%|█████████████████████████████████████| 2.27M/2.27M [00:00<00:00, 10.8GB/s]
0%| | 0.00/2.13M [00:00<?, ?B/s]
1%|▏ | 14.3k/2.13M [00:00<00:30, 70.4kB/s]
2%|▊ | 45.1k/2.13M [00:00<00:15, 132kB/s]
4%|█▋ | 93.2k/2.13M [00:00<00:10, 201kB/s]
10%|███▊ | 208k/2.13M [00:00<00:04, 390kB/s]
14%|█████▎ | 289k/2.13M [00:00<00:04, 426kB/s]
18%|██████▊ | 375k/2.13M [00:01<00:03, 458kB/s]
21%|████████▏ | 447k/2.13M [00:01<00:03, 453kB/s]
24%|█████████▍ | 513k/2.13M [00:01<00:03, 436kB/s]
28%|██████████▉ | 595k/2.13M [00:01<00:03, 455kB/s]
35%|█████████████▌ | 741k/2.13M [00:01<00:02, 588kB/s]
39%|███████████████▍ | 840k/2.13M [00:01<00:02, 589kB/s]
46%|██████████████████ | 987k/2.13M [00:02<00:01, 682kB/s]
50%|███████████████████ | 1.07M/2.13M [00:02<00:01, 626kB/s]
55%|████████████████████▊ | 1.17M/2.13M [00:02<00:01, 617kB/s]
59%|██████████████████████▎ | 1.25M/2.13M [00:02<00:01, 579kB/s]
62%|███████████████████████▍ | 1.31M/2.13M [00:02<00:01, 522kB/s]
66%|█████████████████████████▏ | 1.41M/2.13M [00:02<00:01, 545kB/s]
70%|██████████████████████████▋ | 1.49M/2.13M [00:03<00:01, 530kB/s]
77%|█████████████████████████████▎ | 1.64M/2.13M [00:03<00:00, 639kB/s]
81%|██████████████████████████████▊ | 1.72M/2.13M [00:03<00:00, 596kB/s]
87%|█████████████████████████████████ | 1.85M/2.13M [00:03<00:00, 654kB/s]
92%|██████████████████████████████████▊ | 1.95M/2.13M [00:03<00:00, 636kB/s]
98%|█████████████████████████████████████▏| 2.08M/2.13M [00:03<00:00, 684kB/s]
0%| | 0.00/2.13M [00:00<?, ?B/s]
100%|█████████████████████████████████████| 2.13M/2.13M [00:00<00:00, 10.8GB/s]
No hdf5_path provided, models will not be saved.
Kalunga2016-WithinSession: 100%|██████████| 2/2 [00:10<00:00, 6.10s/it]
Kalunga2016-WithinSession: 100%|██████████| 2/2 [00:10<00:00, 5.27s/it]
score time samples ... n_sessions dataset pipeline
0 0.251111 0.040635 48.0 ... 1 Kalunga2016 CCA
1 0.331111 0.040015 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 11.947 seconds)
Estimated memory usage: 246 MB