Tutorial 0: Getting Started#

This tutorial takes you through a basic working example of how to use this codebase, including all the different components, up to the results generation. If you’d like to know about the statistics and plotting, see the next tutorial.

# Authors: Vinay Jayaram <vinayjayaram13@gmail.com>
#
# License: BSD (3-clause)

Introduction#

To use the codebase you need an evaluation and a paradigm, some algorithms, and a list of datasets to run it all on. You can find those in the following submodules; detailed tutorials are given for each of them.

import numpy as np
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis as LDA
from sklearn.model_selection import GridSearchCV
from sklearn.pipeline import make_pipeline
from sklearn.svm import SVC

If you would like to specify the logging level when it is running, you can use the standard python logging commands through the top-level moabb module

import moabb
from moabb.datasets import BNCI2014_001, utils
from moabb.evaluations import CrossSessionEvaluation
from moabb.paradigms import LeftRightImagery
from moabb.pipelines.features import LogVariance

In order to create pipelines within a script, you will likely need at least the make_pipeline function. They can also be specified via a .yml file. Here we will make a couple pipelines just for convenience

moabb.set_log_level("info")

Create pipelines#

We create two pipelines: channel-wise log variance followed by LDA, and channel-wise log variance followed by a cross-validated SVM (note that a cross-validation via scikit-learn cannot be described in a .yml file). For later in the process, the pipelines need to be in a dictionary where the key is the name of the pipeline and the value is the Pipeline object

pipelines = {}
pipelines["AM+LDA"] = make_pipeline(LogVariance(), LDA())
parameters = {"C": np.logspace(-2, 2, 10)}
clf = GridSearchCV(SVC(kernel="linear"), parameters)
pipe = make_pipeline(LogVariance(), clf)

pipelines["AM+SVM"] = pipe

Datasets#

Datasets can be specified in many ways: Each paradigm has a property ‘datasets’ which returns the datasets that are appropriate for that paradigm

print(LeftRightImagery().datasets)

/home/runner/work/moabb/moabb/moabb/datasets/fake.py:92: RuntimeWarning: Setting non-standard config type: "MNE_DATASETS_FAKEDATASET-IMAGERY-10-2--60-60--120-120--FAKE1-FAKE2-FAKE3--C3-CZ-C4_PATH"
  set_config(key, temp_dir)
/home/runner/work/moabb/moabb/moabb/datasets/fake.py:92: RuntimeWarning: Setting non-standard config type: "MNE_DATASETS_FAKEVIRTUALREALITYDATASET-P300-21-1--60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60-60--120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120-120--TARGET-NONTARGET--C3-CZ-C4_PATH"
  set_config(key, temp_dir)
/home/runner/work/moabb/moabb/moabb/datasets/romani_bf2025_erp.py:218: RuntimeWarning: Setting non-standard config type: "RomaniBF2025ERP"
  path_root = fetch_dataset(
[<moabb.datasets.bnci.BNCI2014_001 object at 0x7fdfc4d5e920>, <moabb.datasets.bnci.BNCI2014_004 object at 0x7fdfc4d5fe20>, <moabb.datasets.beetl.Beetl2021_A object at 0x7fdfc4d5ea40>, <moabb.datasets.beetl.Beetl2021_B object at 0x7fdfcead24a0>, <moabb.datasets.gigadb.Cho2017 object at 0x7fdfc5824df0>, <moabb.datasets.dreyer2023.Dreyer2023 object at 0x7fdfc5827940>, <moabb.datasets.dreyer2023.Dreyer2023A object at 0x7fdfc572a0e0>, <moabb.datasets.dreyer2023.Dreyer2023B object at 0x7fdfc4d5dae0>, <moabb.datasets.dreyer2023.Dreyer2023C object at 0x7fdfc4d5cc70>, <moabb.datasets.mpi_mi.GrosseWentrup2009 object at 0x7fdfc494c8b0>, <moabb.datasets.Lee2019.Lee2019_MI object at 0x7fdfc5648af0>, <moabb.datasets.liu2024.Liu2024 object at 0x7fdfc5649570>, <moabb.datasets.physionet_mi.PhysionetMI object at 0x7fdfc49044c0>, <moabb.datasets.schirrmeister2017.Schirrmeister2017 object at 0x7fdfc564a170>, <moabb.datasets.bbci_eeg_fnirs.Shin2017A object at 0x7fdfc49072b0>, <moabb.datasets.stieger2021.Stieger2021 object at 0x7fdfc4907580>, <moabb.datasets.Weibo2014.Weibo2014 object at 0x7fdfc4906aa0>, <moabb.datasets.Zhou2016.Zhou2016 object at 0x7fdfc4907610>]

Or you can run a search through the available datasets:

print(utils.dataset_search(paradigm="imagery", min_subjects=6))

[<moabb.datasets.alex_mi.AlexMI object at 0x7fdfc5827940>, <moabb.datasets.bnci.BNCI2014_001 object at 0x7fdfc4906aa0>, <moabb.datasets.bnci.BNCI2014_002 object at 0x7fdfc4907610>, <moabb.datasets.bnci.BNCI2014_004 object at 0x7fdfc4906890>, <moabb.datasets.bnci.BNCI2015_001 object at 0x7fdfc4906950>, <moabb.datasets.bnci.BNCI2015_004 object at 0x7fdfc4907340>, <moabb.datasets.gigadb.Cho2017 object at 0x7fdfc49047f0>, <moabb.datasets.dreyer2023.Dreyer2023 object at 0x7fdfc4906830>, <moabb.datasets.dreyer2023.Dreyer2023A object at 0x7fdfc49048e0>, <moabb.datasets.dreyer2023.Dreyer2023B object at 0x7fdfc49040a0>, <moabb.datasets.dreyer2023.Dreyer2023C object at 0x7fdfc4907b20>, <moabb.datasets.fake.FakeDataset object at 0x7fdfc4904790>, <moabb.datasets.mpi_mi.GrosseWentrup2009 object at 0x7fdfc4907580>, <moabb.datasets.Lee2019.Lee2019_MI object at 0x7fdfc49044c0>, <moabb.datasets.liu2024.Liu2024 object at 0x7fdfc4906800>, <moabb.datasets.upper_limb.Ofner2017 object at 0x7fdfc4907730>, <moabb.datasets.physionet_mi.PhysionetMI object at 0x7fdfc49054b0>, <moabb.datasets.schirrmeister2017.Schirrmeister2017 object at 0x7fdfc55e4550>, <moabb.datasets.bbci_eeg_fnirs.Shin2017A object at 0x7fdfc55e5690>, <moabb.datasets.bbci_eeg_fnirs.Shin2017B object at 0x7fdfc55e66b0>, <moabb.datasets.stieger2021.Stieger2021 object at 0x7fdfc55e6230>, <moabb.datasets.Weibo2014.Weibo2014 object at 0x7fdfc55e6200>]

Or you can simply make your own list (which we do here due to computational constraints)

dataset = BNCI2014_001()
dataset.subject_list = dataset.subject_list[:2]
datasets = [dataset]

Paradigm#

Paradigms define the events, epoch time, bandpass, and other preprocessing parameters. They have defaults that you can read in the documentation, or you can simply set them as we do here. A single paradigm defines a method for going from continuous data to trial data of a fixed size. To learn more look at the tutorial Exploring Paradigms

fmin = 8
fmax = 35
paradigm = LeftRightImagery(fmin=fmin, fmax=fmax)

Evaluation#

An evaluation defines how the training and test sets are chosen. This could be cross-validated within a single recording, or across days, or sessions, or subjects. This also is the correct place to specify multiple threads.

evaluation = CrossSessionEvaluation(
    paradigm=paradigm, datasets=datasets, suffix="examples", overwrite=False
)
results = evaluation.process(pipelines)

/home/runner/work/moabb/moabb/moabb/analysis/results.py:95: RuntimeWarning: Setting non-standard config type: "MOABB_RESULTS"
  set_config("MOABB_RESULTS", osp.join(osp.expanduser("~"), "mne_data"))

BNCI2014-001-CrossSession:   0%|          | 0/2 [00:00<?, ?it/s]2026-01-08 15:31:27,115 INFO MainThread moabb.datasets.download MNE_DATA is not already configured. It will be set to default location in the home directory - /home/runner/mne_data
All datasets will be downloaded to this location, if anything is already downloaded, please move manually to this location
/home/runner/work/moabb/moabb/moabb/datasets/download.py:60: RuntimeWarning: Setting non-standard config type: "MNE_DATASETS_BNCI_PATH"
  set_config(key, get_config("MNE_DATA"))

BNCI2014-001-CrossSession:  50%|█████     | 1/2 [00:03<00:03,  3.60s/it]
BNCI2014-001-CrossSession: 100%|██████████| 2/2 [00:06<00:00,  3.21s/it]
BNCI2014-001-CrossSession: 100%|██████████| 2/2 [00:06<00:00,  3.27s/it]
2026-01-08 15:31:33,379 INFO MainThread moabb.evaluations.base AM+LDA | BNCI2014-001 | 1 | 0train: Score 0.786
2026-01-08 15:31:33,503 INFO MainThread moabb.evaluations.base AM+LDA | BNCI2014-001 | 1 | 1test: Score 0.802
2026-01-08 15:31:33,625 INFO MainThread moabb.evaluations.base AM+SVM | BNCI2014-001 | 1 | 0train: Score 0.797
2026-01-08 15:31:33,769 INFO MainThread moabb.evaluations.base AM+SVM | BNCI2014-001 | 1 | 1test: Score 0.774
2026-01-08 15:31:33,907 INFO MainThread moabb.evaluations.base AM+LDA | BNCI2014-001 | 2 | 0train: Score 0.577
2026-01-08 15:31:34,029 INFO MainThread moabb.evaluations.base AM+LDA | BNCI2014-001 | 2 | 1test: Score 0.499
2026-01-08 15:31:34,155 INFO MainThread moabb.evaluations.base AM+SVM | BNCI2014-001 | 2 | 0train: Score 0.551
2026-01-08 15:31:34,294 INFO MainThread moabb.evaluations.base AM+SVM | BNCI2014-001 | 2 | 1test: Score 0.471

Results are returned as a pandas DataFrame, and from here you can do as you want with them

print(results.head())

      score      time  samples  ... n_sessions       dataset  pipeline
0.786458  0.034883    144.0  ...          2  BNCI2014-001    AM+LDA
0.801698  0.033548    144.0  ...          2  BNCI2014-001    AM+LDA
0.576582  0.025918    144.0  ...          2  BNCI2014-001    AM+LDA
0.498650  0.023120    144.0  ...          2  BNCI2014-001    AM+LDA
0.797068  0.160346    144.0  ...          2  BNCI2014-001    AM+SVM

[5 rows x 9 columns]

Total running time of the script: (1 minutes 25.290 seconds)

Estimated memory usage: 797 MB

Gallery generated by Sphinx-Gallery