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.metrics import accuracy_score, roc_auc_score
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)

[<moabb.datasets.bnci.bnci_2014.BNCI2014_001 object at 0x7fafb33bc940>, <moabb.datasets.bnci.bnci_2014.BNCI2014_004 object at 0x7fafa616fee0>, <moabb.datasets.beetl.Beetl2021_A object at 0x7fafa69e1330>, <moabb.datasets.beetl.Beetl2021_B object at 0x7fafa616ff40>, <moabb.datasets.brandl2020.Brandl2020 object at 0x7fafa616feb0>, <moabb.datasets.chang2025.Chang2025 object at 0x7fafa61a9630>, <moabb.datasets.gigadb.Cho2017 object at 0x7fafa61a9570>, <moabb.datasets.dreyer2023.Dreyer2023 object at 0x7fafa61a96f0>, <moabb.datasets.dreyer2023.Dreyer2023A object at 0x7fafa61a9720>, <moabb.datasets.dreyer2023.Dreyer2023B object at 0x7fafa61a8c40>, <moabb.datasets.dreyer2023.Dreyer2023C object at 0x7fafa61a8cd0>, <moabb.datasets.forenzo2023.Forenzo2023 object at 0x7fafa61a8be0>, <moabb.datasets.mpi_mi.GrosseWentrup2009 object at 0x7fafa61a8c70>, <moabb.datasets.guttmann_flury2025.GuttmannFlury2025_ME object at 0x7fafa61a9420>, <moabb.datasets.guttmann_flury2025.GuttmannFlury2025_MI object at 0x7fafa61a9990>, <moabb.datasets.hefmi_ich2025.HefmiIch2025 object at 0x7fafa61a9870>, <moabb.datasets.kaya2018.Kaya2018 object at 0x7fafa61a9270>, <moabb.datasets.kumar2024.Kumar2024 object at 0x7fafa61a9780>, <moabb.datasets.Lee2019.Lee2019_MI object at 0x7fafa61a97e0>, <moabb.datasets.liu2024.Liu2024 object at 0x7fafa61a9540>, <moabb.datasets.physionet_mi.PhysionetMI object at 0x7fafa61a90f0>, <moabb.datasets.schirrmeister2017.Schirrmeister2017 object at 0x7fafa61a98d0>, <moabb.datasets.bbci_eeg_fnirs.Shin2017A object at 0x7fafa61a9840>, <moabb.datasets.stieger2021.Stieger2021 object at 0x7fafa61a98a0>, <moabb.datasets.wairagkar2018.Wairagkar2018 object at 0x7fafa61a9390>, <moabb.datasets.Weibo2014.Weibo2014 object at 0x7fafa61a96c0>, <moabb.datasets.wu2020.Wu2020 object at 0x7fafa61a9660>, <moabb.datasets.yang2025.Yang2025 object at 0x7fafa61a91e0>, <moabb.datasets.Zhou2016.Zhou2016 object at 0x7fafa61a97b0>, <moabb.datasets.zhou2020.Zhou2020 object at 0x7fafa61a9000>]

Or you can run a search through the available datasets:

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

[<moabb.datasets.aguilera_rodriguez2025.AguileraRodriguez2025 object at 0x7fafa95f3eb0>, <moabb.datasets.alex_mi.AlexMI object at 0x7fafa61a9270>, <moabb.datasets.bcicomp2020_upper_limb.BCIComp2020UpperLimb object at 0x7fafa61a9870>, <moabb.datasets.bnci.bnci_2014.BNCI2014_001 object at 0x7fafa61a9990>, <moabb.datasets.bnci.bnci_2014.BNCI2014_002 object at 0x7fafa61a9420>, <moabb.datasets.bnci.bnci_2014.BNCI2014_004 object at 0x7fafa61a94b0>, <moabb.datasets.bnci.bnci_2015.BNCI2015_001 object at 0x7fafa61a8c70>, <moabb.datasets.bnci.bnci_2015.BNCI2015_004 object at 0x7fafa61a9780>, <moabb.datasets.bnci.bnci_2019.BNCI2019_001 object at 0x7fafa61a8be0>, <moabb.datasets.bnci.bnci_2020.BNCI2020_001 object at 0x7fafa61a8cd0>, <moabb.datasets.bnci.bnci_2022_001.BNCI2022_001 object at 0x7fafa61a93f0>, <moabb.datasets.bnci.bnci_2024_001.BNCI2024_001 object at 0x7fafa61a9630>, <moabb.datasets.bnci.bnci_2025.BNCI2025_001 object at 0x7fafa61a9570>, <moabb.datasets.bnci.bnci_2025.BNCI2025_002 object at 0x7fafa61a8c40>, <moabb.datasets.brandl2020.Brandl2020 object at 0x7fafa61a96f0>, <moabb.datasets.chang2025.Chang2025 object at 0x7fafa61a9540>, <moabb.datasets.gigadb.Cho2017 object at 0x7fafa61a9720>, <moabb.datasets.dreyer2023.Dreyer2023 object at 0x7fafa61a90f0>, <moabb.datasets.dreyer2023.Dreyer2023A object at 0x7fafa61a97e0>, <moabb.datasets.dreyer2023.Dreyer2023B object at 0x7fafa61a98d0>, <moabb.datasets.dreyer2023.Dreyer2023C object at 0x7fafa61a98a0>, <moabb.datasets.fake.FakeDataset object at 0x7fafa61a96c0>, <moabb.datasets.forenzo2023.Forenzo2023 object at 0x7fafa61a9390>, <moabb.datasets.gao2026.Gao2026 object at 0x7fafa61a9840>, <moabb.datasets.mpi_mi.GrosseWentrup2009 object at 0x7fafa61a9660>, <moabb.datasets.guttmann_flury2025.GuttmannFlury2025_ME object at 0x7fafa61a91e0>, <moabb.datasets.guttmann_flury2025.GuttmannFlury2025_MI object at 0x7fafa61a9000>, <moabb.datasets.hefmi_ich2025.HefmiIch2025 object at 0x7fafa61a9330>, <moabb.datasets.jeong2020.Jeong2020 object at 0x7fafa61a97b0>, <moabb.datasets.kaya2018.Kaya2018 object at 0x7fafa61a95d0>, <moabb.datasets.kumar2024.Kumar2024 object at 0x7fafa61a91b0>, <moabb.datasets.Lee2019.Lee2019_MI object at 0x7fafa61a93c0>, <moabb.datasets.liu2024.Liu2024 object at 0x7fafa61a9180>, <moabb.datasets.liu2025.Liu2025 object at 0x7fafa61a9090>, <moabb.datasets.ma2020.Ma2020 object at 0x7fafa61a9060>, <moabb.datasets.nguyen2017.Nguyen2017_L object at 0x7fafa61a8eb0>, <moabb.datasets.nguyen2017.Nguyen2017_S object at 0x7fafa61a8fa0>, <moabb.datasets.nguyen2017.Nguyen2017_SL object at 0x7fafa61a9930>, <moabb.datasets.nguyen2017.Nguyen2017_V object at 0x7fafa61a8fd0>, <moabb.datasets.nieto2022.Nieto2022 object at 0x7fafa61a8c10>, <moabb.datasets.upper_limb.Ofner2017 object at 0x7fafa61a8b50>, <moabb.datasets.physionet_mi.PhysionetMI object at 0x7fafa61a9240>, <moabb.datasets.pressel2016.Pressel2016 object at 0x7fafa61a9600>, <moabb.datasets.rozado2015.Rozado2015 object at 0x7fafa61a9120>, <moabb.datasets.schirrmeister2017.Schirrmeister2017 object at 0x7fafa61a9210>, <moabb.datasets.bbci_eeg_fnirs.Shin2017A object at 0x7fafa61a92a0>, <moabb.datasets.bbci_eeg_fnirs.Shin2017B object at 0x7fafa61a8ca0>, <moabb.datasets.stieger2021.Stieger2021 object at 0x7fafa61a8ee0>, <moabb.datasets.tavakolan2017.Tavakolan2017 object at 0x7fafa61a9810>, <moabb.datasets.triana_guzman2024.TrianaGuzman2024 object at 0x7fafa61a8f70>, <moabb.datasets.wairagkar2018.Wairagkar2018 object at 0x7fafa61a8e50>, <moabb.datasets.Weibo2014.Weibo2014 object at 0x7fafa61a8e80>, <moabb.datasets.wu2020.Wu2020 object at 0x7fafa61a8e20>, <moabb.datasets.yang2025.Yang2025 object at 0x7fafa61a9300>, <moabb.datasets.yi2025.Yi2025 object at 0x7fafa61a9900>, <moabb.datasets.zhang2017.Zhang2017 object at 0x7fafa61a8dc0>, <moabb.datasets.zhou2020.Zhou2020 object at 0x7fafa61a8d90>, <moabb.datasets.zuo2025.Zuo2025 object at 0x7fafa61a94e0>]

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
# You can inject custom scoring directly into the paradigm (single or multi-metric).
custom_scorer = [accuracy_score, (roc_auc_score, {"needs_threshold": True})]
paradigm = LeftRightImagery(fmin=fmin, fmax=fmax, scorer=custom_scorer)

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)

[codecarbon WARNING @ 15:49:48] Multiple instances of codecarbon are allowed to run at the same time.
2026-04-27 15:50:15,067 INFO MainThread moabb.evaluations.base AM+LDA | BNCI2014-001 | 1 | 0train: Score 0.729
2026-04-27 15:50:15,067 INFO MainThread moabb.evaluations.base AM+SVM | BNCI2014-001 | 1 | 0train: Score 0.743
2026-04-27 15:50:15,067 INFO MainThread moabb.evaluations.base AM+LDA | BNCI2014-001 | 1 | 1test: Score 0.715
2026-04-27 15:50:15,067 INFO MainThread moabb.evaluations.base AM+SVM | BNCI2014-001 | 1 | 1test: Score 0.715
2026-04-27 15:50:15,067 INFO MainThread moabb.evaluations.base AM+LDA | BNCI2014-001 | 2 | 0train: Score 0.597
2026-04-27 15:50:15,067 INFO MainThread moabb.evaluations.base AM+SVM | BNCI2014-001 | 2 | 0train: Score 0.500
2026-04-27 15:50:15,067 INFO MainThread moabb.evaluations.base AM+LDA | BNCI2014-001 | 2 | 1test: Score 0.521
2026-04-27 15:50:15,067 INFO MainThread moabb.evaluations.base AM+SVM | BNCI2014-001 | 2 | 1test: Score 0.500
/home/runner/work/moabb/moabb/moabb/analysis/results.py:189: H5pyDeprecationWarning: Creating a dataset without passing data or dtype is deprecated. Pass an explicit dtype. Using dtype='f4' will keep the current default behaviour.
  dset.create_dataset(

Results are returned as a pandas DataFrame. When multiple metrics are provided, MOABB adds a primary score plus one column per metric (e.g., score_accuracy_score, score_roc_auc_score).

print(results.head())

      score      time  ...  pipeline                  codecarbon_task_name
0  0.743056  0.126022  ...    AM+SVM  09a672e6-4cf5-4d64-a820-f3dfcc14afdc
1  0.715278  0.120645  ...    AM+SVM  46dc989a-f565-4683-a7b6-618c39394506
2  0.500000  0.222219  ...    AM+SVM  e71d60bf-1f48-4749-bec3-abdb97285eaa
3  0.500000  0.148982  ...    AM+SVM  01546197-5f6b-4ce2-afad-effac2560184
4  0.729167  0.015080  ...    AM+LDA  0d1a69f0-df4a-463f-b4f9-adc1b51da1e1

[5 rows x 15 columns]