braindecode/SignalJEPA-pretrained

⚠️ This repository is deprecated

The weights in this repository are kept online for archival purposes only. They are not compatible with current releases of braindecode and will not receive updates.

Please use the new repositories

The Signal-JEPA weights have been republished:

• braindecode/signal-jepa — full checkpoint including the 62-channel embedding table (for users whose recording channels are a subset of the Lee2019 pre-training layout).
• braindecode/signal-jepa_without-chans — SSL backbone only (channel embeddings trained from scratch on your data).

from braindecode.models import SignalJEPA

model = SignalJEPA.from_pretrained("braindecode/signal-jepa_without-chans")

See the docs for braindecode.models.SignalJEPA and the fine-tuning tutorial Fine-tuning a Foundation Model (Signal-JEPA).

Training data

The data that shaped these weights. The ids are the datasets: field of this card, which braindecode reads into model.card_data.datasets when you call from_pretrained (MOABB class names for MOABB datasets, OpenNeuro / NEMAR accessions, Hub ids otherwise). nm/on ids are datasets on NEMAR (on = NEMAR's copy of the OpenNeuro dataset with the same number). Rows marked private: are data the authors did not release; they are not in datasets:.

datasets: id	Dataset	Stage	Modality
nm000338	Lee et al. 2019 OpenBMI, motor imagery paradigm (54 subjects, 62 EEG channels), MOABB Lee2019_MI	pretraining	EEG
nm000323	Lee et al. 2019 OpenBMI, ERP (P300 speller) paradigm, MOABB Lee2019_ERP	pretraining	EEG
nm000273	Lee et al. 2019 OpenBMI, SSVEP paradigm, MOABB Lee2019_SSVEP	pretraining	EEG

• S-JEPA self-supervised pre-training (16 s examples, 60 % spatial mask, run adeuwv4s) on Lee2019 subjects 1-40, all three paradigms pooled, no labels; subjects 41-47 = SSL validation set used for early stopping / best-checkpoint selection; subjects 48-54 not used for these weights (downstream evaluation only). All 179 original tensors byte-identical; the extra pos_encoder.pos_encoder_spat.weight (19 x 30) is not in the original file and is built from the 19-channel chs_info stored in config.json (not learned on Lee2019).
• Sources: arXiv:2403.11772 (HTML), section DATASETS; https://huggingface.co/api/models/braindecode/signal-jepa/commits/main (commit 9e20126cf5); braindecode check: tensor-byte sha256 check -> compare_sjepa.json; W/bd/braindecode/models/signal_jepa.py.

Cite Braindecode

If you use this braindecode port in scientific work, cite the software on Zenodo, doi:10.5281/zenodo.17699192, and the original model and datasets cited above.

@software{braindecode,
  author = {Aristimunha, Bruno and
            Guetschel, Pierre and
            Wimpff, Martin and
            Gemein, Lukas and
            Rommel, Cedric and
            Banville, Hubert and
            Sliwowski, Maciej and
            Wilson, Daniel and
            Brandt, Simon and
            Gnassounou, Théo and
            Paillard, Joseph and
            {Junqueira Lopes}, Bruna and
            Sedlar, Sara and
            Moreau, Thomas and
            Chevallier, Sylvain and
            Gramfort, Alexandre and
            Schirrmeister, Robin Tibor},
  title = {Braindecode: toolbox for decoding raw electrophysiological brain data
           with deep learning models},
  url = {https://github.com/braindecode/braindecode},
  doi = {10.5281/zenodo.17699192},
  publisher = {Zenodo},
  license = {BSD-3-Clause},
}