emg2qwerty β generic baseline (TDS-Conv-CTC, generic model trained on 100 users)
Pre-remapped braindecode-compatible copy of the upstream
generic.ckpt
from the emg2qwerty release (Sivakumar et al., NeurIPS 2024 D&B Track).
from braindecode.models import EMG2QwertyNet
model = EMG2QwertyNet.from_pretrained("braindecode/emg2qwerty-generic")
Source
Upstream repository: https://github.com/facebookresearch/emg2qwerty Paper: Sivakumar V, Seely J, Du A, Bittner S, Berenzweig A, Bolarinwa A, Gramfort A, Mandel M. emg2qwerty: A Large Dataset with Baselines for Touch Typing using Surface Electromyography. Advances in Neural Information Processing Systems (NeurIPS), Datasets and Benchmarks Track, 2024.
Modification from upstream
Upstream models/generic.ckpt is a PyTorch-Lightning checkpoint
of emg2qwerty.lightning.TDSConvCTCModule, whose inner
nn.Sequential exposes the classifier head as item 4
(model.4.{weight,bias}). braindecode's EMG2QwertyNet
exposes the same head as a named attribute
(final_layer.{weight,bias}).
The remap is a two-key rename, applied once and saved here:
| Upstream key | braindecode key |
|---|---|
model.4.weight |
final_layer.weight |
model.4.bias |
final_layer.bias |
All 49 other keys (BatchNorm, MLP, TDS conv blocks) match
verbatim β both modules expose the backbone as
self.model = nn.Sequential(...), so the keys already share the
model.<index>. prefix and need no rename. Weights are
otherwise unchanged from upstream.
Conversion is reproducible from
neuralbench-repo/scripts/convert_emg2qwerty_checkpoint.py.
Verification
BatchNorm running statistics on the first layer match upstream:
| Stat | This checkpoint | Expected |
|---|---|---|
model.0.batch_norm.running_mean.mean |
0.511 | β 0.51 |
model.0.batch_norm.running_var.mean |
1.146 | β 1.15 |
Forward pass on a 1Γ32Γ8000 random input returns shape
(1, 373, 99) β the 4 s @ 2 kHz window after the TDS encoder +
CTC head.
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 |
|---|---|---|---|
nm000104 |
emg2qwerty: sEMG of touch typing, two 16-ch wristbands at 2 kHz, 108 users, 1,135 sessions, 346 h (Sivakumar et al. 2024, NeurIPS D&B) | training | EMG |
- Generic user model (TDS-Conv-CTC): trained on sessions of the 100 non-test users, 2 sessions per user held for validation; the 8 personalization test users are excluded.
- Sources: arXiv:2410.20081 (HTML) Sec. 4; https://github.com/facebookresearch/emg2qwerty README 'Training'.
License
CC BY-NC-SA 4.0 (Attribution-NonCommercial-ShareAlike 4.0 International), inherited from the upstream emg2qwerty release.
- Attribution (BY) β cite Sivakumar et al. (2024) and link this repository when the weights are used or redistributed.
- NonCommercial (NC) β not licensed for commercial use.
- ShareAlike (SA) β derivatives must be released under the same CC BY-NC-SA 4.0 license.
braindecode itself is BSD-3-Clause; that license applies to the code, not to these weights. The weights are governed solely by CC BY-NC-SA 4.0.
Intended use
Drop-in pretrained backbone for the
emg/qwerty
CTC keystroke-decoding task in NeuralBench, or any other research
workflow consuming braindecode.models.EMG2QwertyNet.
Per the source paper (Table 3, 8 held-out users, no language model): the generic model reaches a test CER of 55.4 % Β± 4.1 on unseen users; fine-tuning it on each user's own sessions brings the test CER to 11.3 % Β± 4.5 (6.95 % Β± 3.6 with a 6-gram character LM). Revision note (2026-11-08): an earlier version of this card said "108-subject pretrained" and "zero-shot CER β 16 %"; both were wrong.
Citation
@inproceedings{sivakumar2024emg2qwerty,
title = {emg2qwerty: A Large Dataset with Baselines for
Touch Typing using Surface Electromyography},
author = {Sivakumar, Viswanath and Seely, Jeffrey and Du,
Alan and Bittner, Sean and Berenzweig, Adam and
Bolarinwa, Anuoluwapo and Gramfort, Alexandre and
Mandel, Michael},
booktitle = {Advances in Neural Information Processing Systems
(NeurIPS), Datasets and Benchmarks Track},
year = {2024},
url = {https://github.com/facebookresearch/emg2qwerty},
}
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, ThΓ©o 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},
}
- Downloads last month
- 43