EEGdash›NeMAR›NM000281

Iss. 281 · 193 subjects · 25253 recordings · CC-BY-NC-SA-4.0

Dataset Brief · emg2pose

NM000281: emg dataset, 193 subjects#

emg2pose: Surface EMG and Hand Pose

Access recordings and metadata through EEGDash.

Citation: Sasha Salter, Richard Warren, Collin Schlager, Adrian Spurr, Shangchen Han, Rohin Bhasin, Yujun Cai, Peter Walkington, Anuoluwapo Bolarinwa, Robert Wang, Nathan Danielson, Josh Merel, Eftychios Pnevmatikakis, Jesse Marshall, Alexandre Gramfort (20). emg2pose: Surface EMG and Hand Pose. 10.82901/nemar.nm000281

Modality: emg Subjects: 193 Recordings: 25253 License: CC-BY-NC-SA-4.0 Source: nemar

Metadata: Complete (100%)

193-participant EMG dataset — emg2pose: Surface EMG and Hand Pose.

Data & curation Sasha Salter · Richard Warren · Collin Schlager · Adrian Spurr · Shangchen Han · Rohin Bhasin · …
Senior author Alexandre Gramfort
Year 20 · Distributed via NeMAR

EMG · 36 ch2000 HzBIDS 1.11.1Task · emg2pose9 sessions

Layer 01Study

What was asked

Hypothesis, independent & dependent variables, paradigm, cohort, and the editorial caveats around what the recordings can and cannot answer.

Layer 02Signal · BIDS

What was recorded

Sidecars, channels & electrodes, coordinate system, event semantics, and quality stats from the NEMAR pipeline when available.

Layer 03Training · ML

What you can train on

Recommended access modes — MNE Raw, braindecode windows, PyTorch DataLoader — plus the targets the metadata makes addressable.

§ 01Access · Get started

Quickstart#

Get Started

Install

pip install eegdash

Access the data

from eegdash.dataset import NM000281

dataset = NM000281(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)

Query & Filter

Filter by subject

dataset = NM000281(cache_dir="./data", subject="01")

Advanced query

dataset = NM000281(
    cache_dir="./data",
    query={"subject": {"$in": ["01", "02"]}},
)

Iterate recordings

for rec in dataset:
    print(rec.subject, rec.raw.info['sfreq'])

Cite This Dataset

If you use this dataset in your research, please cite the original authors.

BibTeX

@dataset{nm000281,
  title = {emg2pose: Surface EMG and Hand Pose},
  author = {Sasha Salter and Richard Warren and Collin Schlager and Adrian Spurr and Shangchen Han and Rohin Bhasin and Yujun Cai and Peter Walkington and Anuoluwapo Bolarinwa and Robert Wang and Nathan Danielson and Josh Merel and Eftychios Pnevmatikakis and Jesse Marshall and Alexandre Gramfort},
  doi = {10.82901/nemar.nm000281},
  url = {https://doi.org/10.82901/nemar.nm000281},
}

§ 02Study · The README

About This Dataset#

emg2pose: Surface EMG and Hand Pose

This dataset is a NEMAR re-host and EMG-BIDS conversion of the public Meta emg2pose release. The original benchmark is described in:

Salter, Warren, Schlager et al. “emg2pose: A Large and Diverse Benchmark for

Surface Electromyographic Hand Pose Estimation.” NeurIPS 2024 Datasets and Benchmarks Track. arXiv:2412.02725. DOI: 10.48550/arXiv.2412.02725.

DOI

The source paper describes emg2pose as a wrist surface electromyography benchmark for hand pose estimation. It contains 2 kHz, 16-channel sEMG recordings paired with hand pose labels captured by a 26-camera motion-capture rig. The public release spans 193 users, 370 hours, 29 behavioral stages, and approximately 80 million labeled frames.

Conversion notes

NEMAR Metadata#

[![DOI](https://img.shields.io/badge/DOI-10.82901%2Fnemar.nm000281-blue)](https://doi.org/10.82901/nemar.nm000281) emg2pose: Surface EMG and Hand Pose This dataset is a NEMAR re-host and EMG-BIDS conversion of the public Meta emg2pose release. The original benchmark is described in: Salter, Warren, Schlager et al. “emg2pose: A Large and Diverse Benchmark for Surface Electromyographic Hand Pose Estimation.” NeurIPS 2024 Datasets and Benchmarks Track. arXiv:2412.02725. DOI: 10.48550/arXiv.2412.02725. The source paper describes emg2pose as a wrist surface electromyography benchmark for hand pose estimation. It contains 2 kHz, 16-channel sEMG recordings paired with hand pose labels captured by a 26-camera motion-capture rig. The public release spans 193 users, 370 hours, 29 behavioral stages, and approximately 80 million labeled frames. Conversion notes —————- The upstream HDF5 recordings were converted using a fast parallel adaptation of facebookresearch/emg2pose PR #6, contributed by Alexandre Gramfort (@agramfort). Each recording is stored as an EMG-BIDS BDF file under sub-/ses-/emg. Each BDF contains: - 16 EMG channels named emg0 through emg15. - 20 joint-angle channels named joint0 through joint19. - Joint-angle channels are represented as radian-valued MISC channels. - BAD_IK annotations mark samples where inverse-kinematics labels are all zero. - The BIDS recording entity stores the source hand side, left or right. The original upstream tarball is preserved in sourcedata as sourcedata/emg2pose_dataset.tar. The arXiv PDF and source bundle for the paper are preserved under sourcedata/paper for provenance. Known limitations —————– Exact anatomical electrode coordinates are not provided by the public release, so EMGPlacementScheme is recorded as Other. The BDF writer pads the final data record with edge values when a recording length is not an exact multiple of the BDF data-record duration; this is a format-level requirement and does not alter the original samples before the final padded block.

License: CC-BY-NC-SA-4.0

Authors:

Sasha Salter
Richard Warren
Collin Schlager
Adrian Spurr
Shangchen Han
… and 10 more

Versions:

Version	DOI	Released
`current`	10.82901/nemar.nm000281

§ 03Cohort · Participants

Cohort#

Dataset Statistics#

Channel counts: 36 ch (n=25253 recordings)

Sampling frequencies: 2000.0 Hz (n=25253 recordings)

Total recording duration: 375 h

§ 04Signal · Electrodes & trace

Signal · Electrodes & live trace#

Fig. 01 Signal & montage 36 ch · EMG · 2000 Hz · 193 subjects, 25253 recordings

Live trace viewer — sub-01 · ses-01 · task-emg2pose · run-01

Showing one representative recording out of 193 subjects and 25253 recordings in this dataset. Browse the full set on OpenNeuro; drop any other _emg.{set,edf,bdf,vhdr} file onto the viewer (or pass ?emg=<url>) to inspect it.

No scalp electrode layout is currently indexed for this dataset. Once the eegdash montage registry ingests it, the interactive viewer will appear here automatically.

NEMAR Processing Statistics#

The plots below are generated by NEMAR’s automated EEG pipeline. The histogram shows pipeline success for data cleaning and ICA decomposition, the percentage of data frames and EEG channels retained after artefact removal, line noise per channel (RMS, dB), and the age/gender distribution of participants.

HED event descriptors word cloud

§ 05Manifest · BIDS tree

Manifest#

File Explorer#

Browse the BIDS file structure of this dataset. Records are fetched on demand from the EEGDash catalog the first time you open the explorer.

Recordings—

Files—

Subjects—

Modalities—

Click to load file structure…

§ 06API · Programmatic access

API Reference#

Signature

eegdash.dataset

class

eegdash.dataset.NM000281(cache_dir, query=None, s3_bucket=None, **kwargs)

Bases: EEGDashDataset

Author (year)—

Canonical—

Importable asNM000281

Sourceeegdash/dataset/registry.py · [source ↗]

class eegdash.dataset.NM000281(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

emg2pose: Surface EMG and Hand Pose

Study:: nm000281 (NeMAR)
Author (year):: —
Canonical:: —

Also importable as: NM000281.

Modality: emg; Subject type: Unknown. Subjects: 193; recordings: 25253; tasks: 1.

Parameters:

cache_dir (str | Path) – Directory where data are cached locally.
query (dict | None) – Additional MongoDB-style filters to AND with the dataset selection. Must not contain the key dataset.
s3_bucket (str | None) – Base S3 bucket used to locate the data.
**kwargs (dict) – Additional keyword arguments forwarded to EEGDashDataset.

data_dir#

Local dataset cache directory (cache_dir / dataset_id).

Type:: Path

query#

Merged query with the dataset filter applied.

Type:: dict

records#

Metadata records used to build the dataset, if pre-fetched.

Type:: list[dict] | None

Notes

Each item is a recording; recording-level metadata are available via dataset.description. query supports MongoDB-style filters on fields in ALLOWED_QUERY_FIELDS and is combined with the dataset filter. Dataset-specific caveats are not provided in the summary metadata.

References

OpenNeuro dataset: https://openneuro.org/datasets/nm000281 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000281 DOI: https://doi.org/10.82901/nemar.nm000281

Examples

>>> from eegdash.dataset import NM000281
>>> dataset = NM000281(cache_dir="./data")
>>> recording = dataset[0]
>>> raw = recording.load()

__init__(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

save(path: str, overwrite: bool = False, offset: int = 0)[source]#

Save datasets to files by creating one subdirectory for each dataset:

path/
    0/
        0-raw.fif | 0-epo.fif
        description.json
        raw_preproc_kwargs.json (if raws were preprocessed)
        window_kwargs.json (if this is a windowed dataset)
        window_preproc_kwargs.json  (if windows were preprocessed)
        target_name.json (if target_name is not None and dataset is raw)
    1/
        1-raw.fif | 1-epo.fif
        description.json
        raw_preproc_kwargs.json (if raws were preprocessed)
        window_kwargs.json (if this is a windowed dataset)
        window_preproc_kwargs.json  (if windows were preprocessed)
        target_name.json (if target_name is not None and dataset is raw)

Parameters:

path (str) –

Directory in which subdirectories are created to store
-raw.fif | -epo.fif and .json files to.
overwrite (bool) – Whether to delete old subdirectories that will be saved to in this call.
offset (int) – If provided, the integer is added to the id of the dataset in the concat. This is useful in the setting of very large datasets, where one dataset has to be processed and saved at a time to account for its original position.

Access modesMNE → braindecode → PyTorch → ML

.rawMNE Raw object — standard tools (filter, epoch, ICA, plot_psd).mne

BaseConcatDatasetEach record is a lazy BaseDataset from braindecode — windowed via create_windows_from_events.braindecode

DataLoaderWraps the windowed dataset into a PyTorch DataLoader; supports parallel workers and on-the-fly augmentations.pytorch

Zarr cacheOptional braindecode Zarr mirror for fast resume; persisted to cache_dir.zarr

Hugging FaceNo per-dataset mirror published yet — browse the EEGDash org listing for sibling datasets. See the datasets loader API.huggingface

Croissant 1.0Machine-readable JSON-LD descriptor — NM000281.croissant.json (MLCommons schema, ingestible by PyTorch / TensorFlow / JAX).mlcommons

Examples using EEGDashcurated · start here

Find datasets with the EEGDash APIQuery the catalogue, filter by task or modality, list candidates.

Load one EEG recordingResolve a single record to an MNE Raw with channels and events.

EEG recording to PyTorch DataLoaderWrap braindecode windows in a DataLoader for model training.

Preprocess EEG and create windowsFilter, resample, epoch — and persist the windowed dataset.

Save and reload prepared dataCache a windowed dataset to disk and reattach it without recompute.

Download a dataset locallyPrefetch BIDS files to a local cache and validate the layout.

Swap any load_dataset(...) call for nm000281 to reproduce the tutorial on this dataset.

Citation

Sasha Salter, Richard Warren, Collin Schlager, Adrian Spurr, Shangchen Han, … (20). emg2pose: Surface EMG and Hand Pose. 10.82901/nemar.nm000281

Provenance

¹Contributed to nemar in BIDS format.

²Curated & ingested by the EEGDash catalog; see CITATION.cff for canonical reference.

³Persistent identifier: 10.82901/nemar.nm000281.

Related & sibling datasets

NM000104EMG · 108 subj NM000106EMG · 100 subj NM000107EMG · 100 subj NM000105EMG · 100 subj NM000159EMG · 16 subj

+ 1 more — see See Also below →

BIDS

BIDS 1.11.1

Sidecars

events · channels

Provenance

CC-BY-NC-SA-4.0 · 10.82901/nemar.nm000281

Machine-readable

schema.org/Dataset · Croissant

Mirrors

OpenNeuro · NEMAR · HF org

Dataset ID	`NM000281`
Title	emg2pose: Surface EMG and Hand Pose
Author (year)	—
Canonical	—
Importable as	`NM000281`
Year	20
Authors	Sasha Salter, Richard Warren, Collin Schlager, Adrian Spurr, Shangchen Han, Rohin Bhasin, Yujun Cai, Peter Walkington, Anuoluwapo Bolarinwa, Robert Wang, Nathan Danielson, Josh Merel, Eftychios Pnevmatikakis, Jesse Marshall, Alexandre Gramfort
License	CC-BY-NC-SA-4.0
Citation / DOI	10.82901/nemar.nm000281
Source links	OpenNeuro \| NeMAR \| Source URL

NM000281: emg dataset, 193 subjects#

Quickstart#

About This Dataset#

NEMAR Metadata#

Cohort#

Dataset Statistics#

Signal · Electrodes & live trace#

NEMAR Processing Statistics#

Manifest#

File Explorer#

API Reference#

Citation

Provenance

Related & sibling datasets

See Also#