https://github.com/eegdash/eegdash
EEG-DaSh: an open data, tool, and compute resource. a Python library and catalog for 700+ BIDS-first EEG, MEG, fNIRS, EMG, and iEEG datasets, ML-ready via PyTorch
https://github.com/eegdash/eegdash
aws aws-s3 data-sharing deep-learning eeg eeg-signals electroencephalography fair-principles machine-learning magnetoencephalography meg nemar neuroinformatics neuroscience open-data openneuro
Last synced: about 8 hours ago
JSON representation
EEG-DaSh: an open data, tool, and compute resource. a Python library and catalog for 700+ BIDS-first EEG, MEG, fNIRS, EMG, and iEEG datasets, ML-ready via PyTorch
- Host: GitHub
- URL: https://github.com/eegdash/eegdash
- Owner: eegdash
- License: bsd-3-clause
- Created: 2024-09-18T18:26:46.000Z (almost 2 years ago)
- Default Branch: develop
- Last Pushed: 2026-07-01T13:08:20.000Z (about 17 hours ago)
- Last Synced: 2026-07-01T15:09:27.081Z (about 15 hours ago)
- Topics: aws, aws-s3, data-sharing, deep-learning, eeg, eeg-signals, electroencephalography, fair-principles, machine-learning, magnetoencephalography, meg, nemar, neuroinformatics, neuroscience, open-data, openneuro
- Language: Python
- Homepage: http://eegdash.org/
- Size: 284 MB
- Stars: 74
- Watchers: 7
- Forks: 9
- Open Issues: 47
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Citation: CITATION.cff
- Codemeta: codemeta.json
Awesome Lists containing this project
README
# EEG-Dash
[](https://pypi.org/project/eegdash/)
[](https://sccn.github.io/eegdash)
[](LICENSE)
[](https://pypi.org/project/eegdash/)
[](https://pepy.tech/project/eegdash)
[](https://github.com/eegdash/EEGDash/blob/main/coverage.json)
EEG-DaSh is a data-sharing archive for MEEG (EEG, MEG) recordings contributed by collaborating labs. It preserves publicly funded research data and exposes it in a form that machine learning and deep learning workflows can use directly.
## Data source
The archive draws on 25 labs and 27,053 participants, with recordings covering both EEG and MEG. Subjects include healthy controls and clinical groups: ADHD, depression, schizophrenia, dementia, autism, and psychosis. Tasks range across sleep, meditation, and cognitive paradigms. EEG-DaSh also pulls in 330 BIDS-formatted MEEG datasets converted from NEMAR.
## Data format
EEGDash queries return a **PyTorch Dataset**. The format plugs directly into PyTorch's `DataLoader` for batching, shuffling, and parallel loading, which matters when training models on large EEG corpora.
## Data preprocessing
EEGDash datasets are [braindecode](https://braindecode.org/stable/index.html) datasets, which are themselves PyTorch datasets. Any preprocessing that works on a braindecode dataset works on an EEGDash dataset. See the braindecode tutorials for the available options.
## EEG-Dash usage
### Install
Requires Python 3.10 or higher. Use whichever environment manager you prefer.
```bash
pip install eegdash
```
Verify the install in a Python session:
```python
from eegdash import EEGDash
```
See the tutorials at [eegdash.org](https://eegdash.org/) for end-to-end examples.
## Education (coming soon)
We run workshops and student training events with US and Israeli partners, online and in person. 2025 dates will go out on the EEGLABNEWS mailing list. [Subscribe here](https://sccn.ucsd.edu/mailman/listinfo/eeglabnews).
## About EEG-DaSh
EEG-DaSh is a collaborative initiative between the United States and Israel, supported by the National Science Foundation (NSF). The partnership brings together experts from the Swartz Center for Computational Neuroscience (SCCN) at the University of California San Diego (UCSD) and Ben-Gurion University (BGU) in Israel.
