eegdash.org

What it is

EEGDash is a Python library for open neuroscience data, positioned as a unified entry point for “700+ BIDS-first EEG/MEG datasets.” It can be installed with pip install eegdash and runs locally in Python 3.10+ environments. Its core value is not general-purpose data labeling or experiment management, but helping researchers quickly search, load, preprocess, and train PyTorch models—especially for reproducibility experiments involving EEG, MEG, and related modalities.

Core capabilities and ecosystem

Functionally, EEGDash takes a search-first approach, built around unified metadata for discovering datasets, file records, modalities, tasks, and queues. It covers five modalities: EEG, MEG, fNIRS, EMG, and iEEG, and provides a curated catalog of 700+ datasets. Its BIDS-first design is a key strength, helping teams and tools maintain consistent metadata. In terms of ecosystem, it explicitly works alongside MNE-Python, braindecode, and PyTorch, and also references sources or toolchains such as EEGPrep, OpenNeuro, NEMAR, and Zenodo. In addition to the Python API, it offers a FastAPI REST metadata service for querying records, counts, dataset summaries, metadata, stats, and more; admin endpoints require a Bearer token.

Pricing, open source, and documentation

The site clearly labels it as open source on GitHub, and community contributions are handled through GitHub. No commercial tiers or paid plans are mentioned, so it is best understood as a free open-source tool. The documentation is relatively strong, with Quick Start, Install, Examples, Concepts, API Reference, Datasets API, many dataset pages, and REST endpoint descriptions. For developers, the onboarding path is clear. However, the main content does not fully explain self-hosted deployment, permission models, version stability, SLA, or enterprise support.

Pros, cons, and who it is for

Its strengths are a rich dataset catalog, a high degree of standardization, close integration with mainstream neuroscience Python toolchains, support for local execution, and automated API querying. Its limitations are also clear: it is highly domain-specific and mainly serves neuroscience data research. If you need a general machine learning data platform, visual experiment management, or enterprise-grade data governance, it may not be a good fit. EEGDash is best suited to neuroscience labs, EEG/MEG algorithm researchers, and teams that need to train and evaluate models across public datasets.

Access from China

The main content does not provide information about access from mainland China, mirrors, payment, or download acceleration. Because it depends on GitHub, public data sources, and data.eegdash.org, the actual experience may vary depending on the network environment, so the conclusion can only be marked as unknown. Possible alternatives or complementary options include using MNE-Python and braindecode directly, searching via OpenNeuro/NEMAR, or building an in-house BIDS data management workflow.