Draft org card for EEGDash — hero, real stats, featured datasets

org-readme/README.md

@@ -0,0 +1,130 @@
+<div align="center">
+
+<img src="https://raw.githubusercontent.com/eegdash/EEGDash/main/docs/source/_static/eegdash_long.svg" width="340" alt="EEGDash" />
+
+### The open catalog of EEG / MEG datasets — indexed, described, and loadable with one line of Python.
+
+[![PyPI](https://img.shields.io/pypi/v/eegdash?style=flat-square&logo=pypi&logoColor=white&color=0072B2)](https://pypi.org/project/eegdash/)
+[![Python](https://img.shields.io/pypi/pyversions/eegdash?style=flat-square&color=0072B2)](https://pypi.org/project/eegdash/)
+[![License](https://img.shields.io/badge/license-BSD--3--Clause-009E73?style=flat-square)](https://github.com/eegdash/EEGDash/blob/main/LICENSE)
+[![GitHub](https://img.shields.io/github/stars/eegdash/EEGDash?style=flat-square&logo=github&color=E69F00)](https://github.com/eegdash/EEGDash)
+[![Downloads](https://static.pepy.tech/badge/eegdash)](https://pepy.tech/project/eegdash)
+
+[**🗺️ Browse the catalog**](https://huggingface.co/spaces/EEGDash/catalog) · [**📚 Docs**](https://eegdash.org) · [**💻 GitHub**](https://github.com/eegdash/EEGDash) · [**📦 PyPI**](https://pypi.org/project/eegdash/)
+
+</div>
+
+---
+
+## Why this exists
+
+Publicly funded neuroscience produces a river of EEG/MEG data — but most of it is stuck: BIDS trees on S3, idiosyncratic loaders, inconsistent metadata, no way to search across studies. **EEGDash is the index.** We catalog every publicly shared EEG/MEG study we can find, normalize the metadata, and expose each one as a single Python object that plugs straight into [braindecode](https://braindecode.org) and PyTorch.
+
+Raw data **never gets rehosted** — each repo here is a pointer to its canonical source (OpenNeuro, NEMAR, lab S3). `EEGDashDataset` handles the download, caching, and conversion on demand.
+
+## By the numbers
+
+<div align="center">
+
+| | | | |
+|:---:|:---:|:---:|:---:|
+| **736** | **40,361** | **222,750** | **85,298** |
+| datasets | subjects | recordings | hours of data |
+
+| **600** | **571** | **73** | **55** | **22** |
+|:---:|:---:|:---:|:---:|:---:|
+| 🤗 mirrored | EEG | iEEG | MEG | fNIRS |
+
+</div>
+
+**Clinical populations covered:** Parkinson's · Alzheimer's · ADHD · Schizophrenia · Depression · Epilepsy · Dementia · Dyslexia · Development · post-surgical · healthy controls.
+
+**Experimental paradigms:** Visual · Auditory · Motor · Multisensory · Tactile · Resting State · Sleep · Anesthesia.
+
+**Data sources:** 546 from [OpenNeuro](https://openneuro.org) · 190 from [NEMAR](https://nemar.org).
+
+## Get started in 30 seconds
+
+```bash
+pip install eegdash
+```
+
+```python
+from eegdash import EEGDashDataset
+
+# Load any dataset in the catalog by its ID...
+ds = EEGDashDataset(dataset="ds002718", cache_dir="./cache")
+print(len(ds), "recordings")
+
+# ...or by its canonical name — every known alias is registered:
+from eegdash.dataset import Wakeman2015
+ds = Wakeman2015(cache_dir="./cache")
+
+# ...or pull a Hub-mirrored Zarr copy directly:
+from braindecode.datasets import BaseConcatDataset
+ds = BaseConcatDataset.pull_from_hub("EEGDash/ds002718")
+
+# Plug into PyTorch — EEGDash datasets ARE braindecode datasets.
+from torch.utils.data import DataLoader
+loader = DataLoader(ds, batch_size=32, shuffle=True)
+```
+
+## Start here
+
+- 🗺️ **[Catalog Space](https://huggingface.co/spaces/EEGDash/catalog)** — searchable, filterable view over all 736 datasets with interactive treemap/sankey/growth plots.
+- 📚 **[Documentation](https://eegdash.org)** — tutorials, per-dataset cards, preprocessing recipes, API reference.
+- 🎓 **[Tutorials](https://eegdash.org/generated/auto_examples/index.html)** — end-to-end examples from load to trained model.
+- 🧪 **[EEG 2025 challenge](https://eegdash.org/eeg2025/)** — benchmarks on Healthy Brain Network data, 22 mini-releases ready to download.
+
+## Featured datasets
+
+| Dataset | What it is | Population | Size | Canonical |
+|---|---|---|---|---|
+| [`ds002718`](https://huggingface.co/datasets/EEGDash/ds002718) | Face recognition (Wakeman & Henson) | Healthy (18) | 4.3 GB | `Wakeman2015` |
+| [`ds003800`](https://huggingface.co/datasets/EEGDash/ds003800) | Resting state / auditory | Parkinson's | small | — |
+| [`ds002799`](https://huggingface.co/datasets/EEGDash/ds002799) | Patient-day recording | Dementia | — | — |
+| [`EEG2025r1`](https://huggingface.co/datasets/EEGDash/eeg2025r1) | HBN multi-task — 136 subjects, 10 paradigms | Development | 20.6 GB | `HBN_r1_bdf` |
+| [`ds004551`](https://huggingface.co/datasets/EEGDash/ds004551) | iEEG recordings | Surgery | — | — |
+| [`ds000117`](https://huggingface.co/datasets/EEGDash/ds000117) | MEG+EEG multi-modal face | Healthy | — | `WakemanHenson_MEEG` |
+
+**[Browse all 600 mirrored datasets →](https://huggingface.co/EEGDash)**
+
+## Backed by
+
+EEGDash is a **U.S.–Israel collaboration** supported by the **National Science Foundation** as part of the EEG-DaSh initiative:
+
+- **Swartz Center for Computational Neuroscience (SCCN)** — University of California, San Diego
+- **Ben-Gurion University of the Negev** — Beer-Sheva, Israel
+
+Built on top of (and giving back to) the wider open-neuroscience ecosystem: [MNE-Python](https://mne.tools), [braindecode](https://braindecode.org), [EEGLAB](https://eeglab.org), [BIDS](https://bids.neuroimaging.io), [OpenNeuro](https://openneuro.org), [NEMAR](https://nemar.org).
+
+## Contribute
+
+The entire catalog is regenerated from **one CSV** (`eegdash/dataset/dataset_summary.csv`) and the EEGDash API. Missing a dataset? Spotted wrong metadata?
+
+- [Open an issue](https://github.com/eegdash/EEGDash/issues) — we add datasets on request.
+- Fix the CSV or push a new `dataset_description.json` — every stub on HF regenerates automatically from the single source.
+- Bigger ideas? [CONTRIBUTING.md](https://github.com/eegdash/EEGDash/blob/main/CONTRIBUTING.md).
+
+## Cite
+
+```bibtex
+@software{eegdash,
+  title = {EEGDash: An open catalog and loader for EEG/MEG datasets},
+  author = {{EEG-DaSh contributors}},
+  url = {https://github.com/eegdash/EEGDash},
+  year = {2025},
+  license = {BSD-3-Clause}
+}
+```
+
+When you use a specific dataset, **follow its upstream citation policy** — the link is in every dataset's card under *How to cite*.
+
+---
+
+<div align="center">
+<sub>
+EEGDash code is <b>BSD-3-Clause</b>. Each dataset retains its upstream license — check the card before redistribution.<br/>
+<em>Open, indexed, loadable.</em>
+</sub>
+</div>