Datasets Index
One line per dataset, with its size and the page that describes it. Sizes are read from the exported fact files of each project; nothing on this page is typed by hand.
AstroSpectro — LAMOST DR5 × Gaia DR3
| Data generation | Spectra | Feature file | Used by |
|---|---|---|---|
| PHY-3500 · G1 · 43,019 spectra | 43,019 | features_20260305T174836Z.csv | PCA, UMAP, t-SNE, HDBSCAN and autoencoder pages |
| G2 | 349,998 processed | features_20260522T153123Z.csv | The reference classifier, its SHAP analysis and the scientific validation |
The two generations are never combined in a single result. The reference classifier also contains Gaia/photometric columns; see the Data Leakage Audit.
Pages: Data & Datasets · LAMOST DR5 · Gaia DR3 cross-match · Master catalogue · Feature catalogue · Quality filters · Dataset versions · Class taxonomy
AstroVision — Galaxy10 DECaLS × SDSS DR16
| Dataset | Content | Size |
|---|---|---|
| Galaxy10 DECaLS | Galaxy cut-outs with morphological classes | 17,736 images |
| SDSS DR16 cross-match | Photometry and spectroscopic redshift added from VizieR | 17,665 galaxies matched |
Pages: Data Overview · Galaxy10 DECaLS · Morphology Classes · Data Preparation · SDSS DR16 Crossmatch
ξ Dark Energy — Pantheon+SH0ES and DESI DR2 BAO
| Dataset | Content | Size |
|---|---|---|
| Pantheon+SH0ES | Type Ia supernovae, after a redshift cut at z > 0.01 | 1,590 supernovae |
| DESI DR2 BAO | Baryon acoustic oscillation distance measurements | 13 measurements |
The BAO analysis fixes the sound horizon at rd = 147.09 Mpc; the Known Limitations page discusses what this implies.
Pages: Data Overview · Pantheon+SH0ES · DESI DR2 BAO · Covariance Handling
See also: Repository Map · Notebook Atlas · Figures Index.