# SDSS SEGUE DR9 Stellar Spectra with SSPP Teff, log g and [Fe/H] Labels (233,013 Stars)

## Facts

- Provider: OneAstronomy, from SDSS-III SEGUE DR9 and the SSPP (Lee et al. 2008)
- License: Public Domain (https://sdss.org/collaboration/image-use-policy/)
- Size: 7 files
- Format: Parquet, JSON (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/OneAstronomy/sspp-dr9-stellar-spectra
- Access: public, readable without an account
- Category: Astronomy & Space
- Source: https://doi.org/10.1088/0004-6256/136/5/2022
- Page: https://scigantic.com/archive/fca3532a-1e9a-5b29-a248-2a2ae5b43b2f
- Keywords: Astronomy & Space, spectroscopy, Parquet, JSON, Hugging Face

## About

Optical stellar spectra from the Sloan Digital Sky Survey SEGUE-1 and SEGUE-2 programs (Data Release 9), each labelled with atmospheric parameters from the SEGUE Stellar Parameter Pipeline (SSPP; Lee et al. 2008, Allende Prieto et al. 2008). 233,013 spectra are split 186,409 train, 23,302 validation and 23,302 test. Each row holds a flux-calibrated 1D spectrum resampled to a common grid of 3,878 wavelength bins over about 3,820 to 9,200 Angstrom (median-normalised flux, raw flux, inverse variance and bad-pixel mask; the grid is in wavelength_grid.json), the adopted effective temperature TEFF_ADOP (4,003 to 9,593 K), surface gravity LOGG_ADOP, metallicity FEH_ADOP (down to -4.35 dex), alpha enhancement AFE, radial velocity RV_ADOP, S/N, ra, dec, HEALPix index and SDSS SPECOBJID. Cuts: ZWARNING 0 or 16, finite Teff, log g and [Fe/H], and S/N at least 10. Data are Parquet under train/, validation/ and test/ (about 15 GB, measured 2026-11-16), readable with Hugging Face datasets, pandas or pyarrow. Uses: data-driven stellar parameter regression (The Cannon-style models, CNNs), metal-poor star searches and Galactic archaeology. SDSS states that all data in its public releases are in the public domain; cite Lee et al. 2008 and acknowledge SDSS-III.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("OneAstronomy/sspp-dr9-stellar-spectra")
print(ds)
```

## Data structure

File types in sampled files: Parquet, JSON

Example file: `test/data.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/fca3532a-1e9a-5b29-a248-2a2ae5b43b2f
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/fca3532a-1e9a-5b29-a248-2a2ae5b43b2f
- Over MCP: `get_schema_card` with id `fca3532a-1e9a-5b29-a248-2a2ae5b43b2f` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [JSON datasets](https://scigantic.com/topics/json)
- [Astronomy & Space datasets](https://scigantic.com/topics/astronomy-and-space)
- [Spectroscopy datasets](https://scigantic.com/topics/spectroscopy)
