# Multimodal Universe DESI x SDSS Cross-Matched Spectra Pairs (13,794 Galaxies)

## Facts

- Provider: Michael J. Smith (UniverseTBD), from Multimodal Universe DESI and SDSS subsets
- License: CC-BY-4.0 (https://data.desi.lbl.gov/doc/acknowledgments/)
- Size: 9 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Smith42/desi_sdss_crossmatched
- Access: public, readable without an account
- Category: Astronomy & Space
- Source: https://arxiv.org/abs/2412.02527
- Page: https://scigantic.com/archive/cd0addff-3d3c-513f-96fa-b1f798e0bac0
- Keywords: Astronomy & Space, spectroscopy, Parquet, Hugging Face

## About

Paired optical spectra of the same objects observed by both the Dark Energy Spectroscopic Instrument (DESI, Early Data Release) and the Sloan Digital Sky Survey (SDSS/BOSS), cross-matched by sky position from the Multimodal Universe DESI EDR SV3 and SDSS subsets. Each of 13,794 rows holds a DESI spectrum (7,781 pixels: flux, inverse variance, line spread function width, wavelength and mask) and an SDSS spectrum with the same fields, the DESI redshift Z with ZERR and ZWARN, the SDSS redshift with Z_ERR, ZWARNING and velocity dispersion VDISP, SDSS spectro-photometric and synthetic ugriz fluxes with inverse variances, Legacy Surveys g, r, z total and fiber fluxes, and Galactic E(B-V). The data are Parquet shards under data/ (one train split, about 1.76 GB, measured 2026-11-16), loadable with Hugging Face datasets, pandas or pyarrow. Uses include cross-instrument spectral translation, testing spectrum foundation models for instrument invariance, and redshift calibration between surveys. DESI data are licensed CC-BY-4.0 and SDSS public data are in the public domain; the packager also tags the repository CC-BY-SA-4.0. Cite DESI EDR, SDSS and the Multimodal Universe paper (arXiv:2412.02527).

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("Smith42/desi_sdss_crossmatched")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00006.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/cd0addff-3d3c-513f-96fa-b1f798e0bac0
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/cd0addff-3d3c-513f-96fa-b1f798e0bac0
- Over MCP: `get_schema_card` with id `cd0addff-3d3c-513f-96fa-b1f798e0bac0` at https://api.scigantic.com/mcp

## Related topics

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