juliensimon/gaia-dr3-young-stellar-objects

---
license: cc-by-4.0
pretty_name: "Gaia DR3 Young Stellar Objects"
language:
- en
description: "Gaia DR3 young stellar object (YSO) candidates — 79,375 pre-main-sequence stars with classification scores, variability parameters, astrometry, and multi-band photometry from ESA Gaia mission."
task_categories:
- tabular-classification
- tabular-regression
tags:
- space
- gaia
- yso
- young-stars
- star-formation
- esa
- astronomy
- open-data
- tabular-data
size_categories:
- 10K<n<100K
configs:
- config_name: default
data_files:
- split: train
path: data/gaia_dr3_young_stellar_objects.parquet
default: true
---

# Gaia DR3 Young Stellar Objects

*Part of the [Astronomy Datasets](https://huggingface.co/collections/juliensimon/astronomy-datasets-69c24caf2f17e36128946743) collection on Hugging Face.*

The Gaia DR3 young stellar object (YSO) catalog, containing **79,375** YSO candidates
identified by the ESA Gaia mission's variability classification pipeline. Each source includes
a YSO classification confidence score, variability statistics (amplitudes, standard deviations,
skewness, kurtosis), astrometry (positions, parallax, proper motions), and multi-band
photometry (G, BP, RP).

## Dataset description

Young stellar objects are pre-main-sequence stars still in the process of forming, often
surrounded by circumstellar disks and exhibiting irregular photometric variability. Gaia's
all-sky photometric survey identified these candidates through automated variability
classification in the `vari_classifier_result` table. The `best_class_score` field gives
the classifier's confidence for the YSO label (higher = more confident).

This dataset joins three Gaia DR3 tables:
- **`vari_classifier_result`** — YSO classification and confidence score
- **`vari_summary`** — variability statistics (mean/median magnitudes, amplitudes, scatter)
- **`gaia_source`** — astrometry (ra, dec, parallax, proper motion) and catalog photometry

## Key columns

| Column | Type | Description |
|--------|------|-------------|
| `source_id` | int64 | Gaia DR3 unique source identifier |
| `best_class_name` | string | Classification label (always "YSO" in this dataset) |
| `best_class_score` | float64 | Classification confidence score (0-1) |
| `ra` | float64 | Right ascension (deg, ICRS, epoch 2016.0) |
| `dec` | float64 | Declination (deg, ICRS, epoch 2016.0) |
| `l` | float64 | Galactic longitude (deg) |
| `b` | float64 | Galactic latitude (deg) |
| `parallax` | float64 | Parallax (mas) |
| `parallax_error` | float64 | Parallax uncertainty (mas) |
| `pmra` | float64 | Proper motion in RA (mas/yr) |
| `pmdec` | float64 | Proper motion in Dec (mas/yr) |
| `phot_g_mean_mag` | float64 | G-band mean magnitude (catalog) |
| `median_mag_g_fov` | float64 | Median G-band magnitude (variability) |
| `median_mag_bp` | float64 | Median BP-band magnitude (variability) |
| `median_mag_rp` | float64 | Median RP-band magnitude (variability) |
| `bp_rp` | float64 | BP-RP color index (derived) |
| `std_dev_mag_g_fov` | float64 | G-band magnitude standard deviation |
| `trimmed_range_mag_g_fov` | float64 | G-band variability amplitude |
| `skewness_mag_g_fov` | float64 | G-band magnitude skewness |
| `kurtosis_mag_g_fov` | float64 | G-band magnitude kurtosis |
| `num_selected_g_fov` | Int32 | Number of G-band observations used |

Full schema includes 40 columns with variability metrics and photometric parameters.

## Quick stats

- **79,375** YSO candidates
- Median G magnitude: 16.99
- Median classification score: 0.512

## Usage

```python
from datasets import load_dataset

ds = load_dataset("juliensimon/gaia-dr3-young-stellar-objects", split="train")
df = ds.to_pandas()

# Classification score distribution
print(df["best_class_score"].describe())

# High-confidence YSOs (score > 0.5)
confident = df[df["best_class_score"] > 0.5]
print(f"High-confidence YSOs: {len(confident):,}")

# Color-magnitude diagram
import matplotlib.pyplot as plt
plt.scatter(df["bp_rp"], df["phot_g_mean_mag"], s=1, alpha=0.3, c=df["best_class_score"], cmap="viridis")
plt.colorbar(label="Classification score")
plt.xlabel("BP - RP (mag)")
plt.ylabel("G (mag)")
plt.gca().invert_yaxis()
plt.title("Gaia DR3 YSO Color-Magnitude Diagram")
plt.show()
```

## Data source

Gaia Collaboration (2023), *Gaia Data Release 3: variability processing and analysis results.*
European Space Agency. Via ESA Gaia Archive — joined from `gaiadr3.vari_classifier_result`,
`gaiadr3.vari_summary`, and `gaiadr3.gaia_source`.

## Related datasets

- [Gaia DR3 Eclipsing Binaries](https://huggingface.co/datasets/juliensimon/gaia-dr3-eclipsing-binaries) — Gaia eclipsing binary candidates
- [Gaia DR3 Variable Star Summary](https://huggingface.co/datasets/juliensimon/gaia-dr3-variable-summary) — all Gaia variable star classifications

## Pipeline

Source code: [juliensimon/space-datasets](https://github.com/juliensimon/space-datasets)

## Support

If you find this dataset useful, please give it a ❤️ on the [dataset page](https://huggingface.co/datasets/juliensimon/gaia-dr3-young-stellar-objects) and share feedback in the Community tab! Also consider giving a ⭐️ to the [space-datasets](https://github.com/juliensimon/space-datasets) repo.

## Citation

```bibtex
@dataset{gaia_dr3_young_stellar_objects,
author = {Simon, Julien},
title = {Gaia DR3 Young Stellar Objects},
year = {2026},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/juliensimon/gaia-dr3-young-stellar-objects},
note = {Based on Gaia DR3 (ESA)}
}
```

## License

[CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)