# ExoMiner++ TESS 2-Minute Threshold-Crossing Events (TCEs), Sectors 1-67: Vetting Features and Labels for 204,729 Transit Signals

## Facts

- Provider: Miguel Martinho and Hamed Valizadegan (NASA Ames, USRA); Hugging Face conversion by 29thDay
- License: CC-BY-4.0 (https://zenodo.org/records/15466293)
- Size: 37 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/29thDay/exominer-plusplus
- Access: public, readable without an account
- DOI: https://doi.org/10.5281/zenodo.15466293
- Category: Astronomy & Space
- Source: https://doi.org/10.5281/zenodo.15466293
- Page: https://scigantic.com/archive/c7bf3158-1e87-5158-b2ef-02b5dae22a96
- Keywords: Astronomy & Space, Parquet, Hugging Face

## About

The ExoMiner++ training and vetting data for transit-like signals, threshold-crossing events (TCEs), that the TESS SPOC pipeline found in 2-minute light curves of Sectors 1 to 67, released by Martinho and Valizadegan (NASA Ames / USRA) on Zenodo (DOI 10.5281/zenodo.15466293) and converted here from TFRecords to Arrow/Parquet for Hugging Face datasets with feature values unchanged. Each TCE is joined to its catalog row and split into labeled (57,162 TCEs) and unlabeled (147,567 TCEs) sets. Labels are known planet (KP), confirmed planet (CP), brown dwarf (BD), eclipsing binary (EB), false positive (FP), non-transiting phenomenon (NTP), nearby eclipsing binary (NEB) and nearby planet candidate (NPC), or UNK for unlabeled. Columns include TCE and TIC identifiers, the paper's 10-fold cross-validation fold, the ExoMiner++ score, scalar diagnostics, phase-folded and binned flux views (31, 301 or 674 bins), an unfolded flux view (20 x 31) and difference images (5 x 33 x 33) for centroid checks. Uses: training and benchmarking exoplanet vetting classifiers that separate planets from eclipsing binaries and instrumental false alarms. The original is CC BY 4.0; cite Martinho and Valizadegan 2025, AJ (DOI 10.3847/1538-3881/ae03a4).

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("29thDay/exominer-plusplus")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `data/labeled-00000-of-00010.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/c7bf3158-1e87-5158-b2ef-02b5dae22a96
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/c7bf3158-1e87-5158-b2ef-02b5dae22a96
- Over MCP: `get_schema_card` with id `c7bf3158-1e87-5158-b2ef-02b5dae22a96` 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)
