Buckets:
| # Project rules | |
| - Keep detection and segmentation labels separate and use stable canonical ontology IDs. | |
| - Preserve ignore regions, partial-label semantics, source provenance, and license scope in loaders and training adapters. | |
| - Do not present DOTA internal splits as official benchmark splits or infer commercial permission from code licenses. | |
| - Do not train a model as part of dataset preparation or validation. | |
Xet Storage Details
- Size:
- 422 Bytes
- Xet hash:
- fce9e606f486730f0b3edac881d10ecae468c51f685c9c60ceccfc3ac2433deb
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.