← Directory
software · Global
Prithvi-EO-2.0-300M-BurnScars
Hugging Face model ibm-nasa-geospatial/Prithvi-EO-2.0-300M-BurnScars
Claim this listing
Represent Prithvi-EO-2.0-300M-BurnScars? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.
What's in it
- Data offered docs-verified
- burned area extent segmentation on HLS imagery (classes: no burn, burned, no data/clouds) · six-band HLS input: Blue, Green, Red, Narrow NIR, SWIR, SWIR 2
- Coverage docs-verified
- continental US (fine-tuning dataset)
- Formats llm-inferred
- GeoTIFF
- Access docs-verified
- Open; Apache-2.0 licence
Useful for
- Use cases llm-inferred
- Post-fire burn scar extent mapping from HLS imagery for damage assessment · Burnt-area delineation as an input to post-fire runoff and erosion risk screening · Burn extent baseline for ecological impact assessment (US-trained; UK applicability unverified)
- Audience llm-inferred
- GIS analysts · environmental consultants · ecologists · hydrologists · remote sensing specialists
Limitations
- Constraints llm-inferred
- Trained and evaluated on continental US chips only; UK performance not established · Benchmark focuses on single-timestamp segmentation · Not deployed by any Inference Provider; self-hosting required · Fine-tuning requires TerraTorch and the supplied config
- Licence
- apache-2.0
- Authentication
- No API key required (repository README, auto-extracted; check upstream requirements)
- GitHub stars
- Unknown
- Last commit
- Unknown
- Last successful page check
- 14/10/2026
- Page health
- healthy
- Package
- Not recorded
- Verification
- Not independently verified
Provenance & evidence
Verified 2026-10-15 via HF API: exists; 4049 downloads/30d, 5 likes.
Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (Hugging Face models)