← Directory
software · Global
TerraMind-1.0-Tokenizer-NDVI
Hugging Face model ibm-esa-geospatial/TerraMind-1.0-Tokenizer-NDVI
Claim this listing
Represent TerraMind-1.0-Tokenizer-NDVI? 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
- NDVI map encoding to discrete tokens (FSQ, 5 dimensions, codebook of 15,360) · NDVI reconstruction from tokens via diffusion decoding
- Coverage llm-inferred
- Global
- Formats llm-inferred
- Python tensor (TerraTorch API)
- Access docs-verified
- Free open model under Apache-2.0 licence; loaded via TerraTorch with pretrained=True; no API key or auth stated in DOCS
Useful for
- Use cases llm-inferred
- Vegetation-condition baseline inputs for TerraMind-based habitat mapping (BNG) · Research prototyping of NDVI tokenisation for generative land-cover or vegetation change analysis on site investigations · Tokenised NDVI features for downstream ML models on vegetation cover around drainage or flood-risk sites
- Audience llm-inferred
- GIS analysts · remote sensing / EO specialists · ecologists · ML engineers
Limitations
- Constraints llm-inferred
- Accepts NDVI maps only, not other spectral bands · Diffusion decoding can add detail that is hallucinated, and more timesteps increase this risk · Not deployed by any Inference Provider per DOCS; requires local Python setup with TerraTorch · Spatial and temporal resolution and the coverage of the training data are not documented in DOCS
- 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; 809 downloads/30d, 4 likes.
Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (Hugging Face models)