dataset · Global

MINDSET

Hugging Face dataset taylor-geospatial/MINDSET

Claim this listing

Represent MINDSET? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.

Data offered llm-inferred: point-level embedding vectors ("aef" list column) keyed by point_id and year, with point geometry and bbox
Coverage llm-inferred: Global (per ENTRY regions; the visible sample rows all fall near -54.8 lat, -68.3 lon)
Formats docs-verified: parquet
Access llm-inferred: Open download under CC-BY-4.0 licence (attribution required); no authentication listed in entry

Use cases llm-inferred: Use the embedding vectors as point features in site-suitability or ground-condition classifiers where no local gridded layer exists · Find environmentally similar analogue sites by nearest-neighbour search on the embedding vectors · Join embeddings to project point datasets (boreholes, monitoring points) by point_id and year for model training
Audience llm-inferred: GIS analysts · hydrologists · ecologists · data scientists

Constraints llm-inferred: Docs text is only a dataset-viewer preview; no dataset card description of the source model, coverage, resolution or update cycle · UK coverage not verified; sampled rows are in South America · Embedding dimensions are not interpretable as named variables · About 121M rows (100M-1B size bucket) needs Polars, Dask or similar for practical use · Not an MCP server or tool, so no tool access

Verified 2026-10-15 via HF API: exists; 327 downloads/30d, 1 likes.

Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (Hugging Face datasets)