| ---
|
| license: apache-2.0
|
| task_categories:
|
| - image-text-to-text
|
| language:
|
| - en
|
| tags:
|
| - satellite
|
| - land-cover
|
| - lfm-vl
|
| - geospatial
|
| pretty_name: NU-TONIC raw SFT init (satellite imagery + land-cover labels)
|
| ---
|
|
|
| # NU-TONIC raw SFT init
|
|
|
| **Satellite imagery** and aligned **land-cover** outputs packaged as **image–text** rows for fine-tuning LFM-VL (leap-finetune `vlm_sft` format). JSONL user prompts name the modality (satellite imagery vs. overhead context) where it matters.
|
|
|
| ## Provenance
|
|
|
| - **Locations:** GeoGuessr-style POIs (default HF source: `stochastic/random_streetview_images_pano_v0.0.2`) via `download_geoguessr_poi_imagery.py`.
|
| - **Optical:** multispectral optical COGs from a public STAC catalog, **blue/green/red** or **visual** preview, percentile-stretched to uint8.
|
| - **Labels:** per-pixel land-cover raster from Earth Engine, **reprojected** to the same 10 m grid as the RGB stack, then tiled and **nearest-neighbor** downsampled with the mask.
|
| - **Context:** optional geographic overhead still per POI (token-based static map API) under `mapbox_stills/`.
|
| - **Rows:** global caption + grounding per tile of satellite imagery, plus **per–land-cover-class** caption (and optional grounding) when a class exceeds `--min-class-fraction`.
|
| - **Downstream format:** `messages[]` with `image` + text, aligned with `refs/satellite-vlm` / VRSBench converter conventions.
|
|
|
| ## Layout
|
|
|
| - `images/*.png` — RGB chips (downsampled, e.g. 224×224).
|
| - `overlays/*.png` — optional bbox visualization per JSONL row (``--write-bbox-overlays``).
|
| - `mapbox_stills/*.png` — overhead context still per POI (when API token set).
|
| - `data/train.jsonl`, `data/validation.jsonl`, `data/test.jsonl` — splits hashed by `poi_id`.
|
| - `metadata/*.json` — per-tile sidecars (coordinates, scene id, `land_cover_label_meta` counts/shapes only).
|
|
|
| ## Build
|
|
|
| See `docs/scripts/SPEC-lfm-vl-sft-dataset.md` and `python data/scripts/build_lfm_vl_sft_dataset.py --help`.
|
|
|
| ## Licenses
|
|
|
| Respect the **STAC** data provider terms, **Earth Engine** terms for label rasters, the **static map** provider ToS for optional stills, and the **MIT** HF pano dataset license for source coordinates.
|
|
|