Skip to content

Latest commit

 

History

History
31 lines (23 loc) · 1.63 KB

File metadata and controls

31 lines (23 loc) · 1.63 KB

Compositors

This document lists compositors available for raster layers in rslearn. Set compositing_method in a layer config to either:

  • a built-in enum name (for example "FIRST_VALID"), or
  • a custom compositor config (class_path + init_args).

See Dataset Configuration for where this field appears.

Built-In Compositors

Compositing Method Notes
FIRST_VALID Default. First non-nodata pixel in item order.
MEAN Per-pixel mean across valid pixels.
MEDIAN Per-pixel median across valid pixels.
SPATIAL_MOSAIC_TEMPORAL_STACK Spatial first-valid per timestep, stacked along T.
TEMPORAL_MEAN Temporal reduction over stacked timesteps.
TEMPORAL_MAX Temporal reduction over stacked timesteps.
TEMPORAL_MIN Temporal reduction over stacked timesteps.

Custom Cloud-Aware Ranking Compositors

These compositors reorder items inside each materialized item group, then apply FIRST_VALID in that ranked order.

Class Path Description
rslearn.dataset.omni_cloud_mask.OmniCloudMaskFirstValid Uses OmniCloudMask model inference on R/G/NIR.
rslearn.dataset.sentinel2_scl.Sentinel2SCLFirstValid Uses Sentinel-2 SCL classes to score cloudiness.