intake loader¶

Opens a dataset from an intake-forecast catalog. Requires pip install "gridstats[extra]".

After loading, preprocessing (variable renaming, sel/isel) is delegated to the same logic as the xarray loader.

Fields¶

Field	Description
`catalog`	Path or URL to the intake catalog file. Required.
`dataset_id`	Entry name within the catalog. Required.
`mapping`	Dict of `{source_name: target_name}` variable renames.
`sel`	Label-based selection (scalars, lists, or `{start, stop}` dicts).
`isel`	Integer-index selection.
`chunks`	Dask chunk sizes for lazy loading.

Example¶

source:
  type: intake
  catalog: /catalogs/oceanum.yaml
  dataset_id: wave_global_era5
  chunks:
    time: 50
  mapping:
    significant_wave_height: hs
  sel:
    time: {start: "2000-01-01", stop: "2020-12-31"}