Simweights

Processor that computes per-event weights using the simweights library. The computed weights are written to a column and the weight group is recorded in the dataset attributes.

Note

Simweights are computed per-file with nfiles=1. On load, the decoder applies a normalization to each weight:

weight /= # files with same weight_group designation

This yields a result identical to running simweights with correct nfiles. Weights are therefore tied to which files are loaded, not to how big the batch was when the file was processed.

Note

Simweights requires specific tables to be available in the envelope under ‘data’. For example, weighting CORSIKA events requires access to ‘CorsikaWeightMap’ and ‘PolyplopiaPrimary’.

Configuration

Selected as name: simweights.

Option	Description	Type	Default
`flux`	Flux-model plugin selection (`name` / `kwargs`). See the flux models below.	mapping	required
`weighter`	Name of the `simweights` weighter matching the simulation type.	str	required
`out`	Column the computed weights are written to.	str	required
`weight_group`	Label recorded in the attributes identifying this weight set.	str	required

- name: simweights
  kwargs:
    flux: { name: gaisser-h4a, kwargs: {} }
    weighter: CorsikaWeighter
    out: weight
    weight_group: corsika

Flux models

The flux model is itself a selectable plugin; the choices are documented under the flux model slot.