Indexed Averager#

Summary#

Compute weighted group means from a precomputed integer index map.

Metadata#

Import path: modacor.modules.base_modules.indexed_averager.IndexedAverager
Source: src/modacor/modules/base_modules/indexed_averager.py
Module ID: IndexedAverager
Module version: 20261001.1
Keywords: indexed, grouped, weighted average, reduction

Required data keys#

None

Modifies#

configured value: signal, uncertainties, axes
configured axis: signal, uncertainties
bin_id: signal
bin_count: signal
positive_weight_count: signal
sum_weights: signal
effective_sample_size: signal

Required arguments#

None

Default configuration#

{
  "axis_key": "Q",
  "bin_id_key": "bin_id",
  "index_key": "bin_index",
  "mask_key": "Mask",
  "output_processing_key": null,
  "stats_keys": null,
  "uncertainty_weight_key": null,
  "use_value_uncertainty_weights": false,
  "use_value_weights": true,
  "value_key": "signal",
  "with_processing_keys": null
}

Argument specification#

Argument	Type	Required	Default	Dependency role	Description
`axis_key`	str or NoneType	No	Q	-	Optional BaseData to average over the same points and attach as the output axis.
`bin_id_key`	str	No	bin_id	-	Output BaseData key recording the original populated bin IDs.
`index_key`	str	No	bin_index	-	BaseData containing precomputed integer group indices.
`mask_key`	str or NoneType	No	Mask	-	Optional mask BaseData; true values are excluded when the key is present.
`output_processing_key`	str or NoneType	No	-	-	Optional distinct output bundle; requires exactly one input bundle.
`stats_keys`	list or str or NoneType	No	-	-	Value and/or axis keys receiving scatter-derived SEM and STD; None selects both.
`uncertainty_weight_key`	str or NoneType	No	-	-	Value uncertainty component used for inverse-variance weighting.
`use_value_uncertainty_weights`	bool	No	False	-	Also weight by inverse variance from one named value uncertainty.
`use_value_weights`	bool	No	True	-	Use the value BaseData weights for both the value and output axis.
`value_key`	str	No	signal	-	BaseData value to average.
`with_processing_keys`	str or list or NoneType	No	-	-	ProcessingData key(s) to reduce.

Notes#

The index map alone determines membership. The optional axis is averaged over the same valid, unmasked observations and weights; bin edges are neither required nor read.