infrared.infrared.Feature Class Reference

Feature in multi-dimensional Boltzmann sampling. More...

Public Member Functions
	__init__ (self, identifier, eval_function, group=[])
	Construct feature.
	eval (self, sample)
	Evaluate feature for given sample and weight.

Protected Attributes
	_identifier
	_eval_function
	_weight
	_group

Properties
	identifier = property
	group = property
	weight = property

Detailed Description

Feature in multi-dimensional Boltzmann sampling.

A feature defines a (partial) evaluation/score of a sample, which can be targeted due to a dedicated weight. It defines a method value, which determines the feature's value for a sample.

Moreover, a feature controls one or several groups of functions.

Features should belong to exactly one Model. Features have weights, but their weights have to be synchronized with their functions. The weight must be changed only by the model in sync with the corresponding functions (set_feature_weight). For this reason, weight is defined as a read-only property.

Constructor & Destructor Documentation

__init__()

infrared.infrared.Feature.__init__	(		self,
			identifier,
			eval_function,
			group = []
	)

Construct feature.

       ifentifier:
       eval_function:

Parameters

group

defaults to empty list []

Member Function Documentation

eval()

infrared.infrared.Feature.eval	(		self,
			sample
	)

Evaluate feature for given sample and weight.

Member Data Documentation

_eval_function

infrared.infrared.Feature._eval_function

protected

_group

infrared.infrared.Feature._group

protected

_identifier

infrared.infrared.Feature._identifier

protected

_weight

infrared.infrared.Feature._weight

protected

Property Documentation

group

infrared.infrared.Feature.group = property

static

identifier

infrared.infrared.Feature.identifier = property

static

weight

infrared.infrared.Feature.weight = property

static

---

The documentation for this class was generated from the following file:

• src/infrared/infrared.py