Optimize Selection (Weight-Guided)
(AI Studio Core)
Synopsis
Adds iteratively features according to input attribute weightsDescription
This operator uses input attribute weights to determine the order of features added to the feature set starting with the feature set containing only the feature with highest weight. The inner operators must provide a performance vector to determine the fitness of the current feature set, e.g. a cross validation of a learning scheme for a wrapper evaluation. Stops if adding the last k features does not increase the performance or if all features were added. The value of k can be set with the parameter generations_without_improval .
Input
example set (IOObject)This is an example set input port
- attribute weights in (IOObject)
- through (IOObject)
through input port, that leaves the content untouched.
Output
- example set (Data Table)
This is an example set output port
weights (Attribute Weights)
performance (Performance Vector)
Parameters
- use early stoppingEnables early stopping. If unchecked, always the maximum number of generations is performed.
- generations without improvalStop criterion: Stop after n generations without improval of the performance.
- use absolute weightsIndicates that the absolute values of the input weights should be used to determine the feature adding order.
- normalize weightsIndicates if the final weights should be normalized.
- use local random seedIndicates if a local random seed should be used.
- local random seedSpecifies the local random seed
- user result individual selectionDetermines if the user wants to select the final result individual from the last population.
- show population plotterDetermines if the current population should be displayed in performance space.
- plot generationsUpdate the population plotter in these generations.
- constraint draw rangeDetermines if the draw range of the population plotter should be constrained between 0 and 1.
- draw dominated pointsDetermines if only points which are not Pareto dominated should be painted.
- population criteria data fileThe path to the file in which the criteria data of the final population should be saved.
- maximal fitnessThe optimization will stop if the fitness reaches the defined maximum.