Synopsis
Generalized Hebbian Algorithm (GHA). Performs an iterative principal components analysis.
Description
Generalized Hebbian Algorithm (GHA) is an iterative method to compute principal components. From a computational point of view, it can be advantageous to solve the eigenvalue problem by iterative methods which do not need to compute the covariance matrix directly. This is useful when the ExampleSet contains many Attributes (hundreds, thousands). The operator outputs a GHAModel
. With the ModelApplier
you can transform the features.
Input
- example set input: expects: ExampleSet
Output
- example set output:
- original:
- preprocessing model:
Parameters
- number of components: Number of components to compute. If '-1' nr of attributes is taken.'
- number of iterations: Number of Iterations to apply the update rule.
- learning rate: The learning rate for GHA (small)
- use local random seed: Indicates if a local random seed should be used.
- local random seed: Specifies the local random seed