amrElm: AMR-ELM.

In rladeira/amrElm: Affinity Matrix Regularized Extreme Learning Machine

Description

Tests the AMR-ELM model.

Usage

## S3 method for class 'amrElm'
predict(model, newdata)

Arguments

model	the model created by amrElm-train, which contains: Z: hidden layer weights H: hidden layer output weights: output layer weights affinity: the affinity used to generate the model (e.g.: cosine affinity) dataTrain: training data for generating affinity matrix. TODO: try to remove X from the parameter list, replacing it by the affinity matrix computed for the training patterns
newdata	the test data, numerical - numerical with zero mean and unit variance and patterns in the lines, attributes in the columns

Value

outupt for the test patterns

Examples

## Not run: testOutput <- predict(model, newdata)

rladeira/amrElm documentation built on May 27, 2019, 9:17 a.m.