ART_Update_Weights: ART_Update_Weights

Description Usage Arguments Details Value

View source: R/ART_Update_Weights.r

Description

Updates the weight matrix of an ART network.

Usage

1
ART_Update_Weights(input, weight, categoryNumber, learningRate)

Arguments

input

Vector of size NumFeatures that contains the input signal into the network.

weight

Matrix of size NumFeatures-by-NumCategories which holds the weights of the network.

learningRate

Rate at which the network should learn new inputs.

categoryNumer

Number of the category that codes the current input.

Details

This function returns a new weight matrix that has "learned" the input in the given category, as well as a value correspoding to whether or not the weight matrix was changed (0 = no change; 1 = change).

The length of the INPUT vector must equal the number of rows in the WEIGHT matrix, the CATEGORYNUMBER must be in the range [1, NumCategories], and the LEARNINGRATE must be in the range [0, 1].

Value

List containing the UPDATEDWEIGHT and WEIGHTCHANGE. The UPDATEDWEIGHT is a matrix of size NumFeatures-by-NumCategories that holds the new weights of the network after the input has been successfully learned. The WEIGHTCHANGE is a value (0 or 1) which relays whether or not the weight matrix was changed during this updating. Here, 0 represents no change and 1 represents a change.


gbaquer3/fuzzyARTMAP documentation built on Aug. 2, 2020, 6:33 p.m.