compute {neuralnet}R Documentation

Computation of a given neural network for a new covariate vector

Description

compute, a method for objects of class nn, typically produced by neuralnet. Computes the outputs of all neurons for a specific arbitrary covariate vector given a trained neural network. Please make sure that the order of the covariates is the same in the new matrix or dataframe as in the original neural network.

Usage

compute(x, covariate, rep = 1)

Arguments

x an object of class nn.
covariate a data.frame or matrix containing the variables to calculate the output of the neural network.
rep an integer indicating the neural network's repetition which should be used.

Value

compute returns a list containing the following components:

neurons a list of the neuron's output for each layer of the neural network.
net.result a matrix containing the overall result of the neural network.

Author(s)

Stefan Fritsch fritsch@bips.uni-bremen.de

Examples

Var1 <- runif(50, 0, 100) 
sqrt.data <- data.frame(Var1, Sqrt=sqrt(Var1))
print(net.sqrt <- neuralnet( Sqrt~Var1,  sqrt.data, hidden=10, threshold=0.01))
compute(net.sqrt, (1:10)^2)$net.result

[Package neuralnet version 1.2 Index]