fit: fit

In jroberayalas/ahnr: An Implementation of the Artificial Hydrocarbon Networks

Description

Function to train an Artificial Hydrocarbon Network (AHN).

Usage

fit(Sigma, n, eta, maxIter = 2000)

Arguments

Sigma	a list with two data frames. One for the inputs X, and one for the outputs Y.
n	number of particles to use.
eta	learning rate of the algorithm. Default is 0.01.
maxIter	maximum number of iterations.

Value

an object of class "ahn" with the following components:

• network: structure of the AHN trained. It contains the hydrogen coefficient H, position of the molecules Pi, data cluster ID per molecule moleculeNumber, number of molecules n, and carbon structure C

• Yo: original output variable.

• Ym: predicted output variable.

• eta: learning rate.

• minOverallError: minimum error achieved.

• variableNames: names of the input variables.

Examples

# Create data
x <- 2 * runif(1000) - 1;
x <- sort(x)

y <- (x < 0.1) * (0.05 * runif(100) + atan(pi*x)) +
    (x >= 0.1 & x < 0.6) * (0.05 * runif(1000) + sin(pi*x)) +
    (x >= 0.6) * (0.05 * runif(1000) + cos(pi*x))

# Create Sigma list
Sigma <- list(X = data.frame(x = x), Y = data.frame(y = y))

# Train AHN
ahn <- fit(Sigma, 5, 0.01, 500)

jroberayalas/ahnr documentation built on May 20, 2019, 2:06 a.m.