gmum.r (version 0.2.1)

GNG: Constructor of GrowingNeuralGas object.

Description

Construct GNG object. Can be used to train offline, or online.

Usage

GNG(x = NULL, labels = c(), beta = 0.99, alpha = 0.5, max.nodes = 1000, eps.n = 6e-04, eps.w = 0.05, max.edge.age = 200, train.online = FALSE, max.iter = 200, dim = -1, min.improvement = 0.001, lambda = 200, verbosity = 0, seed = -1, k = NULL)

Arguments

x

Passed data (matrix of data.frame) for offline training

labels

Every example can be associated with labels that are added to nodes later. By default empty

beta

Decrease the error variables of all node nodes by this fraction (forgetting rate). Default 0.99

alpha

Decrease the error variables of the nodes neighboring to the newly inserted node by this fraction. Default 0.5

max.nodes

Maximum number of nodes (after reaching this size it will continue running, but won't add new nodes)

eps.n

How strongly adapt neighbour node. Default 0.0006

eps.w

How strongly adapt winning node. Default 0.05

max.edge.age

Maximum edge age. Decrease to increase speed of change of graph topology. Default 200

train.online

default FALSE. If used will run in online fashion

max.iter

Uf training offline will stop if exceedes max.iter iterations. Default 200

dim

Used for training online, specifies training example size

min.improvement

Used for offline (default) training. Controls stopping criterion, decrease if training stops too early. Default 1e-3

lambda

Every lambda iteration is added new vertex. Default 200

verbosity

How verbose should the process be, as integer from $[0,6]$, default: 0

seed

Seed for internal randomization

k

Utility constant, by default turned off. Good value is 1.3. Constant controlling speed of erasing obsolete nodes, see http://sund.de/netze/applets/gng/full/tex/DemoGNG/node20.html

Examples

## Not run: 
# X <- gng.preset.sphere(100)
# y <- round(runif(100))
# # Train in an offline manner
# gng <- GNG(X, labels=y, max.nodes=20)
# # Plot
# plot(gng)
# 
# # Train in an online manner with utility (erasing obsolete nodes)
# gng <- GNG(max.nodes=20, train.online=TRUE, k=1.3, dim=3)
# insertExamples(gng, X, labels=y)
# run(gng)
# Sys.sleep(10)
# terminate(gng)
# # Plot
# plot(gng)
# ## End(Not run)