lasvmTrainWrapper

Share:

Description

Use lasvm to train a given problem.

Usage

1
2
3
4
lasvmTrainWrapper(x, y, gamma, cost, degree = 3, coef0 = 0L,
  optimizer = 1L, kernel = 2L, selection = 0L, termination = 0L,
  sample = 0, cachesize = 256L, bias = 1L, epochs = 1L,
  epsilon = 0.001, verbose = FALSE)

Arguments

x

data matrix

y

training labels

gamma

RBF kernel bandwidth

cost

regularization constant

degree

degree for poly kernel

coef0

coefficient for poly kernel

optimizer

type of optimizer

kernel

kernel type

selection

selection strategy

termination

criterion for stopping

sample

parameter for stopping criterion, e.g. seconds

cachesize

size of kernel cache

bias

use bias?

epochs

number of epochs

epsilon

stopping criterion parameter

verbose

verbose output?

Value

a list consisting of SV matrix of support vectors alpha vector of alpha coefficients bias bias term