aboutsummaryrefslogtreecommitdiff
path: root/man/gensvm.refit.Rd
diff options
context:
space:
mode:
authorGertjan van den Burg <gertjanvandenburg@gmail.com>2018-04-04 15:08:12 -0400
committerGertjan van den Burg <gertjanvandenburg@gmail.com>2018-04-04 15:08:12 -0400
commit459ce96fa8a0072d3533bc2dc1566cc1b797401b (patch)
tree229a5d9b137f7fcf3b5112e4a189e972d6dafa26 /man/gensvm.refit.Rd
parentEnsure classes isn't a factor (diff)
downloadrgensvm-459ce96fa8a0072d3533bc2dc1566cc1b797401b.tar.gz
rgensvm-459ce96fa8a0072d3533bc2dc1566cc1b797401b.zip
Documentation improvements
Diffstat (limited to 'man/gensvm.refit.Rd')
-rw-r--r--man/gensvm.refit.Rd51
1 files changed, 47 insertions, 4 deletions
diff --git a/man/gensvm.refit.Rd b/man/gensvm.refit.Rd
index 8a711bc..cae0646 100644
--- a/man/gensvm.refit.Rd
+++ b/man/gensvm.refit.Rd
@@ -4,7 +4,7 @@
\alias{gensvm.refit}
\title{Train an already fitted model on new data}
\usage{
-gensvm.refit(fit, X, y, p = NULL, lambda = NULL, kappa = NULL,
+gensvm.refit(fit, x, y, p = NULL, lambda = NULL, kappa = NULL,
epsilon = NULL, weights = NULL, kernel = NULL, gamma = NULL,
coef = NULL, degree = NULL, kernel.eigen.cutoff = NULL,
max.iter = NULL, verbose = NULL, random.seed = NULL)
@@ -12,16 +12,59 @@ gensvm.refit(fit, X, y, p = NULL, lambda = NULL, kappa = NULL,
\arguments{
\item{fit}{Fitted \code{gensvm} object}
-\item{X}{Data matrix of the new data}
+\item{x}{Data matrix of the new data}
\item{y}{Label vector of the new data}
-\item{verbose}{Turn on verbose output and fit progress. If NULL (the
-default) the value from the fitted model is chosen.}
+\item{p}{if NULL use the value from \code{fit} in the new model, otherwise
+override with this value.}
+
+\item{lambda}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{kappa}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{epsilon}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{weights}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{kernel}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{gamma}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{coef}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{degree}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{kernel.eigen.cutoff}{if NULL use the value from \code{fit} in the new
+model, otherwise override with this value.}
+
+\item{max.iter}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{verbose}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
+
+\item{random.seed}{if NULL use the value from \code{fit} in the new model,
+otherwise override with this value.}
}
\value{
a new fitted \code{gensvm} model
}
+\description{
+This function can be used to train an existing model on new
+data or fit an existing model with slightly different parameters. It is
+useful for retraining without having to copy all the parameters over. One
+common application for this is to refit the best model found by a grid
+search, as illustrated in the examples.
+}
\examples{
x <- iris[, -5]
y <- iris[, 5]