gmum.r (version 0.2.1)

plot.Rcpp_SVMClient: Plot SVM object

Description

Plots trained svm data and models disciriminative

Usage

"plot"(x, X = NULL, mode = "normal", cols = c(1, 2), radius = 3, radius.max = 10, ...)

Arguments

x

Trained SVM object

X

Optional new data points to be predicted and plotted in one of the following formats: data.frame, data.matrix; default: NULL

mode

Which plotting mode to use as string, available are:

• 'normal' - default mode, plots data in cols argument and a linear decision boundry in available
• 'pca' - preforms PCA decomposition and draws data in a subspace of first 2 dimensions from the PCA
• 'contour' - countour plot for non-linear kernels

cols

Data dimensions to be plotted as vector of length 2, default: c(1,2)

radius

Radius of the plotted data points as float, default: 3

radius.max

Maximum radius of data points can be plotted, when model is trained with example weights as float, default: 10

...

other arguments not used by this method.

Examples

## Not run: 
# # here we ause svm is a trained SVM model
# plot(svm)
# plot(svm, X=x, cols=c(1,3))
# plot(svm, mode="pca", radius=5)
# ## End(Not run)