rvm-class function

Class "rvm"

Relevance Vector Machine Class class

Objects from the Class

Objects can be created by calls of the form new("rvm", ...). or by calling the rvm function.

Slots

• tol:: Object of class "numeric" contains tolerance of termination criteria used.
• kernelf:: Object of class "kfunction" contains the kernel function used
• kpar:: Object of class "list" contains the hyperparameter used
• kcall:: Object of class "call" contains the function call
• type:: Object of class "character" contains type of problem
• terms:: Object of class "ANY" containing the terms representation of the symbolic model used (when using a formula interface)
• xmatrix:: Object of class "matrix" contains the data matrix used during computation
• ymatrix:: Object of class "output" contains the response matrix
• fitted:: Object of class "output" with the fitted values, (predict on training set).
• lev:: Object of class "vector" contains the levels of the response (in classification)
• nclass:: Object of class "numeric" contains the number of classes (in classification)
• alpha:: Object of class "listI" containing the the resulting alpha vector
• coef:: Object of class "ANY" containing the the resulting model parameters
• nvar:: Object of class "numeric" containing the calculated variance (in case of regression)
• mlike:: Object of class "numeric" containing the computed maximum likelihood
• RVindex:: Object of class "vector" containing the indexes of the resulting relevance vectors
• nRV:: Object of class "numeric" containing the number of relevance vectors
• cross:: Object of class "numeric" containing the resulting cross validation error
• error:: Object of class "numeric" containing the training error
• n.action:: Object of class "ANY" containing the action performed on NA

Methods

• RVindex: signature(object = "rvm"): returns the index of the relevance vectors
• alpha: signature(object = "rvm"): returns the resulting alpha vector
• cross: signature(object = "rvm"): returns the resulting cross validation error
• error: signature(object = "rvm"): returns the training error
• fitted: signature(object = "vm"): returns the fitted values
• kcall: signature(object = "rvm"): returns the function call
• kernelf: signature(object = "rvm"): returns the used kernel function
• kpar: signature(object = "rvm"): returns the parameters of the kernel function
• lev: signature(object = "rvm"): returns the levels of the response (in classification)
• mlike: signature(object = "rvm"): returns the estimated maximum likelihood
• nvar: signature(object = "rvm"): returns the calculated variance (in regression)
• type: signature(object = "rvm"): returns the type of problem
• xmatrix: signature(object = "rvm"): returns the data matrix used during computation
• ymatrix: signature(object = "rvm"): returns the used response

Author(s)

Alexandros Karatzoglou

alexandros.karatzoglou@ci.tuwien.ac.at

See Also

rvm, ksvm-class

Examples

# create data
x <- seq(-20,20,0.1)
y <- sin(x)/x + rnorm(401,sd=0.05)

# train relevance vector machine
foo <- rvm(x, y)
foo

alpha(foo)
RVindex(foo)
fitted(foo)
kernelf(foo)
nvar(foo)

## show slots
slotNames(foo)