R: Tuning the parameters of the alpha-SVM

Tuning the parameters of the alpha-SVM {CompositionalML}

R Documentation

Tuning the parameters of the `\alpha`-SVM

Description

Tuning the parameters of the \alpha-SVM.

Usage

alfasvm.tune(y, x, a = seq(-1, 1, by = 0.1), cost = seq(0.2, 2, by = 0.2), gamma = NULL,
ncores = 1, folds = NULL, nfolds = 10, stratified = TRUE, seed = NULL, graph = FALSE)

Arguments

`y`	The response variable, it can either be a factor (for classification) or a numeric vector (for regression). Depending on the nature of the response variable, the function will proceed with the necessary task.
`x`	A matrix with the compositional data.
`a`	A vector with a grid of values of the power transformation, it has to be between -1 and 1. If zero values are present it has to be greater than 0. If a=0, the isometric log-ratio transformation is applied.
`cost`	A grid of values for the cost of constraints violation. The cost is the "C"-constant of the regularization term in the Lagrange formulation.
`gamma`	A grid of values for the `\gamma` parameter of the Gaussian kernel. If no values are supplied the default grid is used, ten equidistant values from `1/D^2` to `\sqrt{D}`,
`ncores`	The number of cores to use. If more than 1, parallel computing will take place. It is advisable to use it if you have many observations and or many variables, otherwise it will slow down the process.
`folds`	If you have the list with the folds supply it here. You can also leave it NULL and it will create folds.
`nfolds`	The number of folds in the cross validation.
`stratified`	Do you want the folds to be created in a stratified way? TRUE or FALSE.
`seed`	You can specify your own seed number here or leave it NULL.
`graph`	If graph is TRUE (default value) a plot will appear.

Details

K-fold cross validation is performed to select the optimal parameters for the SVM and also estimate the rate of accuracy. For continuous responses the estimated performance translates to the MSE, while for categorical responses (factors) this is the accuracy (percentage of crrect classification).

Value

If graph is true, a graph with the estimated performance for each value of \alpha. A list including:

`per`	A vector with the estimated performance for each value of `\alpha`.
`performance`	A vector with the optimal performance and the optimal combinations of cost and `\gamma` values.
`best_a`	The value of `\alpha` corresponding to the optimal performance.
`runtime`	The time required by the cross-validation procedure.

Author(s)

Michail Tsagris.

R implementation and documentation: Michail Tsagris mtsagris@uoc.gr.

References

Chang Chih-Chung and Lin Chih-Jen: LIBSVM: a library for Support Vector Machines https://www.csie.ntu.edu.tw/~cjlin/libsvm/

Friedman Jerome, Trevor Hastie and Robert Tibshirani (2009). The elements of statistical learning, 2nd edition. Springer, Berlin.

Tsagris M.T., Preston S. and Wood A.T.A. (2011). A data-based power transformation for compositional data. In Proceedings of the 4th Compositional Data Analysis Workshop, Girona, Spain. https://arxiv.org/pdf/1106.1451.pdf

Examples

x <- as.matrix(iris[, 1:4])
x <- x/ rowSums(x)
y <- iris[, 5]
mod <- alfasvm.tune(y, x, a = c(0, 0.5, 1), cost = c(0.2, 0.4), gamma = c(0.1, 0.2) )
mod

[Package CompositionalML version 1.0 Index]

Tuning the parameters of the \alpha-SVM