selectElasticNetParams: Select the optimal elastic net parameters based on...
In CBIIT/rcellminerElasticNet: rcellminerElasticNet

Select the optimal elastic net parameters based on cross-validation error estimates.

selectElasticNetParams(
  featureMat,
  responseVec,
  standardize = TRUE,
  standardizeY = FALSE,
  fitIntercept = TRUE,
  alphaVals = seq(0.2, 1, length = 9),
  lambdaVals = NULL,
  nFolds = 10,
  nRepeats = 10,
  useLambdaMin = TRUE,
  verbose = TRUE
)

`featureMat`	p x n matrix with input feature vectors along rows.
`responseVec`	n-dimensional response vector to be predicted using a sparse linear combination of input feature vectors specified in featureMat.
`standardize`	Logical flag for feature variable standardization (across observations), passed in to glmnet functions (glmnet documentation: if variables are already in the same units, standardization may not be necessary).
`standardizeY`	Logical flag for response variable standardization across observations.
`fitIntercept`	Logical flag indicating whether intercept term should be fit.
`alphaVals`	a vector of alpha values to be optimized over
`lambdaVals`	a vector of lambda values to be optimized over
`nFolds`	The number of folds used in cross-validation.
`nRepeats`	The number of cross-validation repeats (with different fold assignments).
`useLambdaMin`	a boolean, whether to use the minimum lambda (lambda.min) from cross-validation as the optimum lambda, if FALSE then the largest value of lambda such that error is within 1 standard error of the minimum (lambda.1se) is used.

A list with the following elements:

alpha The optimal alpha parameter.
lambda The optimal lambda parameter.
minCvError The minimum average cross validation error.
cvInfoTab A data frame summarizing minimum cross validation errors and lambda parameter selections for each possible choice of the alpha parameter.
cvFoldIds A matrix indicating the fold ids for each cross validation repeat along the columns.