train_model_glmnet: Training a glmnet model for binomial prediction.

In zktuong/kelvinny: Kelvinny The Incredible!

Description

Training a glmnet model for binomial prediction.

Usage

train_model_glmnet(
  data,
  variable_colname,
  alpha = 1,
  type.measure = "mse",
  type = "response",
  standardize = FALSE,
  cutOff = 0.5,
  nfolds = 10,
  ...
)

Arguments

data	expression data with genes in rows and samples in columns.
variable_colname	Column name for binary class.
alpha	Default = 1 for Lasso penalty. 0 = Ridge penalty. between 0 - 1 = Elastic net. See ?cv.glmnet.
type.measure	See ?cv.glmnet.
type	term for internal prediction type.
standardize	Default = FALSE. Expects data to be pre-scaled. if not, glmnet will do it for you.
cutOff	value for internal prediction.
nfolds	number of folds for cross validation. use ncol(data) for LOOCV.
...	pass to glmnet

Value

Generates a glmnet model.

Examples

model <- train_model_glmnet(data, variable_colname = "disease", alpha = 0.5, cutOff = 0.5, nfolds = ncol(data)) # LOOCV Elastic net regularization

zktuong/kelvinny documentation built on Oct. 22, 2020, 1:27 p.m.