OmicsMarkeR (version 1.4.2)

denovo.grid: Denovo Grid Generation

Description

Greates grid for optimizing selected models

Usage

denovo.grid(data, method, res)

Arguments

data

data of method to be tuned

method

vector indicating the models to generate grids. Available options are "plsda" (Partial Least Squares Discriminant Analysis), "rf" (Random Forest), "gbm" (Gradient Boosting Machine), "svm" (Support Vector Machines), "glmnet" (Elastic-net Generalized Linear Model), and "pam" (Prediction Analysis of Microarrays)

res

Resolution of model optimization grid.

Value

A list containing dataframes of all combinations of parameters for each model:

See Also

"expand.grid" for generating grids of specific parameters desired. However, NOTE that you must still convert the generated grid to a list.

Examples

# random test data
set.seed(123)
dat.discr <- create.discr.matrix(
    create.corr.matrix(
        create.random.matrix(nvar = 50, 
                             nsamp = 100, 
                             st.dev = 1, 
                             perturb = 0.2)),
    D = 10
)

df <- data.frame(dat.discr$discr.mat, .classes = dat.discr$classes)

# create tuning grid
denovo.grid(df, "gbm", 3)