Forward and backwards stepwise search for a set of features and components
with tracking of nonlinear parameters.
buildLKTModel(
data,
usefolds = NA,
allcomponents,
allfeatures,
currentcomponents = c(),
specialcomponents = c(),
specialfeatures = c(),
forv,
bacv,
preset = NA,
presetint = T,
currentfeatures = c(),
verbose = FALSE,
currentfixedpars = c(),
maxitv = 10,
interc = FALSE,
forward = TRUE,
backward = TRUE,
metric = "BIC",
removefeat = c(),
removecomp = c()
)list of values "tracetable" and "currentfit"
is a dataset with Anon.Student.Id and CF..ansbin.
Numeric Vector | Specifies the folds for model fitting in LKT; the features are still calculated across all folds to compute test fold fit externally
is search space for LKT components
is search space for LKT features
components to start search from
add special components (not crossed with features, only paired with special features 1 for 1)
features for each special component (not crossed during search)
the minimuum amount of improvement needed for the addition of a new term
the maximuum amount of loss for a term to be removed
One of "static","AFM","PFA","advanced","AFMLLTM","PFALLTM","advancedLLTM"
should the intercepts be included for preset components
features to start search from
passed to LKT
used for current features as an option to start
passed to LKT
passed to LKT
TRUE or FALSE
TRUE or FALSE
One of "BIC","AUC","AIC", and "RMSE"
Character Vector | Excludes specified features from the test list.
Character Vector | Excludes specified components from the test list.