Parameters for the tabnet model
decision_width(range = c(8L, 64L), trans = NULL)
attention_width(range = c(8L, 64L), trans = NULL)
num_steps(range = c(3L, 10L), trans = NULL)
feature_reusage(range = c(1, 2), trans = NULL)
num_independent(range = c(1L, 5L), trans = NULL)
num_shared(range = c(1L, 5L), trans = NULL)
momentum(range = c(0.01, 0.4), trans = NULL)
mask_type(values = c("sparsemax", "entmax"))
the default range for the parameter value
whether to apply a transformation to the parameter
possible values for factor parameters
These functions are used with tune
grid functions to generate
candidates.
A dials
parameter to be used when tuning TabNet models.