entmax15: Alpha-entmax

In mlverse/tabnet: Fit 'TabNet' Models for Classification and Regression

Alpha-entmax

Description

With alpha = 1.5 and normalizing sparse transform (a la softmax).

Usage

entmax(dim = -1)

entmax15(dim = -1L, k = NULL)

Arguments

dim	The dimension along which to apply 1.5-entmax.
k	The number of largest elements to partial-sort input over. For optimal performance, should be slightly bigger than the expected number of non-zeros in the solution. If the solution is more than k-sparse, this function is recursively called with a 2*k schedule. If NULL, full sorting is performed from the beginning.

Details

Solves the optimization problem: \max_p <input, P> - H_{1.5}(P) \text{ s.t. } P \geq 0, \sum(P) == 1 where H_{1.5}(P) is the Tsallis alpha-entropy with \alpha=1.5.

Value

The projection result P of the same shape as input, such that \sum_{dim} P = 1 \forall dim elementwise.

Examples

## Not run: 
input <- torch::torch_randn(10,5, requires_grad = TRUE)
# create a top3 alpha=1.5 entmax on last input dimension
nn_entmax <- entmax15(dim=-1L, k = 3)
result <- nn_entmax(input)

## End(Not run)

mlverse/tabnet documentation built on July 17, 2025, 4:15 a.m.