get_tau: Optimal threshold (tau) computation for 1.5-entmax

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

Optimal threshold (tau) computation for 1.5-entmax

Description

Optimal threshold (tau) computation for 1.5-entmax

Usage

get_tau(input, dim = -1L, k = NULL)

Arguments

input	The input tensor to compute thresholds over.
dim	The dimension along which to apply 1.5-entmax. Default is -1.
k	The number of largest elements to partial-sort 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. Default is NULL.

Value

The threshold value for each vector, with all but the dim dimension intact.

tabnet documentation built on June 12, 2026, 5:06 p.m.