act_funs: Activation Functions Specification Helper

In kindling: Higher-Level Interface of 'torch' Package to Auto-Train Neural Networks

Activation Functions Specification Helper

Description

This function is a DSL function, kind of like ggplot2::aes(), that helps to specify activation functions for neural network layers. It validates that activation functions exist in torch and that any parameters match the function's formal arguments.

Usage

act_funs(...)

Arguments

...

Activation function specifications. Can be: Bare symbols: relu, tanh Character strings (simple): "relu", "tanh" Character strings (with params): "softshrink(lambda = 0.1)", "rrelu(lower = 1/5, upper = 1/4)" Named with parameters: softmax = args(dim = 2L) Indexed syntax (named): softshrink[lambd = 0.2], rrelu[lower = 1/5, upper = 1/4] Indexed syntax (unnamed): softshrink[0.5], elu[0.5]

Value

A vctrs vector with class "activation_spec" containing validated activation specifications.

kindling documentation built on March 3, 2026, 9:07 a.m.