SensMLP {NeuralSens}

Constructor of the SensMLP Class

Description

Create an object of SensMLP class

Usage

SensMLP(
  sens = list(),
  raw_sens = list(),
  mlp_struct = numeric(),
  trData = data.frame(),
  coefnames = character(),
  output_name = character(),
  cv = NULL,
  boot = NULL,
  boot.alpha = NULL
)

Arguments

sens

list of sensitivity measures, one data.frame per output neuron

raw_sens

list of sensitivities, one matrix per output neuron

mlp_struct

numeric vector describing the structur of the MLP model

trData

data.frame with the data used to calculate the sensitivities

coefnames

character vector with the name of the predictor(s)

output_name

character vector with the name of the output(s)

cv

list list with critical values of significance for std and mean square.

boot

array bootstrapped sensitivity measures.

boot.alpha

array significance level. Defaults to NULL. Only available for analyzed caret::train models.

Value

SensMLP object

References

Pizarroso J, Portela J, Muñoz A (2022). NeuralSens: Sensitivity Analysis of Neural Networks. Journal of Statistical Software, 102(7), 1-36.