tabnet (version 0.4.0)

tabnet_explain: Interpretation metrics from a TabNet model

Description

Interpretation metrics from a TabNet model

Usage

tabnet_explain(object, new_data)
# S3 method for default
tabnet_explain(object, new_data)
# S3 method for tabnet_fit
tabnet_explain(object, new_data)
# S3 method for tabnet_pretrain
tabnet_explain(object, new_data)
# S3 method for model_fit
tabnet_explain(object, new_data)

Value

Returns a list with

• M_explain: the aggregated feature importance masks as detailed in TabNet's paper.

• masks a list containing the masks for each step.

Arguments

object

a TabNet fit object

new_data

a data.frame to obtain interpretation metrics.

Examples

if (torch::torch_is_installed()) {

set.seed(2021)

n <- 1000
x <- data.frame(
  x = rnorm(n),
  y = rnorm(n),
  z = rnorm(n)
)

y <- x$x

fit <- tabnet_fit(x, y, epochs = 20,
                  num_steps = 1,
                  batch_size = 512,
                  attention_width = 1,
                  num_shared = 1,
                  num_independent = 1)


 ex <- tabnet_explain(fit, x)

}