var_importance_callback: Importance Callback

In cmcmaster1/torchtabular: Package for training tabular models using the {torch} package.

Description

Importance Callback

Usage

var_importance_callback(
  names,
  top = 10,
  start_epoch = 1,
  retain_data = TRUE,
  retain_plot = TRUE
)

Arguments

names	variable names as they appear in the model input x
top	number of variables to plot (top n). Default is 10.
start_epoch	the epoch to start plotting from. Default is 1.

Value

two outputs saved to the model context (importance_plot and importance_data)

cmcmaster1/torchtabular documentation built on Dec. 19, 2021, 5:17 p.m.