plot_actives: Active variable plotting

In hmer: History Matching and Emulation Package

Active variable plotting

Description

For a set of emulators, demonstrate which variables are active.

Usage

plot_actives(ems, output_names = NULL, input_names = NULL)

Arguments

ems	The list of emulators to consider
output_names	The names of the outputs to include in the plot, if not all
input_names	The names of the inputs to include in the plot, if not all

Details

Each emulator has a list of ‘active’ variables; those which contribute in an appreciable way to its regression surface. It can be instructive to examine the differences in active variables for a give collection of emulators. The plot here produces an nxp grid for n emulators in p inputs; a square is blacked out if that variable does not contribute to that output.

Both the outputs and inputs can be restricted to collections of interest, if desired, with the optional output_names and input_names parameters.

Value

A ggplot object corresponding to the plot

See Also

Other visualisation tools: behaviour_plot(), diagnostic_wrap(), effect_strength(), emulator_plot(), hit_by_wave(), output_plot(), plot_lattice(), plot_wrap(), simulator_plot(), space_removed(), validation_pairs(), wave_dependencies(), wave_points(), wave_values()

Examples

 plot_actives(SIREmulators$ems)
 # Remove the nR output and aIR input from the plot
 plot_actives(SIREmulators$ems, c('nS', 'nI'), c('aSI', 'aSR'))
 # Note that we can equally restrict the emulator list...
 plot_actives(SIREmulators$ems[c('nS', 'nI')], input_names = c('aSI', 'aSR'))

hmer documentation built on June 22, 2024, 9:22 a.m.