Learn R Programming

spartan (version 3.0.2)

create_ensemble: Internal function to create the ensemble

Description

Internal function to create the ensemble

Usage

create_ensemble(ensemble_emulations, all_emulator_predictions,
  emulator_test_data, measures, emulator_types, pre_normed_mins,
  pre_normed_maxes, algorithm_settings = NULL, normalise = FALSE,
  timepoint = NULL, output_formats = c("pdf"))

Arguments

ensemble_emulations

All emulations to build into the ensemble

all_emulator_predictions

Test set predictions from all emulators, on which the ensemble will be trained / emulators weighted

emulator_test_data

Data on which the ensemble performance will be assessed

measures

Simulation responses the model should predict

emulator_types

Machine learning techniques being employed

pre_normed_mins

The minimum values of each parameter prior to data normalisation. Used to rescale the results

pre_normed_maxes

The maximum values of each parameter prior to data normalisation. Used to rescale the results

algorithm_settings

Object output from the function emulation_algorithm_settings, containing the settings of the machine learning algorithms to use in emulation creation. Used here to obtain settings relevant to ensemble creation - namely number of generations and whether the ensemble should be saved to file, as well as whether plots should be produced showing ensemble performance.

normalise

Whether the predictions generated when testing the ensemble should be normalised for presenting test results

timepoint

Simulation timepoint for which an ensemble is being created

output_formats

File formats in which result graphs should be produced

Value

Generated ensemble object