funGp (version 0.2.1)

funGp-package: Gaussian Process Models for Scalar and Functional Inputs

Description

Construction and smart selection of Gaussian process models with emphasis on treatment of functional inputs.

Base functionalities

• Main methods fgpm: creation of funGp regression models predict: output estimation at new input points based on a funGp model simulate: random sampling from a funGp Gaussian process model update: modification of data and hyperparameters of a funGp model

• Plotters plotLOO: validation plot for a funGp model plotPreds: plot of predictions based on a funGp model plotSims: plot of simulations based on a funGp model

Model selection

• Main method fgpm_factory: structural parameter optimization

• Plotters pre-optimization decay: regularized initial pheromones decay2probs: normalized initial pheromones

• Plotters post-optimization plotX: absolute and relative quality of the optimized model plotEvol: evolution of the algorithm

• Correction post-optimization of input data structures which_on: indices of active inputs in a model structure delivered by fgpm_factory get_active_in: extraction of active input data based on a model structure delivered by fgpm_factory

Useful material

• Manual Gaussian Process Regression for Scalar and Functional Inputs with funGp - The in-depth tour

• Paper Gaussian process metamodeling of functional-input code for coastal flood hazard assessment

• Tech. report Ant Colony Based Model Selection for Functional-Input Gaussian Process Regression

Authors

Jos<U+00E9> Betancourt, Fran<U+00E7>ois Bachoc and Thierry Klein

Contributors

D<U+00E9>borah Idier and J<U+00E9>r<U+00E9>my Rohmer