Skip to contents

Initalization

The function to initiate the underlying Python environment.

init_py()
'python' environment initialization

Basic Nodes

Functions to customize stuctures of GP and DGP emulators.

kernel()
Initialize a Gaussian process node
Hetero()
Initialize a heteroskedastic Gaussian likelihood node
Poisson()
Initialize a Poisson likelihood node
NegBin()
Initialize a negative Binomial likelihood node

GP, DGP, and Linked (D)GP Emulations

Functions to train, validate, and make predictions from GP, DGP, and Linked (D)GP emulators.

gp()
Gaussian process emulator construction
dgp()
Deep Gaussian process emulator construction
lgp()
Linked (D)GP emulator construction
continue()
Continue the training of a DGP emulator
combine()
Combine layers
predict(<dgp>) predict(<lgp>) predict(<gp>)
Predictions from GP, DGP, or linked (D)GP emulators
validate()
Validate a constructed GP, DGP, or linked (D)GP emulator
plot(<dgp>) plot(<lgp>) plot(<gp>)
Validation plots of a constructed GP, DGP, or linked (D)GP emulator
prune()
Static pruning of a DGP emulator

Sequential Design

Functions to implement sequential designs for (D)GP and bundles of (D)GP emulators.

design()
Sequential design of a (D)GP emulator or a bundle of (D)GP emulators
alm()
Locate the next design point for a (D)GP emulator or a bundle of (D)GP emulators using ALM
mice()
Locate the next design point for a (D)GP emulator or a bundle of (D)GP emulators using MICE
vigf()
Locate the next design point for a (D)GP emulator or a bundle of (D)GP emulators using VIGF
pei()
Locate the next design point for a (D)GP emulator or a bundle of (D)GP emulators using PEI
update()
Update a GP or DGP emulator
draw()
Validation plots of a sequential design
pack()
Pack GP and DGP emulators into a bundle
unpack()
Unpack a bundle of (D)GP emulators

Helpers

summary(<gp>) summary(<dgp>) summary(<lgp>)
Summary of a constructed GP, DGP, or linked (D)GP emulator
write()
Save the constructed emulator
read()
Load the stored emulator
set_seed()
Random seed generator
set_linked_idx()
Set linked indices
set_imp()
Reset number of imputations for a DGP emulator
window()
Trim the sequences of model parameters of a DGP emulator
trace_plot()
Plot of DGP model parameter traces
nllik()
Calculate negative predicted log-likelihood