Package index

Core workflow

Prepare the data, estimate the kernel hyperparameters, predict.

data_process()

Process raw epidemiological data for the GP model

infer_kernel_params()

Quick exact-marginal-likelihood estimate of the kernel hyperparameters

gp_predict()

Predict the latent rate and a count prediction interval (CG)

Kernels and distances

space_kernel()

Build the spatial correlation matrix

time_kernel()

Build the temporal correlation matrix

rbf_kernel()

Radial basis function kernel

periodic_kernel()

Periodic kernel

get_spatial_distance()

Pairwise spatial distances

get_temporal_distance()

Pairwise temporal distances

Building blocks

The pieces the workflow is assembled from, exported for direct use.

build_plugin_field()

Build a plug-in latent field from observed counts

gp_marginal_loglik()

Exact separable-GP marginal log-likelihood of a field, with a nugget

default_kernel_priors()

Default priors for the kernel hyperparameters

quick_mvnorm()

Quick multivariate normal draw over two dimensions

quick_mvnorm_chol()

Quick multivariate normal draw over two dimensions (Cholesky precomputed)