The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
stars input
(x/y grid dimensions), alongside the existing
vector-geometry input, including a new raster_adjacency()
internal helper and dispatch in
genclust()/sfclust().chapa, a real NDWI2 vegetation-index raster
dataset from El Chaparrillo, Spain, and a new article vignette
demonstrating raster clustering
(vg13-ndwi-chaparrillo.Rmd).NA response values (e.g. raster
cells outside a study region) are now automatically excluded from
clustering via response / valid_ids, instead
of requiring the user to pre-filter their data.update() methods for sfclust
objects: niter continues MCMC sampling from where a
previous fit left off, and sample refits the full INLA
models for a specific stored sample without further sampling.spnames/fnames) so spatial dimension names no
longer need to match a fixed convention.plot_clusters_fitted()/fitted() gained
inv_link/aggregation options, returning cluster-level
(mean_cluster) and inverse-link
(mean_cluster_inv) summaries.sfclust() now
dispatches to sfclust.data.frame() (core interface) and
sfclust.stars() (spatial convenience wrapper), replacing
the previous single-path implementation.
data_all()/filter_df() replace the old
stnames-based long-format conversion.NULL fitted models and raster
inputs.fitted()/plot() no longer require INLA to
be installed when applied to an already-fitted sfclust
object (e.g. one loaded from a saved .rds); the
inverse-link transform for standard families (gaussian, binomial,
poisson) is now computed natively instead of via
INLA::inla.link.inv*().These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.