Skip to content

Latest commit

 

History

History
76 lines (53 loc) · 1.93 KB

NEWS.md

File metadata and controls

76 lines (53 loc) · 1.93 KB

ubms 1.2.7

  • Package links now go to biodiverse Github org
  • Use reformulas instead of lme4
  • Return eta from ranef when submodel has spatial component
  • Support terra format rasters in predict
  • Minor bugfixes

ubms 1.2.6

  • Fix deprecated Stan code

ubms 1.2.5

  • Fix issue with version number not being a string
  • Adding missing roxygen tag

ubms 1.2.4

  • Adjust Makevars to fix problems with clang 16

ubms 1.2.3

  • Fix bug in Stan code revealed by new StanHeaders
  • Drop C++ specification per CRAN instructions
  • Remove deprecated ggplot2 functions
  • Small updates to overview vignette

ubms 1.2.2

  • Fix broken URLs in vignettes
  • New kfold function
  • Add ability to not save log likelihoods in posterior
  • Add new Laplace prior (thanks Justin Cally)
  • Better handling of factors with levels that don't appear in data
  • Fix CRAN clang warnings
  • Misc. bugfixes

ubms 1.1.0

  • Support for restricted spatial regression (RSR)
  • Add vignette for RSR
  • Update default priors and allow setting custom priors
  • Support for offsets
  • New utility functions and bugfixes

ubms 1.0.2

  • Make default priors less informative
  • Fix issues in sim_z C++ code that triggered clang-UBSAN errors
  • Fix various minor issues with checks/tests on R-oldrel

ubms 1.0.1

  • Added vignette comparing ubms and JAGS output
  • Added configuration for pkgdown site
  • Small adjustments to wording in DESCRIPTION and docs for CRAN submission

ubms 0.1.9

  • Add time-to-detection occupancy model (stan_occuTTD)
  • Wrote examples for all fitting functions
  • Many new tests added
  • Wrote a vignette demonstrating random effects

ubms 0.1.8

  • Added distance-sampling model (stan_distsamp)
  • Added multinomial-Poisson mixture model (stan_multinomPois)
  • Wrote overview vignette

ubms 0.1.4

  • Added dynamic occupancy model (stan_colext)
  • Consolidated single-season models into one Stan file to speed up compilation
  • Using rasters as predict input is now supported