Version 1.4.0 of the bayesplot R package is now available. Until CRAN binaries are built (usually a few days) you can install the update from source using
install.packages("bayesplot", type = "source")
The release notes are copied below and also available on GitHub or the bayesplot webpage.
I would also like to mention that I’m very happy that TJ Mahr (@tjmahr on Discourse, GitHub, Twitter) has agreed to join me in developing bayesplot. TJ’s expertise in R and ggplot2 (among other things!) will be a huge asset as we continue to try to make bayesplot as useful as possible both for individual users and for developers of packages interfacing with Stan.
Release notes for bayesplot v1.4.0
(GitHub issue/PR numbers in parentheses)
- New plotting function
mcmc_parcoord()
for parallel coordinates plots of
MCMC draws (optionally including HMC/NUTS diagnostic information). (#108) -
mcmc_scatter
gains annp
argument for specifying NUTS parameters, which
allows highlighting divergences in the plot. (#112) - New functions with names ending with suffix
_data
don’t make the plots,
they just return the data prepared for plotting (more of these to come in
future releases):-
ppc_intervals_data()
(#101) -
ppc_ribbon_data()
(#101) -
mcmc_parcoord_data()
(#108) -
mcmc_rhat_data()
(#110) -
mcmc_neff_data()
(#110)
-
-
ppc_stat_grouped()
,ppc_stat_freqpoly_grouped()
gain afacet_args
argument for controlling ggplot2 faceting (many of themcmc_
functions
already have this). - The
divergences
argument tomcmc_trace()
has been deprecated in favor
ofnp
(NUTS parameters) to match the other functions that have annp
argument. - Fixed an issue where duplicated rhat values would break
mcmc_rhat()
(#105).