Jonah Gabry
Jonah Gabry
Maybe using geom_bar() instead of geom_histogram() let us get around that problem?
I added a few mentions of this to the doc in 8352a81, but I'm going to leave this issue open because I want to eventually add a vignette (or section...
And prior PCs will be more flexible once we have #151. Currently y needs to be included, which is not always desirable.
Thanks @VeenDuco. > Only in R/mcmc-diagnostics.R there are still some occurrences of labels = c("low", "ok", "high") instead of labels = c("low", "mid", "high"). Can you point me to where...
I've received a few questions about why we would expose the `_data` functions if the data is already stored in the ggplot object, so I'm adding some comments here to...
Thanks @ecoronado92! I've wanted to have some more intuitive demonstrations of the LOO PIT stuff for a while, so this would be great to have as a vignette. I'll try...
Sorry I haven't chimed in yet! After I moved we also ended up having some issues with the latest RStan release and I've been spending a lot of time responding...
Ok great thanks! > How should I proceed in either case - a pull request in the script for `ppc_loo_pit_overlay` or would you prefer I start tinkering my demo linked...
Cool thanks! Yeah I think the red curve looks much better. @avehtari is this what you had in mind for the corrected density estimation?
Awesome thanks!