-
-
Notifications
You must be signed in to change notification settings - Fork 23
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
new Pareto-k diagnostics vignette #372
Conversation
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 44fc43e is merged into master:
|
Thanks! Do you think this vignette needs additional review? If yes, could @n-kall do it since you have been implementing these methods. |
Sure, I can look at this |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I haven't yet had a chance to run the vignette, but as the tests were passing (except on R 3.6) I have gone through and made some suggestions.
Most suggestions are removing the |>
operator which is not available in R < 4.1.0. If we want to use the native pipe, I think would need to bump the required version to 4.1.0 (as far as I understand)
vignettes/pareto_diagnostics.Rmd
Outdated
dr<-array(data=replicate(4,as.numeric(arima.sim(n = N, | ||
list(ar = c(phi)), | ||
sd = sqrt((1-phi^2))))), | ||
dim=c(N,4,1)) |> | ||
as_draws_df() |> | ||
set_variables('xn') |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't think we can use the |>
operator unless we bump the required R version.
Instead this could be written
dr<-array(data=replicate(4,as.numeric(arima.sim(n = N, | |
list(ar = c(phi)), | |
sd = sqrt((1-phi^2))))), | |
dim=c(N,4,1)) |> | |
as_draws_df() |> | |
set_variables('xn') | |
dr <- array(data = replicate(4, as.numeric(arima.sim(n = N, | |
list(ar = c(phi)), | |
sd = sqrt((1 - phi^2))))), | |
dim = c(N, 4, 1)) | |
dr <- as_draws_df(dr) | |
variables(dr) <- "xn" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
As posterior
suggests dplyr
, can I use %>%
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, I think so
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good point. dplyr is importing magrittr so should work. That will be a much better solution
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Switched to %>%
vignettes/pareto_diagnostics.Rmd
Outdated
We examine the draws with the default `summarise_draws()`. | ||
|
||
```{r summarise_draws} | ||
drt |> summarise_draws() |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
drt |> summarise_draws() | |
summarise_draws(drt) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Would it be better here to use default_convergence_measures()
to focus on the convergence diagnostics at first?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I like it as it is now
Co-authored-by: Noa Kallioinen <[email protected]>
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 44fc43e is merged into master:
|
Thanks Noa. I made the suggested, except for using I also fixed the two references in |
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 44fc43e is merged into master:
|
Co-authored-by: Noa Kallioinen <[email protected]>
Thanks Noa, you have magical eyes for spotting errors! |
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 44fc43e is merged into master:
|
Looks good! All the vignettes are failing to build with R 3.6 but I think that is due to the |
Fixes #322