Skip to content

Verbosity enhancements#518

Merged
fweber144 merged 8 commits into
stan-dev:masterfrom
fweber144:verbosity_continued
May 19, 2025
Merged

Verbosity enhancements#518
fweber144 merged 8 commits into
stan-dev:masterfrom
fweber144:verbosity_continued

Conversation

@fweber144
Copy link
Copy Markdown
Collaborator

This is a follow-up for PR #516: Thanks to #516, the PR presented here can add "extra" verbosity (i.e., when global option projpred.extra_verbose is set to TRUE) about the current fold when running cv_varsel() with validate_search = TRUE or K-fold CV with validate_search = FALSE. There are some other minor enhancements to verbosity which are explained in the commit messages.

File NEWS.md is unchanged here. I will update it as soon as #511 (or #508) has been merged. I would propose to change the following entry:

* Minor enhancements concerning verbosity (e.g., the number of projected draws---resulting from clustering or thinning---is now printed out during the different steps of the computations). Also introduced global option `projpred.verbose` which may be used to set argument `verbose` of `varsel()` and `cv_varsel()` globally. (GitHub: #506)

to:

* Several enhancements concerning verbosity (e.g., the number of projected draws---resulting from clustering or thinning---is now printed out during the different steps of the computations and verbose-mode output is redirected to `stderr()` instead of `stdout()`). Also introduced global option `projpred.verbose` which may be used to set argument `verbose` of `varsel()` and `cv_varsel()` globally. (GitHub: #506, #518)

fweber144 added 8 commits May 19, 2025 21:17
PSIS-LOO CV or K-fold CV with `validate_search = TRUE`); these messages mention
the observation/fold for which computations are currently taking place.

Also rename `verbose_search` to `verbose_obs` (in `one_obs()`) and
`verbose_fold` (in `one_fold()`).

Also use a more consistent naming of objects containing text for verbose-mode
messages (`vtxt_[...]` instead of `verb_[...]`).
…sages);

this makes those messages more similar to the `print.vselsummary()` output
`one_obs()` and `one_fold()` (and high verbosity)
call that is used for creating `summaries_fast`
also set `file = stderr()` in `utils::txtProgressBar()` calls (this is the
behavior in the progressr package and also mentioned in the documentation of
argument `file` of `utils::txtProgressBar()`)
…to `~ 1`";

this is done not only for the intercept-only model, but for all submodel
formulas
@fweber144 fweber144 merged commit d61ff5c into stan-dev:master May 19, 2025
@fweber144 fweber144 deleted the verbosity_continued branch May 19, 2025 19:43
fweber144 added a commit that referenced this pull request Jun 6, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant