Simon P. Couch

use different package than rpart.plot to plot decision trees

8

Follow up with Max on which package this is. :)

For the "What makes a model" section, the "Predict with your model" slides and classwork differ slightly: https://github.com/tidymodels/workshops/blob/4a6d2b911ac2932e508324baac5082d99dabaf65/slides/03-what-makes-a-model.qmd#L527-L535 https://github.com/tidymodels/workshops/blob/4a6d2b911ac2932e508324baac5082d99dabaf65/classwork/03-classwork.qmd#L123-L129

Add alt-text to pictures, plots, etc; see https://posit.co/blog/knitr-fig-alt/ for examples.

Use cli errors over rlang / home-grown machinery

Consider running `usethis::use_tidy_dependencies()` and/or replace compat files with `use_standalone()`

removal from CRAN?

3

Hi there! Maintainer of a margins dependency, broom, here. I [noticed today](https://github.com/tidymodels/broom/issues/1200) that margins was removed from CRAN a few weeks ago. Just wanted to confirm that yall don't plan...

consider switching to xy interface for `kernlab::ksvm()`

1

Related to #1114. > This [SO post](https://stackoverflow.com/questions/78372245/tidymodels-workaround-for-kernlab-protection-stack-overflow-error) runs into troubles with the formula interface of `kernlab::ksvm()` which could be resolved by using the matrix interface of the kernlab function. We...

Closes #775. I'm unable to run the reprex in the linked issue with either version due to a `summarize(across())` error coming from the engine.

revisiting `tunable()` for model specifications

1

Going ahead and PRing now to introduce tests and run GHA with them so I can demonstrate they still pass once I make some upcoming changes. EDIT: Closes #1104.

issue with `tunable` entries in `tune_args.model_spec`

1

e.g.: ``` r library(tidymodels) tune_args( # known tunable linear_reg(penalty = tune()) %>% # not known tunable set_engine("glmnet", dfmax = tune()) ) #> # A tibble: 2 × 6 #> name...