When using the select_by_one_std_err() function in tidymodels pipeline, in the "..." argument, the documentation in the help says I need to put the parameter to sort the model from most simple to most complex. So, if using glmnet's penalty parameter, since a bigger penalty will be a simpler model, I should put desc(penalty) in as the "..." argument? Am I interpreting this correctly? I'm happy to provide a reprex if need be. Thank you!
From the function signature
\dots For select_by_one_std_err() and select_by_pct_loss(), this argument is passed directly to dplyr::arrange() so that the user can sort the models from most simple to most complex. See the examples below. At least one term is required for these two functions.
Like all \dots arguments, it's optional.
See the FAQ: How to do a minimal reproducible example reprex for beginners. Having to reverse engineer the code to illustrate the question is an obstacle to getting specific answers, in this case whether penalty is the appropriate argument. It does however, have to appear in the output, rather than the argument to the select_by_one_std_err function.
I’ll work on making a reprex. In this case the ... argument is required, which also confused me at first.
Yes.
My emphasis to point out that it's not optional.
1 Like
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.
If you have a query related to it or one of the replies, start a new topic and refer back with a link.