Meeting chat log
00:12:38 Daniel Chen: it's essentially doing the multiple recipes and collecting the model metrics for you across all your preprocessing steps/models
00:12:40 Daniel Chen: ?
00:14:29 Daniel Chen: fn
The function to run. Acceptable values are: tune::tune_grid(), tune::tune_bayes(), tune::fit_resamples(), finetune::tune_race_anova(), finetune::tune_race_win_loss(), or finetune::tune_sim_anneal().
00:15:00 Daniel Chen: seems like there's only a few functions that are availiable to be used
00:16:36 Daniel Chen: but they're using the string instead of quoted form because they're matching on string to see which functions are allowed: https://github.com/tidymodels/workflowsets/blob/main/R/workflow_map.R#L101
00:16:53 Ildiko Czeller: makes sense, thanks
00:16:55 Toryn Schafer (she/her): Thanks, Daniel!
00:32:13 Daniel Chen: i guess they're using tidyposterior, instead of tidymodels. so i guess that's what's adding to the confusion
00:35:06 Daniel Chen: cross
A logical: should all combinations of the preprocessors and models be used to create the workflows? If FALSE, the length of preproc and models should be equal.
00:49:17 jiwan: tune_grid(
object,
preprocessor,
resamples,
...,
param_info = NULL,
grid = 10,
metrics = NULL,
control = control_grid()
)
00:50:04 Daniel Chen: https://tune.tidymodels.org/reference/tune_grid.html
00:50:52 jiwan: A data frame of tuning combinations or a positive integer. The data frame should have columns for each parameter being tuned and rows for tuning parameter candidates. An integer denotes the number of candidate parameter sets to be created automatically