Pivot metrics

[topepo]

A function to collect metrics then pivot_wider() on them so that metrics are in columns.

[simonpcouch]

Been sitting with this and wanted to open a discussion before reviewing any technical bits: would this be better situated as an argument to collect_metrics()?

My hesitancy with pivot_metrics() is that the name currently suggests to me that its input is the output of collect_metrics() or metrics(), and the only alternatives I can think of for a name change seem ambiguous with collect_metrics(). I would also make the case this the name of this function should start with collect_*(), as this feels like a collection action and that naming would better support tab-completion. This function does feel more to me like an alternative usage of collect_metrics() than a different function altogether.

Transitioning this to a pivot argument with logical values, or type = c("long", "wide"), would also mean we wouldn't have to maintain a separate usage of this column identification boilerplate:

tune/R/collect.R

Lines 485 to 495 in dde5d35

	param_names <- .get_tune_parameter_names(x)
	id_names <- grep("^id", names(x), value = TRUE)
	group_cols <- .get_extra_col_names(x)
	all_bad <- is_cataclysmic(x)
	if (all_bad) {
	rlang::abort("All models failed. Run `show_notes(.Last.tune.result)` for more information.")
	}
	# The mapping of tuning parameters and .config.
	config_key <- .config_key_from_metrics(x)

[simonpcouch]

Closing in favor of #839. :)

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.