Relist plate scoring functions and only use the functions of the current plate in the loop #33
Conversation
…ions of the current plate in the loop."" This reverts commit b9f6c60.
banfai
changed the title
| @@ -266,7 +264,7 @@ optimize_multi_plate_design <- function(batch_container, across_plates_variables | |||
| if (!quiet && length(plate_levels) > 1) cat(curr_plate, "... ") | |||
|
|
|||
| bc <- optimize_design(bc, | |||
| scoring = scoring_funcs, | |||
| scoring = scoring_funcs %>% purrr::map(curr_plate) %>% rlang::set_names(paste(names(.), " Plate_", curr_plate)), | |||
There was a problem hiding this comment.
not sure I understand this code.
scoring_funcs is a list of lists now, and curr_plate is a a plate id (character or numeric). right?
what does scoring_funcs %>% purrr::map(curr_plate) do?
maybe we need to transpose the scoring_funcs first with purrr::list_transpose, and then we:
assert_thatlength ofplate_levelsis the same ofscoring_funcs?for (plate_index in seq_len(length(plate_levels)
There was a problem hiding this comment.
not sure I understand this code.
scoring_funcsis a list of lists now, andcurr_plateis a a plate id (character or numeric). right?what does
scoring_funcs %>% purrr::map(curr_plate)do?
First part is correct.
purrr::map(curr_plate) is equivalent to purrr::map(purrr::pluck, curr_plate)
Maybe it should be written like that for readability.
maybe we need to transpose the
scoring_funcsfirst withpurrr::list_transpose, and then we:
assert_thatlength ofplate_levelsis the same ofscoring_funcs?for (plate_index in seq_len(length(plate_levels)
I need to check this.
This reverts the revert commit b9f6c60.
Can this be rebased/squashed to somehow remove these idiotic commits?
fixes #32