Conversation
|
@roninsightrx @jasmineirx Added some functionality to aim/iterate/report t_gt_mic metrics common to beta-lactam therapeutics. Would appreciate suggestions / feedback as to how this should be better formulated. Not quite there yet but steps in the right direction. |
R/dose_grid_search.R
Outdated
| if(target_design$type %in% target_types_time) { | ||
| if(is.null(target_design$range)) { | ||
| target_design$range <- c(target_design$min,target_design$max) | ||
| } else { | ||
| if(min(target_design$range) >= 100) { | ||
| target_design$variable <- ifelse( | ||
| grepl("_free", target_design$type), | ||
| "CONCF", | ||
| "CONC" | ||
| ) | ||
| target_design$type <- "cmin" | ||
| target_design$value <- min(target_design$range)/100 * tail(covariates$MIC$value, 1) | ||
| target_design$range <- rep(target_design$value, 2) | ||
| } |
There was a problem hiding this comment.
since this won't change per-loop and it is basically only manipulating the target_design object, can we move it out of the loop, and maybe even to where the target design object is created?
| skip_hessian = TRUE, # faster | ||
| int_step_size = 0.01, | ||
| ... | ||
| int_step_size = 0.01 |
There was a problem hiding this comment.
Since a few months we don't allow just random arguments to PKPDmap anymore. This was now causing an issue in a project. We should probably also remove the passing of ... from the calling function, and see if that doesn't affect any existing features and scenarios.
3 participants
like "time % of free conc above MIC".