Skip to content

Latest commit

 

History

History
60 lines (49 loc) · 2.71 KB

ROADMAP.md

File metadata and controls

60 lines (49 loc) · 2.71 KB

xpose development roadmap

New features

  • Add outliers labeling features e.g. aes(color = outlier(CWRES, ±6))
  • Add feature to exp-transform plot scales as exp = "xy"
  • Add reference distribution in eta_distrib when guide = TRUE
  • Add data checkout diagnostics
  • Add support for time-to-event (TTE) data
  • Add support for categorical data
  • Add gam and diagnostics
  • Add diagnostics for bootstrap, nonpb
  • Add diagnostics for cdd
  • Add diagnostics for frem
  • Add diagnostics for lasso
  • Add diagnostics for llp
  • Add diagnostics for mcmp
  • Add diagnostics for mimp
  • Add diagnostics for npc
  • Add diagnostics for npde
  • Add diagnostics for randtest
  • Add diagnostics for resmod
  • Add diagnostics for scm, boot_scm, xv_scm
  • Add diagnostics for simeval
  • Add diagnostics for sir
  • Add diagnostics for sse
  • Add diagnostics to compare several runs e.g. compare(xpdb1, xpdb2)

Improvements

High priority

  • Update xpose functions to use rlang instead of the standard evaluation version of tidyverse functions (e.g. mutate_())

  • Collect (from the lst file) and make use of parameter labels and units in plot labels

  • Get shrinkage in summary from PsN shrinkage and .shk file

  • Improve get_file() to handle 2x same method, same issue with all file dependent functions

  • Improve integration of additional modeling tools by using more generic functions

  • Improve plotting of single line point when surrounded by NA (vpc/ind_plots):

    1. Identify line points that are "alone"
    2. Have a special geom_point() mapped after the line aesthetics that is only plotted if single points are required?

Low priority

  • Improve the use of color, shape scales in ind_plots() and vpc(), e.g. add color scale options to the xp_theme
  • Find a way to add doses (bolus and infusion) to individual plots (shaded areas and/or vertical lines)
  • Improve xp_theme() use with a + method as in ggplot2 (e.g. here)
  • Allow eta_*() functions to read ETA values from table data or .phi file?
  • Should @nobs @nsim @nind values be updated when using filtering of the data?
  • Use the kappa() function to calculate the condition number directly?

Satelite packages development

  • xpose_shiny: a click based shiny interface for xpose
  • xpose_monolix: add support for monolix [help wanted]
  • others ...

Will not do

  • Use delayed assignment for data importation
  • Use general aes assignment (e.g. color = will impact point + lines) targeting ggplot()