search

blind-contours / CVtreeMLE

:deciduous_tree: :dart: Cross Validated Decision Trees with Targeted Maximum Likelihood Estimation
MIT License
5 stars 1 forks source link

Reduce number of dependencies #26

Open wleoncio opened 1 year ago

wleoncio commented 1 year ago

Eyeing a potential CRAN submission (or at least a welcome bloat-reduction), would you consider reducing the number of dependencies in your package 20+ is a lot, IMO)?

blind-contours commented 1 year ago

This package requires a lot of dependencies by it's nature - integrating trees as a data-adaptive parameter (pre, partykit etc.) with machine learning (sl3, ranger, and others) then all the data formatting and parallelization (furrr, data.table, purr, stringr, stats, dplyr etc.). I see a couple ways moving forward where I can clean some things up. 1. Move simulations out of the package, this knocks out 1-2 packages that are required and moves them to suggests, 2. Getting rid of the plotting functions that come default in the package would get rid of 3. Also making the readme less pretty with knit I could also get rid of that. I'll make these changes before the submission to CRAN. Also SL3 is being updated for CRAN submission and that should get rid of the misc. dependencies we need to install for that package (should be packaged in SL3).

wleoncio commented 1 year ago

Thank you for the reply, I'm happy to see a plan for this! :)