Roadmap: gammodels package

[mdancho84]

Modelgam Project Roadmap - MVP

• [X] Create Package Structure
• [ ] Develop MVP Algorithms (#2) - Get basic use-cases from business-science/modeltime#71, Test on several datasets
• [ ] Develop R
  • [x] Top Priority - gen_additive_mod() algorithm with gam engine (#2)
  • [x] Fix Workflow Issue (#3)
  • [ ] Add classification / logistic regression - role = "classification"
  • [ ] Determine if any parameters needed for gen_additive_mod()
• [ ] Create Tests (#4)
• [x] Automate Tests with GitHub Actions ()
• [ ] Make AWESOME README ()
• [ ] Add Vignettes - Getting Started (#5)
• [ ] Make Hex Logo ()
• [x] Add Pkgdown Documentation ()
• [ ] Finalize DESCRIPTION - Update authors
• [ ] Send to CRAN

Future Backends

We can use caret as a baseline for adding more backends (engines) http://topepo.github.io/caret/train-models-by-tag.html#generalized-additive-model

• [ ] bam
• [ ] gamm
• [ ] mboost / boosted gams https://github.com/boost-R

Recipes GAM Steps

Evaluate the feasibility of using a recipe interface for applying smoothing functions. Something like using step_gam_smooth() as a way of replacing s() inside a formula.

[BenWynne-Morris]

https://github.com/topepo/caret/blob/master/models/files/gam.R

[AlbertoAlmuinha]

Hi @BenWynne-Morris ,

Please, subscribe to #2 , I have written you a message there.

Regards