HappyTree: Type Safe and End to End Decision Tree

[ bsd3, data, library ] [ Propose Tags ]




Maintainer's Corner

Package maintainers

For package maintainers and hackage trustees


  • No Candidates
Versions [RSS] 0.2018.1.5, 0.2018.1.7, 0.2018.1.8
Change log ChangeLog.md
Dependencies base (>=4.7 && <5), constraints (==0.9.1), generics-sop (==, safe (==0.3.15), singletons (==2.3.1) [details]
License BSD-3-Clause
Copyright 2018 Marisa Kirisame
Author Marisa Kirisame
Maintainer lolisa@marisa.moe
Category Data
Home page https://github.com/MarisaKirisame/HappyTree#readme
Bug tracker https://github.com/MarisaKirisame/HappyTree/issues
Source repo head: git clone https://github.com/MarisaKirisame/HappyTree
Uploaded by MarisaVeryMoe at 2018-01-08T16:23:39Z
Reverse Dependencies 1 direct, 0 indirect [details]
Downloads 1941 total (10 in the last 30 days)
Rating (no votes yet) [estimated by Bayesian average]
Your Rating
  • λ
  • λ
  • λ
Status Docs available [build log]
Last success reported on 2018-01-08 [all 1 reports]

Readme for HappyTree-0.2018.1.8

[back to package description]


Happy Tree is a end-to-end decision tree library in Haskell.

That mean, it's main selling point is, while most decision tree library support splitting on Discrete Category/Continuous Variable (Read: essentially Int and Double), Happy Tree let you split on any type, as long as you specify how to split on it.

Want to decide on List or AST? No Problem!

Want to use your custom quantile search like thingy to speed up splitting on continuous variable? Piece of Cake!

Want to use different split strategy at the same time, splitting on the finest choice? You Name It: It form a monoid.

Known Problem

Can only split on finitely many type now. Cannot split on

data Perfect a = Here a | More (Perfect (a, a)).