README.md

In ja-thomas/mlComp: Machine Learning competitions and challenges in R

mlComp - Machine Learning Challenges based on mlr and OpenML

• Install the development version

  splus devtools::install_github("ja-thomas/mlComp")

General overview

Machine Learning competitions and challenges in R. Create mlr learner to beat performance threshold on OpenML tasks. Users can set difficulty level and time limit.

library(mlComp)
chall = Challenge$new(id = 3, difficulty = "very easy", time.limit = 3600)
lrn = makeLearner("classif.rpart")
chall$submit(lrn)

# random challenge (From OpenML-CC18)
chall.rand = Challenge$new(difficulty = "very easy", time.limit = 3600)

# daily challenge
chall.rand = Challenge$new("daily", difficulty = "very easy", time.limit = 3600)

Data Preprocessing

The dataset can not be changed by the user, but preprocessing can be done via mlrCPO

library(mlComp)
library(mlrCPO)
chall = Challenge$new(id = 3, difficulty = "very easy", time.limit = 3600)
lrn = cpoModelMatrix(~ 0 + .^2) %>>%  # interactions
   makeLearner("classif.rpart")
chall$submit(lrn)

ja-thomas/mlComp documentation built on May 31, 2019, 7:41 a.m.