search

mlr-org / mlr3pipelines

Dataflow Programming for Machine Learning in R
https://mlr3pipelines.mlr-org.com/
GNU Lesser General Public License v3.0
140 stars 25 forks source link

create PipeOps for oversampling and undersampling #102

Closed berndbischl closed 5 years ago

mb706 commented 5 years ago

PipeOpDownsample already exists

mb706 commented 5 years ago

See slack discussion

mb706 commented 5 years ago

Plan is to do oversampling by sampling without replacement; if more samples are needed than there are instances, the instances are replicated fully multiple times and then sampling without replacement is performed.