0

I wanto to perform class balancing using h2o autoML. I know there is a parameter class_sampling_factors that allow to specify the under/over sampling factor for each class.

If this parameter is not specified, h2o makes both under and over sampling. Is there any reference to understand how the procedure works? I would like to understand if in case of really imbalanced classes, the procedure allows for a mix of under/over sampling

A1010
  • 193
  • 8
  • [Are unbalanced datasets problematic, and (how) does oversampling (purport to) help?](https://stats.stackexchange.com/q/357466/1352) – Stephan Kolassa Jul 20 '20 at 11:40
  • You are recommended to take a look at: https://stats.stackexchange.com/questions/476911/how-are-artificially-balanced-datasets-corrected-for/476948#476948 – Match Maker EE Jul 20 '20 at 12:07
  • Thank you for the feedback. Considering your answer, I will really appreciate if you can answer also at this question (https://datascience.stackexchange.com/questions/78011/how-to-do-class-balancing). Thanks! – A1010 Jul 20 '20 at 15:25

0 Answers0