search

scicloj / tablecloth

Dataset manipulation library built on the top of tech.ml.dataset
https://scicloj.github.io/tablecloth
MIT License
294 stars 26 forks source link

Aproximate Bayesian Bootstrap for missing values #46

Closed genmeblog closed 2 years ago

genmeblog commented 3 years ago

Another method for missing imputation.

Algorithm (for each column)

  1. Bootstrap non missing samples (len = number of non missing)
  2. Bootstrap from above number of missing values.
  3. Use this values to replace missing.
genmeblog commented 2 years ago

moved here: https://github.com/techascent/tech.ml.dataset/issues/273