Allow for supervised/semi-supervised dimension reduction with labels
Opened this issue · 1 comments
ShanSabri commented
Is it possible to provide this implementation of UMAP label information (even partial labeling) to help guide the embedding?
tkonopka commented
It's an interesting question. At the moment, that is not possible (version 0.2.0).
Python package umap-learn has support for supervised and semi-supervised embedding. It should be possible to port that functionality to the R package. both into the method="umap-learn" wrapper and the method="naive" implementation. Unfortunately, I can't give an estimate or timeline for that now.