0

The title pretty much says it all, I have a df with 40+ dimension which I'd like to process into the Umap algorithm in order to have a 2-d output.

I would like to know if it is possible to weight the input columns differently for the purpose of studying the possible different Umap outcomes.

Thank you for your time

P.S. I work in python

Stuart Berg
  • 17,026
  • 12
  • 67
  • 99
Luca
  • 17
  • 5
  • Please provide enough code so others can better understand or reproduce the problem. – Community Sep 27 '22 at 20:23
  • Yes, it's possible to scale the input columns. Your question is related to the subject of feature normalization, [discussed in the UMAP FAQ](https://umap-learn.readthedocs.io/en/latest/faq.html#should-i-normalise-my-features). Is there any specific difficulty you're having while attempting to scale the inputs? – Stuart Berg Sep 29 '22 at 15:01

1 Answers1

1

Why not simply applying UMAP to A:

A = X*W

where X is your Nx40 matrix and W=diag(w) is a 40x40 diagonal matrix of weights w=[w1, w2,..., w40]?

Consider using normalized weights wi, i=1,2,...,40 such that sum(w) == 1, to distribute normally your information.

SystemSigma_
  • 1,059
  • 4
  • 18