Skip to content

Use scipy.stats.yeojohnson PowerTransformer #26308

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
lorentzenchr opened this issue Apr 30, 2023 · 5 comments · Fixed by #31227 · May be fixed by Projet-open-source/scikit-learn#3
Closed

Use scipy.stats.yeojohnson PowerTransformer #26308

lorentzenchr opened this issue Apr 30, 2023 · 5 comments · Fixed by #31227 · May be fixed by Projet-open-source/scikit-learn#3
Labels
help wanted Moderate Anything that requires some knowledge of conventions and best practices module:preprocessing Refactor Code refactor

Comments

@lorentzenchr
Copy link
Member

lorentzenchr commented Apr 30, 2023

Inside PowerTransformer, we should use scipy.stats.yeojohnson instead of our own implementation.

scipy.stats.yeojohnson was release with scipy 1.2.0. With PR #24665, we now have even 1.5.0 as minimum.

Edit: Note that scipy/scipy#18389 to be release in scipy 1.12 will also resolve #23319.

@lorentzenchr
Copy link
Member Author

@lsorber are you interested?

@lsorber
Copy link

lsorber commented Apr 30, 2023

Sure, but I'd first like to see the outcome of scipy/scipy#18389 before contributing a PR.

@thomasjpfan thomasjpfan added module:preprocessing Refactor Code refactor and removed Needs Triage Issue requires triage labels May 4, 2023
@vishalj0501
Copy link

Hello, I'm a first timer and almost new to open source,
Can I take up this issue ?

@lorentzenchr
Copy link
Member Author

@lsorber kind of brought up this issue and already indicated his interest.

@xuefeng-xu
Copy link
Contributor

PR #27818 uses scipy.stats.yeojohnson instead of our own implementation in PowerTransformer. However, we need to bump the minimum version of scipy from 1.5.0 to 1.9.0 to resolve this issue #14959.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
help wanted Moderate Anything that requires some knowledge of conventions and best practices module:preprocessing Refactor Code refactor
Projects
None yet
5 participants