-
-
Notifications
You must be signed in to change notification settings - Fork 25.9k
[MRG + 1] Bug fix for unnormalized laplacian #4995
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
Conversation
So far, _set_diag() always set the diagonal of the Laplacian to 1. This is only valid for the normalized Laplacian. For the unnormalized Laplacian, the diagonal should not be changed.
can you add a test? |
I added a test that checks w.r.t. manual dense eigendecomposition. I don't know if that's sufficient. |
LGTM |
please add an entry to whatsnew.rst |
added information for this bugfix
…into patch-1 Conflicts: doc/whats_new.rst
I added the entry to whatsnew.rst and pulled all the changes from scikit-master, such that merging should be possible. |
Thanks. We need another review, though. |
So far, _set_diag() always set the diagonal of the Laplacian to 1. This is only valid for the normalized Laplacian. For the unnormalized Laplacian, the diagonal should not be changed.
added information for this bugfix
This reverts commit b97e5b4.
…o patch-1 Conflicts: doc/whats_new.rst
…into patch-1 Conflicts: doc/whats_new.rst
@amueller rebase is done. |
LGTM. Merging without appveyor, as it is drowning from the sprint. |
[MRG + 1] Bug fix for unnormalized laplacian
+1 for focusing on regressions. |
So far, _set_diag() always set the diagonal of the Laplacian to 1. This is only valid for the normalized Laplacian. For the unnormalized Laplacian, the diagonal should not be changed.