Skip to content

Conversation

johannes-mueller
Copy link
Member

Fix #155, follow blue-yonder/tsfresh#1097

Signed-off-by: Johannes Mueller <johannes.mueller4@de.bosch.com>
@johannes-mueller
Copy link
Member Author

johannes-mueller commented Feb 20, 2025

@maierbn scipy-1.15.2 does no longer report an optimization failure for one of the four failure cases for the beta calculation. What do we do with this?

@maierbn
Copy link
Contributor

maierbn commented Feb 20, 2025

but it does for the other three? Then I would say we just keep it as it is.

@johannes-mueller
Copy link
Member Author

For P_A = 1.0 it now finds beta = -10.378464316310737 It fails for P_A = 1.0 + 1e-15. I would suggest to drop the 1.0 case or replace it with something like 1.01

Obviously the optimization behavior in scipy>=1.15 changed slightly

Signed-off-by: Johannes Mueller <johannes.mueller4@de.bosch.com>
@johannes-mueller johannes-mueller merged commit 7d45355 into develop Feb 24, 2025
14 checks passed
@johannes-mueller johannes-mueller deleted the uplift-scipy-restriction branch February 24, 2025 08:17
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Dependency fix needed
2 participants