Skip to content
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

Add matrix profile as an optional import. #985

Merged
merged 2 commits into from
Dec 21, 2022

Conversation

bvanelli
Copy link
Contributor

There are many issues regarding matrixprofile not being supported anymore:

#972
#937
#952
#932

These are all related to the outdated dependency matrixprofile.

This draft MR introduces matrix profile as an optional dependency. It also makes sure many users can run this code in newer versions of python or different cpus.

Could any maintainer take a look?

@bvanelli bvanelli marked this pull request as ready for review November 21, 2022 18:57
@bvanelli bvanelli changed the title WIP: Add matrix profile as an optional import. Add matrix profile as an optional import. Nov 21, 2022
@michaelosthege
Copy link

ping @kempa-liehr is there any ETA for a new release with Python 3.10 / 3.11 support?

It looks like this PR might get the tsfresh package much closer to that.

@arturdaraujo
Copy link

arturdaraujo commented Dec 18, 2022

This PR seems pretty good to me. Matrixprofile is the main package causing dependency problems, so after this PR the package will work with most packages on python 3.9 and 3.10.

what do you think? @kempa-liehr

@arturdaraujo
Copy link

What the mantainers of package think? This PR will let tsfresh support 3.9 and 3.11 as @michaelosthege said.
This PR is just waiting to be approved
@nils-braun @kempa-liehr @MaxBenChrist @kempa-liehr @dbarbier @earthgecko

@arturdaraujo
Copy link

@erezinman can you check this PR? I think you were talking about something like this, right?

@kempa-liehr sorry for being annoying, I just don't want to see tsfresh become obsolete as it is the best time series feature extraction out there.

@arturdaraujo
Copy link

@kempa-liehr Please have a look at this

@kempa-liehr kempa-liehr merged commit dc9c7ff into blue-yonder:main Dec 21, 2022
@kempa-liehr
Copy link
Collaborator

@bvanelli, @michaelosthege, @aiwalter, @PJPRoche, @williamgilpin, @arturdaraujo, @enrique-wesper, @0liu, @vanbenschoten, @CyrilLeMat, @erezinman, @Luke12062, @antonindurieux

Thanks for the discussion, solutions, and encouraging feedback! Please accept my apologies for letting this problem dangle for so long. We are currently rebuilding the maintenance team and are confident to support the community more efficiently!

Cheers,
Andreas

@bvanelli
Copy link
Contributor Author

Whoa, that is nice. Thanks for taking the time everyone. My solution for now excludes the feature completely but since this got merged I might consider in the future to extend mp with stomp, please ping me in an issue if someone finds problems with the current implementation.

Wish everyone merry Christmas and happy new year!

@aiwalter
Copy link

would it be possible to make a new tsfresh release with this changes?

@kempa-liehr
Copy link
Collaborator

@aiwalter, v0.20.0 has now been released: https://github.com/blue-yonder/tsfresh/releases/tag/v0.20.0

Happy New Year!

@arturdaraujo
Copy link

Don't forget about releasing on conda-forge! Anyway, what a day! Thanks @kempa-liehr

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.

5 participants