-
Notifications
You must be signed in to change notification settings - Fork 171
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
cloudpickle v1.5.0 breaks TensorFlow Probability #390
Comments
3 tasks
CloudPickler should be imported from the top level package: from cloudpickle import CloudPickler |
Thank you for the correction! |
@pierreglaser we might need to implement the hack you suggested to implement a Python 3.6 compatible |
Yes most probably. |
matthewfeickert
added a commit
to scikit-hep/pyhf
that referenced
this issue
Jul 2, 2020
* Explicitly disallow cloudpickle v1.5.0 to avoid breaking TensorFlow Probability - This is a temporary solution to unblock development, but should be removed once TensorFlow Probability v0.11 has been released - tensorflow/probability#991 - cloudpipe/cloudpickle#390
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi. I'm just reporting that
cloudpickle
v1.5.0
currently breaks TensorFlow Probabilityv0.10.0
entirely. This is more of a heads up as you'll probably be getting some other reports.This is currenlty detailed in the following Issues:
pyhf
Issue 913A minimal failing example is found in
pyhf
Issue 913 but also reproduced here:This is obviously already known to the
cloudpickle
devs, but for anyone else, as can be seen above this fails becauseis invalid and should be
from cloudpickle.cloudpickle_fast import CloudPickler
(edit)
The text was updated successfully, but these errors were encountered: