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
Replicate errors found in the 3.7 env in travis #5834
Conversation
I do replicate locally. I'll try to see what is all that about |
FYI same failures are happening on maint/0.17, so we'll want to backport |
yes there has been a numpy realease and it's not compatible with scipy=1.2.0 (I think) If this PR is green I suggest to merge it, and make a PR removing those 2 lines + fixing whatever is wrong. |
We probably don't need to fix anything, just wait for the SciPy 1.2.0 + NumPy 1.16.0 wheels to land. 1.16.0 came out yesterday IIRC so it might take a day or two ... ? |
@larsoner yes, we can just do the PR to keep track of it, re run the CI in a couple of days and if its green. Done!! |
maybe there was a nicer way to rollback to 1.15.4, but .. |
Actually I get it to go away locally if I add:
to setup.cfg. This makes sense if the Can you see if this fixes things for you? If so we should add it for a day or two rather than pin the NumPy version. |
Codecov Report
@@ Coverage Diff @@
## master #5834 +/- ##
=========================================
Coverage ? 88.58%
=========================================
Files ? 369
Lines ? 69125
Branches ? 11649
=========================================
Hits ? 61232
Misses ? 5038
Partials ? 2855 |
It does fix it on my end. |
I guess that if anything is crucial, some other test would actually break, some buffer would go crazy, or something. |
I looked at the topmost error in the log, it was our suite turning a warning into an error |
This message is coming from scikit-learn apparently they are using |
Try to replicate the errors we are facing in 3.7 travis runs in the following PRs: