-
Notifications
You must be signed in to change notification settings - Fork 36
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Failures on master #526
Comments
Actually, I re-tested now. The failures of the
|
The remaining failures are due to a non-deterministic nature of allocations, which showed up in Julia 1.9 vs 1.8 (see #530). It remains to check where the occasional additional allocation comes from, or else to relax the tests to allow deviations of +/- 1 allocations. |
Ok, we are getting there! Any ideas on what the last remaining failure is about?
|
Probably some ChainRules rule for |
AxisArrays
@test_broken
problems #546The BaseKernels failures on master can be traced back to moving ChainRules from 1.52.0 to 1.53.0. I think some weird interaction betwen ChainRules and Zygote pullbacks is at play.
The upside is that they are resolved by using JuliaStats/Distances.jl#246.
The text was updated successfully, but these errors were encountered: