-
Notifications
You must be signed in to change notification settings - Fork 10
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
NLopt zero-order algorithms are using Zygote #160
Comments
Should be fixed in the latest release of NonconvexNLopt JuliaRegistries/General#87164. |
Check again in half an hour, update NonconvexNLopt to 0.1.6 and try again. |
If it is still an issue, please re-open this issue |
Thanks for the quick response, but it's still failing.
and here is the error listing (I can't copy the entire error listing because it exceeds my terminal's scrollback buffer length):
|
Sorry, but I can't find the way to reopen the issue. |
Please change your algorithm to |
The supported algorithm symbols are in the documentation https://julianonconvex.github.io/Nonconvex.jl/stable/algorithms/nlopt/#Construct-an-instance. |
With |
No worries, glad it worked. I opened another issue to implement throwing a better error in cases like yours. The above error is very non-informative. |
This is my first attempt at using
Nonconvex.jl
and I'm hitting a few hiccups. My objective function is not differentiable so I'm attempting to use 0-order algorithms from NLopt:This produces the following error output:
So it appears that
Nonconvex
is trying to useZygote
to differentiate my function even though I'm using a 0-order algorithm that doesn't require derivatives. Is there a way to informNonconvex
that it shouldn't try to get derivative information?The text was updated successfully, but these errors were encountered: