Prototyping a Faster Optuna Implementation in Rust #5362
HideakiImamura
announced in
Announcements
Replies: 2 comments 1 reply
-
That looks very exciting! I am currently using optuna in a context where the tpesampler takes approx 10x the time needed to evaluate the objective. The rust implementation would make a lot of sense to use for me Is it possible to have access to some beta? |
Beta Was this translation helpful? Give feedback.
0 replies
-
Thanks, is it the package showcased in medium? I can’t see the tpe sampler
or the python binding
Alexis
Le lun. 27 mai 2024 à 17:12, Kento Nozawa ***@***.***> a
écrit :
… https://github.com/optuna/rustuna
—
Reply to this email directly, view it on GitHub
<#5362 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AEU46W5UINFY6QV2QVTPUBDZENLN3AVCNFSM6AAAAABFKFA5D2VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TKNZSGEYTK>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
We’ve started prototyping the faster Optuna implementation in Rust, featuring Python, JavaScript, and C bindings.
Masashi Shibata (@c-bata) explains the motivation and the potential benefits that a Rust-based Optuna implementation could provide. 🚀
https://medium.com/optuna/prototyping-a-faster-optuna-implementation-in-rust-e76efba3761b
Beta Was this translation helpful? Give feedback.
All reactions