How does keras tuner count objective? #581
Unanswered
GustawOhler
asked this question in
Q&A
Replies: 1 comment
-
|
The best score of a trial is the mean of the best epoch of all executions. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
When I tuned my model with Bayesian tuner it seemed to me that it takes the best value across all executions. It didn't fit my needs so I changed to Random tuner and now it seems that it takes some sort of average but I don't know if it's from best scores from all executions or all epochs of all executions. So - like in title - how does the keras tuner count objective for comparing trials?
Beta Was this translation helpful? Give feedback.
All reactions