Closed topepo closed 2 years ago
It turns out that kernlab gives different results for this iterative tuning on M1 compared to Intel.
We should try to create a reprex demonstrating this. 🥵
This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.
The Bayesian optimization and simulated annealing search methods start to produce different results. The only differences in packages were that kernlab changed from 0.9-29 to 0.9-30. The changes to kernlab seem minimal but produce different results.
For BO, the last two iterations changed.
For SA, the new best is at iteration 40 instead of 27.
There are no changes needed to the automatically generated results.