search

experimental-design / bofire

Experimental design and (multi-objective) bayesian optimization.
https://experimental-design.github.io/bofire/
BSD 3-Clause "New" or "Revised" License
188 stars 22 forks source link

Random search with nonlinear inequality constraints #346

Closed bertiqwerty closed 7 months ago

bertiqwerty commented 7 months ago

The data model was missing the nonlinear inequality constraint

jduerholt commented 7 months ago

Just a side node, the UniversalSampler is also able to deal with NonlinearConstraints, maybe we should also integrate this one into the RandomStrategy. What do you think?

bertiqwerty commented 7 months ago

Just a side node, the UniversalSampler is also able to deal with NonlinearConstraints, maybe we should also integrate this one into the RandomStrategy. What do you think?

What is the use case?

jduerholt commented 7 months ago

What is the use case?

It uses the Spacefilling criterion from the DoE package, as it is not fitting to the DoE Strategy, Aron made a seperate strategy for it.

bertiqwerty commented 7 months ago

What is the use case?

It uses the Spacefilling criterion from the DoE package, as it is not fitting to the DoE Strategy, Aron made a seperate strategy for it.

Could also be part of Random. Would be easier to find, I guess.

jduerholt commented 7 months ago

Could also be part of Random. Would be easier to find, I guess.

Will keep it in mind, when I refactor random and the samplers in the next weeks.