cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Discussions

Solve problems, and share tips and tricks with other JMP users.
Choose Language Hide Translation Bar
matteo_patelmo
Level IV

Bayesian Optimization factor range setting

I would like to use Bayesian Optimization starting from 2 data points.  The factors for these data points (historical data) span only a very limited range, but I would like to expand the exploration to a wider range for some of the factors. 

It seems that BO by default is stuck with the initial explored range. Is there a way to set the range to be explored, so that I can go outside the initially explored hypercube?

thanks
Matteo

1 ACCEPTED SOLUTION

Accepted Solutions

Re: Bayesian Optimization factor range setting

Hi @matteo_patelmo,

The range that BayesOpt is set to explore is agnostic of the actual entries you've used - you just need to go on each column for each factor and set 'Coding' to say the limits that you want explored (see screenshot below).

 

Ben_BarrIngh_0-1763048213807.png

Thanks!

Ben

“All models are wrong, but some are useful”

View solution in original post

4 REPLIES 4

Re: Bayesian Optimization factor range setting

Hi @matteo_patelmo,

The range that BayesOpt is set to explore is agnostic of the actual entries you've used - you just need to go on each column for each factor and set 'Coding' to say the limits that you want explored (see screenshot below).

 

Ben_BarrIngh_0-1763048213807.png

Thanks!

Ben

“All models are wrong, but some are useful”
matteo_patelmo
Level IV

Re: Bayesian Optimization factor range setting

Thanks for the instantaneous reply. 

Matteo

 

Re: Bayesian Optimization factor range setting

You can also modify the factor range using the Profilers on the Batch Selection tab.

Victor_G
Super User

Re: Bayesian Optimization factor range setting

Hi @matteo_patelmo,

JMP DoE and Bayesian Optimization platforms use similar metadata from the datatable to make the analysis or design creation easier: Factors column properties (Design Role, Coding, Factor Changes, Mixture, etc...), Response Limits, and Constraint script added to the datatable as soon as you have specified any constraint and created the DoE datatable.

The Bayesian Optimization platform use many of these metadata to work.
You can see some explanations here How do I set up a constraint when using Bayesian Optimization? 

Hope this complementary answer may help you, 

Victor GUILLER

"It is not unusual for a well-designed experiment to analyze itself" (Box, Hunter and Hunter)

Recommended Articles