Re: Bayesian optimization add-inThank you @yuichi_katsumur very useful add-in
Re: Bayesian optimization add-inNice work @yuichi_katsumur ! @summers1985 , although there is not currently automatic integration with the XGBoost add-in, it is straightforward to run both add-ins manually. If you run the XGBoost Add-In with the Autotune option, it will create a table of fitting results over a fast flexible filling design. Make sure the performance metric you want is included in the table, save it to a JMP table, then use this table as input to the Bayesian Optimizataion add-in. This lets you profile the current results and suggest potentially improved hyperparameter settings. Iterate between the two add-ins as needed to find ranges of optimal hyperparameters. I tend to prefer a manual approach like this as it let's you see what is happening across the whole hyperparameter space at each step.
Re: Bayesian optimization add-inIs it possible to specify the range and conditions of experimental points added from the search results?
For example, if you want to optimize the combination of mixing ratios of materials A to C using Bayesian optimization, you would like to set the ratio range of A to C and the sum of A to C.
Re: Bayesian optimization add-inHello @Mae-Tetsu , Thank you for your question. Please try to set mixture column properties before performing the Bayesian optimization.
Please see the following page for detailed information for column properties.
https://www.jmp.com/support/help/ja/16.2/#page/jmp/mixture-example.shtml#ww529344
Re: Bayesian optimization add-inHi
Please provide me with a source for an article or book on the subject of Bayesian optimization
Regards
Re: Bayesian optimization add-inHI @Raaed,
Thank you for your comment. Here are some review articles I found;
S. Greenhill, S. Rana, S. Gupta, P. Vellanki and S. Venkatesh, "Bayesian Optimization for Adaptive Experimental Design: A Review," in IEEE Access, vol. 8, pp. 13937-13948, 2020, doi: 10.1109/ACCESS.2020.2966228.
B. Shahriari, K. Swersky, Z. Wang, R. P. Adams and N. de Freitas, "Taking the Human Out of the Loop: A Review of Bayesian Optimization," in Proceedings of the IEEE, vol. 104, no. 1, pp. 148-175, Jan. 2016, doi: 10.1109/JPROC.2015.2494218.
Re: Bayesian optimization add-inThank you very much
Re: Bayesian optimization add-inCan I create a plugin by linking the bayes method and xgboost?
Re: Bayesian optimization add-in@summers1985 Thank you for your comment. Unfortunately, the current version does not have the ability to work with the XGBoost add-in. If there are many requests, I would like to consider creating an add-in.
Bayesian optimization add-in
About this add-in
This add-in is an add-in for performing Bayesian optimization using JMP's Gaussian process model.
System requirements
- JMP 15 and later
- Display language is English and Japanese only
About Bayesian optimization
Bayesian optimization is a method used to repeatedly perform sequential experiments to find the optimal solution. The main flow of Bayesian optimization is as follows.
- Create a Space Filling plan (or other plan) with an appropriate number of experiments.
- do an experiment.
- A Gaussian process model is estimated from the obtained data.
- From the Gaussian process model g(x) (and its covariance matrix ∑) estimated from the data, find the point x* where A(g(x)) calculated from an appropriate function A(.) is maximum.
- Add x* as an experimental point and return to step ii.
For more information on Bayesian optimization, please refer to the following papers.
B. Shahriari, K. Swersky, Z. Wang, R. P. Adams and N. de Freitas, "Taking the Human Out of the Loop: A Review of Bayesian Optimization," in Proceedings of the IEEE, vol. 104, no. 1, pp. 148-175, Jan. 2016, doi: 10.1109/JPROC.2015.2494218.
How to use add-ins
(Please also see attached document)
- Open the JMP data table that contains the experimental data you have already conducted. *Note: The current add-in only supports one objective variable Y. Also, explanatory variable X can only be specified on a continuous scale.
- Select Add-ins > Bayesian Optimization.
- Configure columns and options.
- Once the analysis is complete, a results window will appear. If you want to add candidate experimental points to the table, click the Add row to table button.
- If you want to perform the next experiment, you can perform Bayesian optimization under the same conditions by running the table script "Run Bayesian Optimization Add-in".
screenshot
How to use this add-in
This add-in allows you to perform Bayesian optimization using Gaussian process model in JMP.
Bayesian optimization is one technique used when repeatedly performing sequential experiments and searching for an optimal solution.
Before conducting a real experiment, it is highly recommended to check whether it works well with some pseudo data. There are cases where it is impossible to find the optimal point well with few experiments.
- Open the JMP data table where you entered the experimental data you have already done.
- Select Add-Ins > Bayesian Optimization.
- Set columns and options.
- When the analysis is finished, a window will be displayed. If you want to add candidate experimental points to the table, click the “Add the row to table” button.
- If you want to repeat the Bayesian optimization, click “Run Bayesian Optimization Add-in” in the table script.
For more information, please see the attached pdf file.
change history
29 September, 2023 Version 1.3: Added a checkbox to the launch window to add nugget effects. Fixed the variance calculation formula to be correct when tertiary structure is specified.
07 October, 2023 Version 1.3.1: Fixed issue when decimal point is comma
08 December, 2023 Version 1.3.2: Fixed an issue where an error was displayed when there was a combination factor.
Design of ExperimentsRe: Bayesian optimization add-inHello,
I would like to use the Bayesian optimization add-in for the design of mixing, but is there any way to do that?
Re: Bayesian optimization add-inHello @KOKI , Thank you for your question. Please try to set the mixture column properties before running the Bayesian optimization. If you have any further questions, please let me know.
Re: Bayesian optimization add-in@yuichi_katsumur
Thank you for your answer! I've been able to understand your suggestion by actually running Bayesian add-in with a sample data set. The issue has been resolved.
Re: Bayesian optimization add-inHow do I know the Rsquare after finding candidate experiment points with Bayesian Optimization add-in?
Re: Bayesian optimization add-inHello @ThomasBaah ,
Thank you for your question. This add-in uses the Gaussian process in JMP. One measure of goodness of fit is how well the points lie along the 45-degree diagonal line in the actual versus predicted plot. JMP does not report the R2 statistic for the Gaussian process. For information on the Actual vs. Predicted plot, see the following page.
https://www.jmp.com/support/help/en/18.0/#page/jmp/actual-by-predicted-plot-2.shtml#
Re: Bayesian optimization add-in@yuichi_katsumur Thank you!
Is there any better way to extract plots in bayesian optimization add-in? I'm bit new to JMP
Re: Bayesian optimization add-in@ThomasBaah You can use either of the following methods.
https://www.jmp.com/support/help/en/18.0/#page/jmp/copy-and-paste-reports.shtml#
https://www.jmp.com/support/help/en/18.0/#page/jmp/save-reports-as-powerpoint-presentations.shtml#
