Gaussian Processes (moving it from Specialized to Predictive Modeling)Gaussian Processes are a very powerful modeling technique in predictive analytics.
https://scikit-learn.org/stable/modules/gaussian_process.html
While JMP has GP, it is only for specialized modeling (not possible to fit using training/validation/test)
Advanced Statistical ModelingData Exploration and VisualizationPredictive Modeling and Machine LearningRe: Gaussian Processes (moving it from Specialized to Predictive Modeling)- Kernels would be great, e.g. choosable: periodic, linear trend, noisy, smoothness-parameter and free kernels.
- Additionally it would be great if you could include the uncertainty of the input data points.
- An easy function to find the (next n) point(s) of highest variance in a specified input-range would help to find the best next experimental point(s). Of course this can by done by a script, but a menue entry for this would make it much easier for non-scripters.
Re: Gaussian Processes (moving it from Specialized to Predictive Modeling)Addition of different Kernels (matern, etc) would be highly benefitial for general modellig as well.
Re: Gaussian Processes (moving it from Specialized to Predictive Modeling)Adding the ability to analyze the curves with DOE variables (supplemental) like FDE and Fit curves would be great. If it were another model option in FDE this would be taken care of.
