What inspired this wish list request?
I have been working on a project where I need to fit a specific nonlinear model (that is not available in Fit Curve) to hundreds of different curve data sets. Each curve needs to be fit separately, not jointly. This is not easy to do in the Nonlinear platform
What is the improvement you would like to see?
Similar to what is in Fit Curve, allow for specification column into a group role that identifies each separate curve in the data table. Then allow for fitting of each curve separately in an efficient and compact way. This would include making it easy to interrogate the model fits, find situation where the model fit did not converge for specific curves and allow for manual intervention, saving the prediction formula for both individual fitted curves and for all curves (in a single column, using the Group role column in a conditional).
Why is this idea important?
During development of a pharmaceutical drug product, we generate hundreds to thousands of dissolution curves on which we want to do parametric modeling of those curves with models that are not available in Fit Curve. Being able to do this quickly and efficiently is very important. For example, on a current project I have 10000 curves that need to be fit, and with my JSL scripting approach, it is taking hours to run.