I am currently building a curve-fitting analysis program in KNIME, where I use R View to manually correct abnormal curves (based on certain parameters obtained from the fitting process). However, the issue is that every time I run the program, it processes all the curves (which could be several hundred, and out of those, over a hundred may need manual correction). The process cannot be interrupted. Sometimes, I would like to pause after processing a portion and resume the next time I have more time. Additionally, if I make a mistake during manual correction, I would like the ability to modify only the specific curve, while keeping the others unchanged. How can I implement these features?
Hi,
That sounds like a really interesting task! Could you share your workflow?
As far as I know, it is difficult or almost impossible in KNIME to pause a loop, edit parts within the loop, and then continue. So, it might be better to split up the tasks.
I would let the loop run through and store the metrics you use to decide if the fit is good in an additional table (use two or more output ports at the loop end). Right after the loop, you can split the data and redo the manual fitting with the (hopefully) smaller dataset.
1 Like