Hello all,
I am trying to replicate the Goal Seek Solver functionality in Excel. I am trying to change 3 different variables. Those variables will interact with a data set and produce a result. I am trying to optimize this so that the result is as close to zero as possible. I can work with formulas to set up the scenario, but I am not sure how to get Knime to run hundreds of test scenarios to create the solution to test and repeat.
I hope this makes sense.
I think you’re going to need to provide more context about your specific scenario. KNIME isn’t Excel, and the way that one approaches these kinds of situations needs to be adjusted accordingly.
It would also be helpful if you provided some example input data, some example output data, and whatever workflow you’ve been working on.
I suspect you’ll need to use a Parameter Optimization Loop:
I was worried you would ask me to give an example. It’s a little hard to put into words. Here is an approximation of what I want:
Imagine the below data set:
|A|Q|
|B|R|
|C|S|
|D|T|
Now imagine I have the variable X which is the ratio of A:B:C:D (e.g. A= B * X, B= C * X) ,etc.
I am able to adjust the first value (A) and the ratio (X). Q,R,S.T are given. My goal is to come up with the values A and X to minimize the difference between the two colunms (i.e. the sum of (A-Q),(B-R),(C-S),(D-T).
Does that make sense?
Yes, I saw this, but I can’t make heads or tails of it. It seems like he is trying to solve for a different kind of issue, but I am not positive.
This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.
