RProp MLP Learner

I’m not sure why this doesn’t work. I am trying to build a model to optimize a neural net model by adjusting the number of nodes per layer and the number of layers. The attached workflow is a simplified example of what I am trying to do, but it does not work. The Flow variable is not being used correctly. I saw another post on this topic, but unfortunately I have not been able to fix my model.
Also, I do not see a Flow Variable categoriy in RProp for the number of nodes per layer. Is there a way to create a custom Flow Variable?


NeuralNetsIris.knwf (20.4 KB)

Hello @acito,

to use a flow variable you need to select the flow variabel via the drop down menu. See screenshot below.

If you add a value behind the dropdown, a new flow variable is created, using the value defined in the flow variable tab as variable name and the selected setting in the configuration window as variable value.

To control the number of nodes per layer you can use the option nrhiddenneurons in the Flow Variabel tab.


Thanks very much for the helpful and quick reply.

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.