Hello to you all. I am facing an issue with a case switch node:
Trying to perform a Time Series analysis I want to perform differencing or detrending of the signal, before model it with different techniques. Depending on a variable the workflow executes differencing or detrending algorithms. When detrending, the inactive Branch Works perfectly and propagates the inactive ports till the end case of the workflow. But when differencig, the inactive Branch is not able to propagate the inactive ports through the metanodes and components located in between, so the final end case cannot handle the data that comes from the active brach.
It seems that there is a problem with some components to propagate the inactive ports. After analysing more carefully looks like it is due another IF SWITCH node anidated inside an “Decompose Signal” metanode.
¿Any clue / workaround?
Thanks in advance.
Welcome to the KNIME Forum! Which version of KNIME are you using? I remember such behavior being fixed a couple of months ago.
Can you share a minimal workflow with me that highlights the problem? Maybe we can get to the bottom of this.
Sorry for the delay, I was abroad and I couldn’t answer. Investigating more closely into the subject looks like the inactivity propagation in the Branch stops when it finds a “Decompose Signal” component, the same one we can download from the Knime website. Anyway I rejected the case solution because even when the workflow took the the Branch where the inactivity propagated correctly at the end of the case always got stoped, not executing the following nodes.
Anyway i found a workaround by concatenating two loops.
Thank you very much for your response.
This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.