I just downloaded your script. The courses at the top are still the same. I also did reparametrize in my own script at the two locations as mentioned by you but still getting the same old result.
I’m wondering how would reparametrize solve the issue as the evaluation location of the contour curves would remain exactly the same for the computation in the entire script.
I guess there’s some issue with how the data structure is being managed in the script. Isn’t there a way to check the orange error bubble and find the exact spot where the issue is coming?
How did you tackle the issue ? I mean how did you check the script to reach to the conclusion that there was issue with the masks ? What was your thinking process like?
You compared outputs of any particular components by reading and analyzing their data ?
I used the Tree Statistics component at various outputs. One thing it does is give the number of paths, which allowed me to see the problem coming out of your Split Tree, where there were 28 paths instead of 29 and realized it was the mask.