If you open your graph and set the value of the random shuffle to 8, and the increasing weight to 55, what do you see in the outputs? My guess: null.
Somehow that combination of values has introduced a null along your logic stream, and so the only outputs you’re getting are *. You should do a quick ‘cross product’ study to ensure that the bulk of your outputs aren’t going to return null, and consider adding an ‘if null, return big scary value that is the worst value you could ever get + 1’ to your evaluation metrics, which will prevent GD form thinking the occasional null is good due to a low score elsewhere.
Other things to check:
- Have you remembered the right data?
- Do you have any python nodes calling an external application which GD doesn’t have access to (ie: Revit API requests, or RSA), or code which is attempting to leverage multiple cores (something GD does automatically).