Errors when reject_f4z applied in find_graphs

[ZheCai]

Hi,

When the reject_f4z option was applied in find_graphs, I encountered the following error:

find_graphs(f2_blocks, numadmix=5, outpop="Out", stop_gen=200, stop_gen2=30, reject_f4z=3) ✔ 1: score 7065.869 best score 7065.869 none tot 1 Error: Repaired names have length 4 instead of length 8.

Is there something wrong with my arguments?

Thanks!

[uqrmaie1]

There is nothing wrong with your arguments, this was just a bug that should be fixed now. The idea behind the reject_f4z option is that it's faster to test whether a graph topology is broadly consistent with the given f4-statistics patterns than to fit a graph using qpgraph(), and so it's possible to pre-filter newly proposed graph topologies.

In practice, I didn't notice any improvements from this, but it might help in some cases, which is why I left the option in there.