You can use non-bipartite terms in a GOF for a bipartite fit (e.g., degree, geodesics). It would be helpful to state explicitly how these non-bp statistics are calculated in the bp context.
If it's as simple as the bp property of the graph is ignored, and both modes are treated as part of an undifferentiated network, I'm happy to add that to the docs.
If it's more complicated, then someone else will need to do this.
You can use non-bipartite terms in a GOF for a bipartite fit (e.g., degree, geodesics). It would be helpful to state explicitly how these non-bp statistics are calculated in the bp context.
If it's as simple as the bp property of the graph is ignored, and both modes are treated as part of an undifferentiated network, I'm happy to add that to the docs.
If it's more complicated, then someone else will need to do this.