Hi!

I use graph_tool.topology.shortest_distance for an all pairs shortest path calculation, what is the main run-time footprint of my algorithm and way to large.

How would you speed it up / tackle this?

I tried to sub-sample with manual source, target pairs but that's terribly inefficient since it does not use the internal bookkeeping.

Nice would be graph_tool.topology.shortest_distance(G, U, V), where U and V are lists of same length with sources / targets but that's not implemented.

Thank a lot in advance!

Stephan