Hi,

> As you said, there does not seem to be easy choices other then automatising
> at the filesystem level (saving hundreds of graphs from R, processing
> hundreds of graphs in python while writing them, and reading the hundreds of
> graphs from R again. Oh well.
Well, there is one more possibility. If your graph does not have attributes,
you can simply convert it into an edgelist using get.edgelist in the R side,
and then you can re-construct the graph on the Python side from the edge
list. I don't know NetworkX well enough but I'm 100% sure that it has some
function for creating graphs from numeric edge lists. This way you can avoid
the overhead of saving the graphs to disk.

-- 
T.

_______________________________________________
igraph-help mailing list
[email protected]
https://lists.nongnu.org/mailman/listinfo/igraph-help

Reply via email to