ehh in contradiction to what I said earlier ~ vectorized operations provide a giant speed up. ~ http://i.imgur.com/Oo92T.png -> i'm using numpy arrays in both cases. How about having the declarations be more simple and then packing / unpacking only for the math portion? Also you could split this into chunks + do it in parallel to get some additional speed boosts. On 17 December 2012 09:15, Guillaume Gay <guillaume@mitotic-machine.org>wrote:
Le 17/12/2012 15:14, Tiago de Paula Peixoto a écrit :
On 12/17/2012 11:59 AM, Guillaume Gay wrote:
Is there a way to do the following differently:
edge_x = g.new_edge_property() edge_x.a = np.array([x[e.source()] for e in g.edges()])
with x being a vector property map. I plain English, is there a (more efficient) way to copy the vertex property map of every edge source to an edge property map?
Something like this is not yet available, you have to use a python loop for now.
However I plan to add something like this at some point, since it occurs often enough. If you wish, you may open a ticket, and I'll address when time permits.
Cheers, Tiago
_______________________________________________ graph-tool mailing listgraph-tool@skewed.dehttp://lists.skewed.de/mailman/listinfo/graph-tool
Ok I'll do that, thanks!
G.
_______________________________________________ graph-tool mailing list graph-tool@skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool