I'll do these modifications and submit a PR.
Best,
François.
2015-05-04 12:36 GMT+02:00 Tiago de Paula Peixoto ti...@skewed.de:
On 03.05.2015 13:18, François wrote:
Hello,
The shortest_path and shortest_distance functions are fixed to be able
to handle more than one target. In order to
Hello,
The shortest_path and shortest_distance functions are fixed to be able to
handle more than one target. In order to be be more explicit, I propose for
those two functions to return a dictionary when there is more than one
target. This dictionary would be keyed by target vertex. What do you
Hello,
I realize that my C++ skill aren't sufficient to produce quality /
maintainable / efficient code for this feature. Would you take care of this
?
I've update my github repo https://github.com/Fkawala/gcloud-python,
it compiles but does not work.
The current error stack is:
Traceback
Hello,
I did try to compile my fork, however it fails in a file that I didn't
modified: graph_blockmodel_covariates.
The error I get is below.
graph_blockmodel_covariates.cc:695:16: error: 'dense_hash_map' was not
declared in this scope
typedef vectordense_hash_mapsize_t, size_t,
On 19.04.2015 20:40, François Kawala wrote:
I did try to compile my fork, however it fails in a file that I didn't
modified: graph_blockmodel_covariates.
This is a bug when sparse_hash is not enabled. I have fixed it now in
git.
Best,
Tiago
--
Tiago de Paula Peixoto ti...@skewed.de
On 15.04.2015 01:14, François Kawala wrote:
About your comment on line 196 (which probably also applies to lines
238 and 274), what is proper way to receive a set from python ?
You have to expose the class to Python using boost::python:
On 15.04.2015 00:27, François wrote:
I forked the grap_tool github repo, and tried to update the grpah_distance.cc.
The result is visible there
https://github.com/Fkawala/graph-tool/blob/master/src/graph/topology/graph_distance.cc.
However,
I'm not familiar with CPP, thus the result might
What would you recommend ?
I would implement a new class called djk_max_multiple_targets_visitor that
would be used in get_dists only when there are more than one target. Does
it sounds good?
F
2015-04-13 20:32 GMT+02:00 Tiago Peixoto [via Main discussion list for the
graph-tool project]
On 13.04.2015 20:23, François Kawala wrote:
I guess that I should update accordingly the *get_dists *function.
Should I submit a pull request once that done ?
Sure, but be careful not to compromise the performance in the case where
there is only a single target.
Best,
Tiago
--
Tiago de
My first guess is to generalize the djk_max_visitor
https://git.skewed.de/count0/graph-tool/blob/master/src/graph/topology/graph_distance.cc#L78
as
described in this [gist
https://gist.github.com/Fkawala/6c0a40320d5036e6324e], does it sound a
good starting point ?
F.
2015-04-12 18:17 GMT+02:00
Wrong gist link, sorry, the good one is :
https://gist.github.com/Fkawala/6c0a40320d5036e6324e
2015-04-12 19:18 GMT+02:00 François Kawala francois.kaw...@gmail.com:
My first guess is to generalize the djk_max_visitor
11 matches
Mail list logo
