Open thenealon opened 6 years ago
http://doc.sagemath.org/html/en/reference/graphs/sage/graphs/graph_generators.html has a list of generators and factories.
Concerning 2, most generators or factories will give graphs of a specific order. Do you want a function something
such that something(n)
yields all graphs of order n
that can be constructed by any of the generators?
Honestly, I'm not sure what I want. I just noticed that there are often graphs that aren't in graph objects which I expected there to be; often, these are graphs for which there is a generator. So, I was thinking about ways of putting in these 'missing' graphs. The idea in your comment would certainly be useful (at least to me), and shouldn't be too hard to write.
When I first read this, I thought you might want a way to identify that, for example, graphs.CompleteGraph(n)
is in gt for n <= 15
, but not n > 15
. We could have a list corresponding to each generator.
Maybe everytime GT loads there might be some more comment on what's available - or how to see what's available?
Currently we have a comment about Sloane and DIMACS graphs. maybe it would be useful to say a few more thinks - or at least list ways to get some more hints.
Maybe we could overload the Sage help function, and also offer a few comments by typing help(generators), help(graph_objects), etc, and list these options everytime GT loads?
In both GT.sage and Sage itself there are many graph generators, both deterministic and random
1) By definition, we're unable to include all graphs provided by a generator to graph objects. It would be useful to have a list somewhere of the generators, so that one can see which classes of graphs we may be missing in a given conjecturing run.
2) Is there a useful way to provide a utility that adds a sample of graphs from our generators? In the deterministic setting in particular, a utility that returned all generator graphs of a given order (or range of orders) might be quite useful.