Ross Barnowski
Ross Barnowski
I am unable to reproduce this error from the reported example on `master`. What version of NetworkX are you using? Also, the traceback reports an `IndexError` - it's not immediately...
Ok, I've tried this one again and now I *am* able to reproduce it with `pydot==1.4.2`. It turns out the length of the name is not the only factor though,...
I actually think that the current behavior may have use-cases in the wild, and don't think it should be changed. There may be cases where 0 is a meaningful value...
I confirm that I can reproduce this on `main` (7fe275fd)
Is this issue still relevant? @neocsbee could you provide more specific information from the paper that is at odds with the current implementation?
I agree this would be a really nice addition! Re: API I'm not sure what's best, though I do think it makes sense for the "layers" to be lists instead...
@MridulS does the need for non-local go away if you use a namedtuple instead of a dataclass? If so that might be a nice alternative
This is interesting and IIUC it boils down to the question: how are directed edges treated during conversion to an *undirected* multigraph? Should each *directed* edge between two nodes be...
> What if we provide a keep_multiple_directed_edges parameter that defaults to False, where keep_multiple_directed_edges==True gives the behavior @nlevhari was expecting? That's certainly one approach - the most important thing IMO...
> If you like, I can give you an example graph for which your steinter_tree function returns not a tree but a graph with cycles. @cw299 please do - a...