-
Notifications
You must be signed in to change notification settings - Fork 107
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Vf2 vs vf2pp #89
base: main
Are you sure you want to change the base?
Vf2 vs vf2pp #89
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I finally had a chance to look at this. Very cool!
I have a couple of comments below. Nice job.
``` | ||
|
||
To do this same experiment on DiGraphs, we need to define an in-degree sequence and an out-degree sequence. | ||
We will use the function *nx.directed_configuration_model(in_degree_sequence, out_degree_sequence, create_using, seed)*. Both degree sequences must have the same sum. Again this function returns a MultiDiGraph so we will have to convert it into a DiGraph. Let's do this in an example: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm surprised that the resulting degree sequences are the same for the two graphs. The multiedges are not likely to be the same -- and the selfloop edges are also likely to be a different number of edges with the two seeds. So when you convert to a non-multigraph, and remove the selfloops, you change the degree sequences. And it seems like you would be changing them in different ways. Have you checked that the two graphs have the same degree sequence? If they don't, the function will correctly return True
, but it won't actually run any of the algorithm to find this out. I think it doesn't even create the data structures that would cost time for overhead in VF2++, but I'm not sure about that.
Co-authored-by: Dan Schult <[email protected]>
An initial draft of benchmarking of vf2 vs vf2++. There are still some details to fix in the plots (mainly x labels and some titles) and many TODOs.