Skip to content
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

Toplogical metrics #50

Closed
Tracked by #49
barneydobson opened this issue Feb 28, 2024 · 6 comments · Fixed by #74
Closed
Tracked by #49

Toplogical metrics #50

barneydobson opened this issue Feb 28, 2024 · 6 comments · Fixed by #74
Assignees
Labels
sa_paper Sensitivity analysis paper

Comments

@barneydobson
Copy link
Collaborator

barneydobson commented Feb 28, 2024

There are a range of topological metrics to compare the pipe layout of different networks. I have reviewed some of these and will add them here soonest - @cheginit I think has also calculated these in his busn-estimator.

@barneydobson barneydobson added the sa_paper Sensitivity analysis paper label Feb 28, 2024
@barneydobson
Copy link
Collaborator Author

From Chahinian:
image

@barneydobson
Copy link
Collaborator Author

From Duque:
image

@barneydobson
Copy link
Collaborator Author

From Blumensaat:

image

@barneydobson
Copy link
Collaborator Author

From Chegini:

image

@barneydobson
Copy link
Collaborator Author

Other metrics could be graph edit distance rooted at the outlet (see #53 )

@barneydobson
Copy link
Collaborator Author

..can I just check that comparing the distribution of nx.betweenness_centrality of two graphs via a KS test is actually a semi reasonable thing to do?

That's a loaded question 😄

First, regarding computing BC, networkx can be very slow (computing BC is computationally expensive in general), that's why I use networkit. Second, for comparing graphs in the context of optimization, there are more suitable metrics that we can choose from. For example, there is an interesting discussion here. You can also check out the distance measures or s-metric in networkx. It appears that there's a new backend for networkx that speeds up some slow operations in networkx, called graphblas

Originally posted by @cheginit in #59 (comment)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
sa_paper Sensitivity analysis paper
Projects
None yet
Development

Successfully merging a pull request may close this issue.

1 participant