Open
Conversation
|
|
Two tests with expected values computed via NetworkX: - test_personalized_page_rank: all teleport weight on node 1 - test_personalized_page_rank_partial: weight split between nodes 1 and 3
phdavis1027
force-pushed
the
feat-personalized-page-rank
branch
from
b290f9c to
7d375a0
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Two tests with expected values computed via NetworkX:
What changes were proposed in this pull request?
Add a personalization parameter to
unweighted_page_rankIf
personalizationis None, then we dispatch via trait to the exact same calculations as the previous, uniform variant to avoid rounding errors from non-commutative f64 arithmetic.Why are the changes needed?
To compute personalized page rank (e.g., in fraud-detection situations)
Does this PR introduce any user-facing change? If yes is this documented?
Yes, it adds the
personalizationparameter tounweighted_pagerank. This is added to docstrings in bothpythonandalgorithms/centralityHow was this patch tested?
The unit tests added were validated by computing the same values using networkx.
Are there any further changes required?
I am quite happy to also add a weighted variant