Closed guoyangqin closed 2 years ago
Hard to tell, as I'm not familiar with the details of the scipy implementation. The problem is O(n^x) where x in {3,4} depending on the implementation (see [1]). A genuine C implementation might cure any overhead that stems from looping in interpreted languages.
[1] http://www.assignmentproblems.com/doc/LSAPIntroduction.pdf
Thanks for this package, it saves me a huge amount of time.
Just wondering why is it so much faster? Is it because of a better algorithm or being based on C. If latter, why didn't scipy.optimize.linear_sum_assignment or sklearn.utils.linearassignment develop based on C? Thanks.