Skip to content

(Kuhn-Munkres) numpy implementation, rectangular matrix is supported (|X| <= |Y|). 100x100000 in 0.153 s.

Notifications You must be signed in to change notification settings

mayorx/hungarian-algorithm

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 

Repository files navigation

hungarian algorithm

Description

  • a python numpy implementation of hungarian algorithm (also known as Kuhn–Munkres algorithm).
  • Rectangular matrix is supported.
  • 0.153 second when |X|=100 and |Y|=100,000 on my Mac.

Example

matcher = KMMatcher([
    [2., 3., 0., 3.],
    [0., 4., 4., 0.],
    [5., 6., 0., 0.],
    [0., 0., 7., 0.]
])
matcher.solve(verbose=True)

result:

match 0 to 3, weight 3.0000
match 1 to 1, weight 4.0000
match 2 to 0, weight 5.0000
match 3 to 2, weight 7.0000
ans: 19.0000

or using default setting

weights = np.random.randn(n, m)
matcher = KMMatcher(weights)
best = matcher.solve()

Performance

(only tested on my Mac)

N = |X|
M = |Y|
N M time(seconds)
10 100,000 0.017
10 1,000,000 0.250
100 10,000 0.026
100 100,000 0.153
100 1,000,000 2.004
1,000 1,000 1.634

Reference

topcoder c++ implementation

About

(Kuhn-Munkres) numpy implementation, rectangular matrix is supported (|X| <= |Y|). 100x100000 in 0.153 s.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages