A Python 3 implementation of the Hungarian Algorithm for optimal matching in bipartite weighted graphs.

Based on the graph theory implementation in these notes combined with the matrix interpretation in these notes.

For a detailed overview, see this Jupyter notebook.

Usage

Installation: pip3 install hungarianalg

Import: from hungarianalg.alg import hungarian

Function call: result = hungarian(matrix)

Properties:

• Optimal Matching: result.match
• Revenues: result.revenues
• Row Weights: result.row_weights
• Col Weights: result.col_weights
• Total Revenue: result.revenue_sum

See example.py for a comprehensive example.