equal
deleted
inserted
replaced
100 /// |
100 /// |
101 /// \brief %Preflow algorithm class. |
101 /// \brief %Preflow algorithm class. |
102 /// |
102 /// |
103 /// This class provides an implementation of Goldberg-Tarjan's \e preflow |
103 /// This class provides an implementation of Goldberg-Tarjan's \e preflow |
104 /// \e push-relabel algorithm producing a \ref max_flow |
104 /// \e push-relabel algorithm producing a \ref max_flow |
105 /// "flow of maximum value" in a digraph. |
105 /// "flow of maximum value" in a digraph \ref clrs01algorithms, |
|
106 /// \ref amo93networkflows, \ref goldberg88newapproach. |
106 /// The preflow algorithms are the fastest known maximum |
107 /// The preflow algorithms are the fastest known maximum |
107 /// flow algorithms. The current implementation uses a mixture of the |
108 /// flow algorithms. The current implementation uses a mixture of the |
108 /// \e "highest label" and the \e "bound decrease" heuristics. |
109 /// \e "highest label" and the \e "bound decrease" heuristics. |
109 /// The worst case time complexity of the algorithm is \f$O(n^2\sqrt{e})\f$. |
110 /// The worst case time complexity of the algorithm is \f$O(n^2\sqrt{e})\f$. |
110 /// |
111 /// |