min_path_error.py

import flowpaths as fp
import networkx as nx

def main():
    # Create a simple graph
    graph = nx.DiGraph()
    graph.graph["id"] = "simple_graph"
    graph.add_edge("0", "a", flow=6, length=2)
    graph.add_edge("0", "b", flow=7, length=4)
    graph.add_edge("a", "b", flow=2, length=6)
    graph.add_edge("a", "c", flow=5, length=3)
    graph.add_edge("b", "c", flow=9, length=1)
    graph.add_edge("c", "d", flow=6, length=8)
    graph.add_edge("c", "1", flow=7, length=9)
    graph.add_edge("d", "1", flow=6, length=4)

    # We create a Minimum Path Error solver with default settings, 
    # by specifying that the flow value of each edge is in the attribute `flow` of the edges,
    # and that the number of paths to consider is 3.
    mpe_model = fp.kMinPathError(graph, flow_attr="flow", k=3, weight_type=float)

    # We solve it
    mpe_model.solve()

    # We process its solution
    process_solution(mpe_model)

    edges_to_ignore = [("a", "c")]
    # We solve again, by telling the model to ignore the edges in `edges_to_ignore`
    # when computing the path slacks (i.e. edge errors)
    mpe_model_2 = fp.kMinPathError(graph, flow_attr="flow", k=3, weight_type=int, elements_to_ignore=edges_to_ignore, length_attr="length")
    mpe_model_2.solve()
    process_solution(mpe_model_2)

    edges_to_ignore = [("a", "c")]
    # We solve again, by telling the model to ignore the edges in `edges_to_ignore`
    # when computing the path slacks (i.e. edge errors)
    mpe_model_3 = fp.kMinPathError(graph, flow_attr="flow", k=3, weight_type=int, elements_to_ignore=edges_to_ignore)
    mpe_model_3.solve()
    process_solution(mpe_model_3)

def process_solution(model: fp.kMinPathError):
    if model.is_solved():
        print(model.get_solution())
        print(model.solve_statistics)
        print("model.is_valid_solution()", model.is_valid_solution())
    else:
        print("Model could not be solved.")

if __name__ == "__main__":
    main()