Find vertices along maximum cost path in directed graph

Question

I have a weighted directed graph like this:

I want to find the list of vertices along the maximum cost path from start to finish.

In this example I should get:

'enableBrowserExtension' -> 'newWindow' -> 'newTab' -> 'startPage' -> 'typed' -> 'selectTab' -> 'clickTextField' -> 'changeField-frullatore' -> 'clickButton-VAI' -> 'submit' -> 'formSubmit' -> 'mouseClick' -> 'link'

I'm trying with this code found here:

import networkx as nx

def inverse_weight(graph, weight='weight'):
    copy_graph = graph.copy()
    for n, eds in copy_graph.adjacency():
        for ed, eattr in eds.items():
            copy_graph[n][ed][weight] = eattr[weight] * -1
    return copy_graph

def longest_path_and_length(graph, s, t, weight='weight'):
    i_w_graph = inverse_weight(graph, weight)
    path = nx.dijkstra_path(i_w_graph, s, t)
    length = nx.dijkstra_path_length(i_w_graph, s, t) * -1
    return path, length

if __name__ == '__main__':
    DG = nx.DiGraph()
    DG.add_edge('enableBrowserExtension', 'enableBrowserExtension', weight=3)
    DG.add_edge('enableBrowserExtension', 'newWindow', weight=1)
    DG.add_edge('newWindow', 'newTab', weight=1)
    DG.add_edge('newTab', 'startPage', weight=1)
    DG.add_edge('startPage', 'typed', weight=1)
    DG.add_edge('typed', 'selectTab', weight=2)
    DG.add_edge('selectTab', 'newTab', weight=1)
    DG.add_edge('newTab', 'typed', weight=1)
    DG.add_edge('typed', 'typed', weight=1)
    DG.add_edge('selectTab', 'clickTextField', weight=2)
    DG.add_edge('clickTextField', 'changeField', weight=1)
    DG.add_edge('changeField', 'clickButton', weight=1)
    DG.add_edge('clickButton', 'submit', weight=1)
    DG.add_edge('submit', 'formSubmit', weight=1)
    DG.add_edge('formSubmit', 'mouseClick', weight=1)
    DG.add_edge('mouseClick', 'link', weight=2)
    DG.add_edge('link', 'formSubmit', weight=1)
    DG.add_edge('formSubmit', 'selectTab', weight=1)
    DG.add_edge('selectTab', 'mouseClick', weight=1)
    DG.add_edge('mouseClick', 'clickLink', weight=1)
    DG.add_edge('clickLink', 'link', weight=1)
    DG.add_edge('link', 'mouseClick', weight=2)
    DG.add_edge('mouseClick', 'changeField', weight=1)
    DG.add_edge('changeField', 'mouseClick', weight=1)
    DG.add_edge('mouseClick', 'selectTab', weight=1)

    path, length = longest_path_and_length(DG, 'enableBrowserExtension', 'link')

But I get the error:

ValueError: ('Contradictory paths found:', 'negative weights?')

I also tried this java code but it only returns the total maximum cost as integer, I want the name of the vertices along the path.

Is there a way to fix this?

score 1 · Answer 1 · answered Mar 12 '20 at 22:26

1

Since you have positive-value loops in the graph, the maximum-cost path is infinite. You need to remove the trivial loops and finesse the others. Perhaps you have an unimplemented requirement that no node is visited more than once?

The various ways of infinite looping are why you get the "contradictory paths".

answered Mar 12 '20 at 22:26

Prune

76,765
14
60
81

Thanks for your reply, do you have a code example I can run? I tried to remove trivial loops like `('typed', 'typed')` but I still get the same error. – Ford1892 Mar 12 '20 at 23:01
No, I don't; that's not what Stack Overflow is for. However, if you post a [minimal, reproducible example](https://stackoverflow.com/help/minimal-reproducible-example), we'll be happy to fix a *specific* sticking point in *your* code. "I tried to fix it, but that didn't work" isn't specific. Also, your original graph is notably larger than "minimum". – Prune Mar 12 '20 at 23:03

Find vertices along maximum cost path in directed graph

1 Answers1