NetworkX (Python):如何通过指定规则改变边缘的权重

I have a weighted graph:

我有一个加权图

F=nx.path_graph(10)
G=nx.Graph()
for (u, v) in F.edges():
    G.add_edge(u,v,weight=1)

get the nodes list:

得到节点列表:

[(0, 1), (1, 2), (2, 3), (3, 4), (4, 5), (5, 6), (6, 7), (7, 8), (8, 9)]

I want to change each edge's weight by this rule:

我想通过以下规则来改变每条边的重量:

remove one node, such as node 5, clearly, edge (4, 5) and (5, 6) will be delete, and the weight of each edge will turn to:

删除一个节点，如节点5，显然，删除边(4,5)和(5,6)，各边的权值变为:

{# these edges are nearby the deleted edge (4, 5) and (5, 6)

{#这些边在被删除的边(4,5)和(5,6)附近

(3,4):'weight'=1.1,

(3、4):“重量”= 1.1,

(6,7):'weight'=1.1,

(6、7):“重量”= 1.1,

#these edges are nearby the edges above mentioned

#这些边在上面提到的边附近

(2,3):'weight'=1.2,

(2、3):“重量”= 1.2,

(7,8):'weight'=1.2,

(7、8):“重量”= 1.2,

#these edges are nearby the edges above mentioned

#这些边在上面提到的边附近

(1,2):'weight'=1.3,

(1、2):“重量”= 1.3,

(8,9):'weight'=1.3,

(8、9):“重量”= 1.3,

# this edge is nearby (1,2)

#这条边在(1,2)附近

(0,1):'weight'=1.4}

(0,1):“重量”= 1.4 }

How to write this algorithm?

如何编写这个算法?

PS: path_graph is just an example. I need a program suit any graph type. Furthermore, the program need to be iterable, it means I can remove one node from origin graph each time.

path_graph只是一个示例。我需要一个适合任何图形类型的程序。此外，程序需要是可迭代的，这意味着我可以每次从原点图中删除一个节点。

Regards

问候

1 个解决方案

#1

You can access the edge weight as G[u][v]['weight'] or by iterating over the edge data. So you can e.g.

您可以以G[u][v]['weight']的形式访问边缘权值，或者通过迭代边缘数据。所以你可以如。

In [1]: import networkx as nx

In [2]: G=nx.DiGraph()

In [3]: G.add_edge(1,2,weight=10)

In [4]: G.add_edge(2,3,weight=20)

In [5]: G[2][3]['weight']
Out[5]: 20

In [6]: G[2][3]['weight']=200

In [7]: G[2][3]['weight']
Out[7]: 200

In [8]: G.edges(data=True)
Out[8]: [(1, 2, {'weight': 10}), (2, 3, {'weight': 200})]

In [9]: for u,v,d in G.edges(data=True):
   ...:     d['weight']+=7
   ...:     
   ...:     

In [10]: G.edges(data=True)
Out[10]: [(1, 2, {'weight': 17}), (2, 3, {'weight': 207})]