CODEWITHSUNDEEP
python-3.xnetworkx
Trenton McKinney
Trenton McKinney

Reputation: 62393

Why don't edges added with G.add_edges_from(), match G.edges()?

Scenario

Code

import networkx as nx

edges = [(1, 3), (1, 4), (1, 5), (1, 6), (1, 7), (1, 8), (1, 9), (1, 10), (1, 11), (1, 12), (1, 13), (1, 14), (1, 15), (1, 16), (1, 17), (1, 18), (1, 19), (1, 20), (1, 21), (1, 22), (1, 23), (1, 24), (1, 25), (1, 26), (1, 27), (1, 28), (1, 29), (1, 30), (1, 31), (1, 32), (1, 33), (1, 34), (1, 35), (1, 36), (1, 37), (1, 38), (1, 39), (1, 40), (1, 41), (1, 42), (1, 43), (1, 44), (1, 45), (1, 46), (1, 47), (1, 48), (1, 49), (16, 18), (16, 35), (16, 36), (16, 48), (18, 16), (18, 24), (18, 35), (18, 36), (19, 5), (19, 8), (19, 11), (19, 13), (19, 15), (19, 17), (19, 20), (19, 21), (19, 24), (19, 30), (19, 31), (19, 35), (19, 36), (19, 37), (19, 48), (28, 1), (28, 5), (28, 7), (28, 8), (28, 11), (28, 14), (28, 15), (28, 17), (28, 20), (28, 21), (28, 24), (28, 25), (28, 27), (28, 29), (28, 30), (28, 31), (28, 35), (28, 36), (28, 37), (28, 44), (28, 48), (28, 49), (36, 5), (36, 24), (36, 35), (36, 37), (37, 24), (37, 35), (37, 36), (39, 1), (39, 24), (39, 33), (39, 35), (39, 36), (39, 38), (39, 40), (39, 41), (39, 45), (42, 1), (43, 24), (43, 29), (43, 35), (43, 36), (43, 37), (43, 47), (43, 48), (45, 1), (45, 39), (45, 41)]

print(len(edges))
>>> 119

# unique index 0 values from each edges tuple
print(set([x[0] for x in edges]))
>>> {1, 16, 18, 19, 28, 36, 37, 39, 42, 43, 45}

# create empty graph
G = nx.Graph()

# add edges
G.add_edges_from(edges)

# get graph edges
ge = G.edges()

# length of ge
print(len(ge))
>>> 112

# unique index 0 values from each ge tuple
print(set([x[0] for x in ge]))
>>>{1, 5, 7, 8, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 24, 25, 27, 28, 29, 33, 35, 36, 37, 38, 39, 41, 43}

# output the edges of G
print(ge)
>>> EdgeView([(1, 3), (1, 4), (1, 5), (1, 6), (1, 7), (1, 8), (1, 9), (1, 10), (1, 11), (1, 12), (1, 13), (1, 14), (1, 15), (1, 16), (1, 17), (1, 18), (1, 19), (1, 20), (1, 21), (1, 22), (1, 23), (1, 24), (1, 25), (1, 26), (1, 27), (1, 28), (1, 29), (1, 30), (1, 31), (1, 32), (1, 33), (1, 34), (1, 35), (1, 36), (1, 37), (1, 38), (1, 39), (1, 40), (1, 41), (1, 42), (1, 43), (1, 44), (1, 45), (1, 46), (1, 47), (1, 48), (1, 49), (5, 19), (5, 28), (5, 36), (7, 28), (8, 19), (8, 28), (11, 19), (11, 28), (13, 19), (14, 28), (15, 19), (15, 28), (16, 18), (16, 35), (16, 36), (16, 48), (17, 19), (17, 28), (18, 24), (18, 35), (18, 36), (19, 20), (19, 21), (19, 24), (19, 30), (19, 31), (19, 35), (19, 36), (19, 37), (19, 48), (20, 28), (21, 28), (24, 28), (24, 36), (24, 37), (24, 39), (24, 43), (25, 28), (27, 28), (28, 29), (28, 30), (28, 31), (28, 35), (28, 36), (28, 37), (28, 44), (28, 48), (28, 49), (29, 43), (33, 39), (35, 36), (35, 37), (35, 39), (35, 43), (36, 37), (36, 39), (36, 43), (37, 43), (38, 39), (39, 40), (39, 41), (39, 45), (41, 45), (43, 47), (43, 48)])

Expectations and questions

  1. edges and ge should have the same length
    • Why don't they?
  2. edges and ge should have the same node combinations
    • Why don't they?
  3. Are my expectations incorrect, and if so, why?
  4. What, if anything, can be done to make ge and edges match?

Upvotes: 3

Views: 466

Answers (1)

yatu
yatu

Reputation: 88236

This is because you're generating an undirected graph, and some of the edges in the edge list are the same just in reversed order. You can check this by sorting the sublists, and building a set from the result:

len(set(tuple(sorted(i)) for i in edges))
# 112

As you can see you have the same amount of unique combinations of nodes as edges in the graph.

If you generated a directed graph instead, you'd get as you expect, since in this case order does matter:

G = nx.DiGraph()
G.add_edges_from(edges)

len(G.edges())
# 119

Upvotes: 4

Related Questions