CODEWITHSUNDEEP
pythonpandasnetworkx
LdM
LdM

Reputation: 704

Network from pandas dataframe with colours

I would need to create a network using the following data

String        List1                           
string1 [string3, string2, string20, string4]   
string2 [string100, string1, string4]
string3 [string1, string80, string2]
string4 [string13, string28, string12, string16]
string5 [string3, string8, string12, string6]
string6 []

with indirect links which shows links among the variables/nodes. So, for example, string1 is linked with string3, string2, string30, string4. String2 is linked with string100, string1 and string4, and so on.

I have tried to use Networkx

import itertools.combinations as comb

edges = set()
for col in df:
    for _, data in df.groupby(col):
        edges.update(comb(data.index, 2))

G = nx.Graph()
G.add_nodes_from(df.index)
G.add_edges_from(edges)

but it does not do what I would expect as output.

Upvotes: 0

Views: 93

Answers (1)

warped
warped

Reputation: 9481

indeed you should use df.explode. networkxhas a function that you can feed the exploded df into directly:

G = nx.from_pandas_edgelist(df.explode('List1'), source='String', target='List1')

That said, the way you construct the dataframe (i.e. lists as cell entries) causes a lot of headaches, so it is best avoided.

Upvotes: 2

Related Questions