Create python network graph from 3 dataframe columns unweighted using networkx and pyvis

Question

Im working to create a python network graph utilizing the 3 pandas dataframe string columns - unweighted (python networks and pyviz)

I would like to link (nodes) the projects to country, and projects to company.

my table/dataframe is as follow;

Company	Project	Country
T-mobile	5G upgrade	Germany
T-mobile	Network upgrade	Kuwait

[Please do not upload images of code/data/errors.](//meta.stackoverflow.com/q/285551) — J_H, Feb 28 '23 at 03:29
Your question isn't very clear: What do you mean by _"link"_? — Timus, Feb 28 '23 at 12:54

score 0 · Accepted Answer · answered Feb 28 '23 at 14:37

Let's say the DataFrame containing your data is df (and it has three columns, i.e. "Company", "Project" and "Country". This would be a possible solution:

import networkx as nx
import pandas as pd

G = nx.from_pandas_edgelist(
    pd.DataFrame({
        'source': pd.concat((df['Project'], df['Project'])),
        'target': pd.concat((df['Company'], df['Country']))
    }),
)

Here are the nodes and edges of G:

>>> G.nodes
NodeView(('5G upgrade', 'T-mobile', 'Network upgrade', 'Germany', 'Kuwait'))
>>> G.edges
EdgeView([('5G upgrade', 'T-mobile'), ('5G upgrade', 'Germany'),
('T-mobile', 'Network upgrade'), ('Network upgrade', 'Kuwait')])

Create python network graph from 3 dataframe columns unweighted using networkx and pyvis

1 Answers1