Suppose I have this dataframe where column a-number represents nodes with directed edges to nodes in b-number:
a_number b_number
0 343 991
1 991 633
2 343 633
3 633 628
4 343 633
5 628 916
6 697 886
7 916 572
8 697 884
9 886 125
How can I generate an image representation of this graph that looks like this:
Networkx is the go-to library for graphs in python: https://networkx.org/documentation/stable/index.html
First do the import:
import networkx as nx
To start a graph like that declare a inicialize DiGraph (directed graph):
G = nx.DiGraph()
Then add some nodes:
G.add_node(343)
G.add_node(991)
G.add_node(633)
Then some edges:
G.add_edge(343,991)
G.add_edge(991,633)
G.add_edge(343,633)
Finaly draw the graph G:
nx.draw(G, with_labels = True, font_size=14 , node_size=2000)
use the with_labels = True so you can have the node numbers, node_size=2000 to make the nodes bigger and font_size=14 to make the font also bigger
This is the output of the code:
Now to read from the dataframe, just do a cycle like:
for i, (x, y) in df.iterrows():
G.add_node(x)
G.add_node(y)
G.add_edge(x,y)
If the nodes or edges already exists it will not add a new one, so you don't need to worry about it
You can also use the graphviz library:
from graphviz import Digraph
dot = Digraph()
for i, (a, b) in df.iterrows():
dot.edge(str(a), str(b))
dot.render('graph.gv', view=True)
