Networkx: Raising the labels above the node using position

Question

I am very new to networkx (just started today!): I am using these two links and replicating:

This is for creating the network

and This is for how I tried adjusting the label positions

So mine looks like this:

layout = nx.spring_layout(g,k=0.2,iterations=50)
for l in layout:  # raise text positions
    layout[l][1] += 0.5

I adjusted the 0.5 to smaller and even larger values, but nothing happens, no adjustment. No change at all. What am I doing wrong?

The two codes combined looks like this:

import matplotlib.pyplot as plt
import pandas as pd
import networkx as nx
import os

plt.figure(figsize=(12, 12))

df = pd.read_csv(df_path)

# 1. Create the graph
g = nx.from_pandas_edgelist(df, source='name', target='club') 
# 2. Create a layout for our nodes 
layout = nx.spring_layout(g,k=0.2,iterations=50)
for l in layout:  # raise text positions
    layout[l][1] += 0.5
# 3. Draw the parts we want
nx.draw_networkx_edges(g, layout, edge_color='#AAAAAA')

clubs = [node for node in g.nodes() if node in df.club.unique()]
size = [g.degree(node) * 80 for node in g.nodes() if node in df.club.unique()]
nx.draw_networkx_nodes(g, layout, nodelist=clubs, node_size=size, node_color='lightblue')

people = [node for node in g.nodes() if node in df.name.unique()]
nx.draw_networkx_nodes(g, layout, nodelist=people, node_size=100, node_color='#AAAAAA')

high_degree_people = [node for node in g.nodes() if node in df.name.unique() and g.degree(node) > 1]
nx.draw_networkx_nodes(g, layout, nodelist=high_degree_people, node_size=100, node_color='#fc8d62')

club_dict = dict(zip(clubs, clubs))
nx.draw_networkx_labels(g, layout, labels=club_dict)

# 4. Turn off the axis because I know you don't want it
plt.axis('off')

plt.title("Revolutionary Clubs")

Thank you so much in advance!!

and on a side note does anyone have a great tutorial on networkx? I have been googling and I haven't found much. And if you know of networkx tutorials that show how to build interactive networks that would be even better!

Answer 1

You first need to draw the graph and then add the value (or create a second variable for the position of the labels). If you read again the code for positioning the labels , you will see that they first draw the graph and then modify the layout and draw the labels.

Your code simply moves everything, ie labels and edges along the y axis. I've corrected the position of the adjustment in your code:

import matplotlib.pyplot as plt
import pandas as pd
import networkx as nx
import os

plt.figure(figsize=(12, 12))

df = pd.read_csv(df_path)

# 1. Create the graph
g = nx.from_pandas_edgelist(df, source='name', target='club') 
# 2. Create a layout for our nodes 
layout = nx.spring_layout(g,k=0.2,iterations=50)
#
# ----- removed correction
#

# 3. Draw the parts we want
nx.draw_networkx_edges(g, layout, edge_color='#AAAAAA')

clubs = [node for node in g.nodes() if node in df.club.unique()]
size = [g.degree(node) * 80 for node in g.nodes() if node in df.club.unique()]
nx.draw_networkx_nodes(g, layout, nodelist=clubs, node_size=size, node_color='lightblue')

people = [node for node in g.nodes() if node in df.name.unique()]
nx.draw_networkx_nodes(g, layout, nodelist=people, node_size=100, node_color='#AAAAAA')

high_degree_people = [node for node in g.nodes() if node in df.name.unique() and g.degree(node) > 1]
nx.draw_networkx_nodes(g, layout, nodelist=high_degree_people, node_size=100, node_color='#fc8d62')

club_dict = dict(zip(clubs, clubs))

# ------> and move it here
for l in layout:  # raise text positions
    layout[l][1] += 0.1  # probably small value enough
nx.draw_networkx_labels(g, layout, labels=club_dict)

# 4. Turn off the axis because I know you don't want it
plt.axis('off')

plt.title("Revolutionary Clubs")