从 CSV 文件中的邻接矩阵绘制 NetworkX 图

Question

I have been battling with this problem for a little bit now, I know this is very simple - but I have little experience with Python or NetworkX. My question is very simple, I am trying to plot a large dataset (about 200 rows/columns) of a matrix that looks like this. The first row and first column are identical.

  A,B,C,D,E,F,G,H,I,J,K
A,0,1,1,0,1,1,1,1,0,1,0
B,1,0,0,0,1,1,1,1,0,1,0
C,1,0,0,0,1,1,1,1,0,1,0

It just a matrix showing how people are connected, and all I want is to import and plot this csv file, with it's corresponding labels in NetworkX.

I have this file ( people.cs v), and looking at previous answers here , it seems the best way to do this is by putting the data in an array with numpy.

There seems to be a problem with this:

import numpy as np
import networkx as nx
import matplotlib.pyplot as plt
from numpy import genfromtxt
import numpy as np

mydata = genfromtxt('mouse.csv', delimiter=',')

I get the following output:

File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/numpy/lib/npyio.py", line 1272, in genfromtxt
  fhd = iter(np.lib._datasource.open(fname, 'rbU'))
File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/numpy/lib/_datasource.py", line 145, in open
  return ds.open(path, mode)
File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/numpy/lib/_datasource.py", line 472, in open
  found = self._findfile(path)
File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/numpy/lib/_datasource.py", line 323, in _findfile
  if self.exists(name):
File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/numpy/lib/_datasource.py", line 417, in exists
  from urllib2 import urlopen
File "/System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/urllib2.py", line 94, in <module>
  import httplib
File "/System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/httplib.py", line 69, in <module>
  from array import array
      File "/Users/Plosslab/Documents/PythonStuff/array.py", line 4, in <module>
      NameError: name 'array' is not defined

Answer 1

I made a small csv called mycsv.csv that has the following:

,a,b,c,d
a,0,1,0,1
b,1,0,1,0
c,0,1,0,1
d,1,0,1,0

You don't have a ',' as the first character on the first row, but instead you have a space, so if this is an error on my part let me know. The general idea will be the same. Read in the csv as such:

from numpy import genfromtxt
import numpy as np
mydata = genfromtxt('mycsv.csv', delimiter=',')
print(mydata)
print(type(mydata))

This prints:

[[ nan  nan  nan  nan  nan]
 [ nan   0.   1.   0.   1.]
 [ nan   1.   0.   1.   0.]
 [ nan   0.   1.   0.   1.]
 [ nan   1.   0.   1.   0.]]
<type 'numpy.ndarray'>

Now that we have the csv read in as a numpy array we need to extract just the adjacency matrix:

adjacency = mydata[1:,1:]
print(adjacency)

This prints:

[[ 0.  1.  0.  1.]
 [ 1.  0.  1.  0.]
 [ 0.  1.  0.  1.]
 [ 1.  0.  1.  0.]]

You can just slice your numpy array as needed if my small example isn't exactly as yours.

To plot the graph you will need to import matplotlib and networkx:

import matplotlib.pyplot as plt
import networkx as nx

def show_graph_with_labels(adjacency_matrix, mylabels):
    rows, cols = np.where(adjacency_matrix == 1)
    edges = zip(rows.tolist(), cols.tolist())
    gr = nx.Graph()
    gr.add_edges_from(edges)
    nx.draw(gr, node_size=500, labels=mylabels, with_labels=True)
    plt.show()

show_graph_with_labels(adjacency, make_label_dict(get_labels('mycsv.csv')))

Here's a short tutorial on graphs with python.

Answer 2

This can be done easily by using pandas and networkx .

For example, I have created a small csv file called test.csv as

A,B,C,D,E,F,G,H,I,J,K
A,0,1,1,0,1,1,1,1,0,1,0
B,1,0,0,0,1,1,1,1,0,1,0
C,1,0,0,0,1,1,1,1,0,1,0
D,0,0,0,0,1,0,1,1,0,1,0
E,1,0,0,0,1,1,1,1,0,1,0
F,0,0,1,0,1,0,0,0,0,1,0
G,1,0,0,0,0,0,0,1,0,0,0
H,1,0,0,0,1,1,1,0,0,1,0
I,0,0,0,1,0,0,0,0,0,0,0
J,1,0,0,0,1,1,1,1,0,1,0
K,1,0,0,0,1,0,1,0,0,1,0

You can read this csv file and create graph as follows

import pandas as pd
import networkx as nx
input_data = pd.read_csv('test.csv', index_col=0)
G = nx.DiGraph(input_data.values)

For plotting this graph use

nx.draw(G)

You would be getting a plot something similar to this.

Answer 3

This is identical to Scott's excellent answer but handles correctly nodes without edges.

import matplotlib.pyplot as plt
import networkx as nx

def show_graph_with_labels(adjacency_matrix, mylabels):
    rows, cols = np.where(adjacency_matrix == 1)
    edges = zip(rows.tolist(), cols.tolist())
    gr = nx.Graph()
    all_rows = range(0, adjacency_matrix.shape[0])
    for n in all_rows:
        gr.add_node(n)
    gr.add_edges_from(edges)
    nx.draw(gr, node_size=900, labels=mylabels, with_labels=True)
    plt.show()