“Simplest” way to graphically represent a matrix

I am doing an assignment involving percolation–long story short, I have an array of arrays that represents a grid, and every entry is -1, 0, or 1. I have the assignment essentially done, but one part asks for a graphical representation of the matrix using colors for each possible entry. But from the way the assignment is worded, it sounds to me like perhaps I shouldn’t be using anything not found in standard Python 2.7 and numpy.

I couldn’t think of how to do it so I just went ahead and import pylab and plotted every coordinate as a large colored square in a scatter plot. But I just have this nagging concern that there’s a better way to do it–like there’s a better package to use for this task, or like there’s a way to do it with just numpy. Suggestions?

If it helps, my current code is below.

def show_perc(sites):
    n = sites.shape[0]
    # Blocked reps the blocked sites, etc., the 1st list holds x-coords, 2nd list holds y.
    blocked = [[],[]]
    full = [[],[]]
    vacant = [[],[]]
    # i,j are row,column, translate this to x,y coords.  Rows tell up-down etc., so needs to
    # be in 1st list, columns in 0th.  Top row 0 needs y-coord value n, row n-1 needs coord 0.
    # Column 0 needs x-coord 0, etc.
    for i in range(n):
        for j in range(n):
            if sites[i][j] > 0.1:
                vacant[0].append(j)
                vacant[1].append(n-i)
            elif sites[i][j] < -0.1:
                full[0].append(j)
                full[1].append(n-i)
            else:
                blocked[0].append(j)
                blocked[1].append(n-i)
    pl.scatter(blocked[0], blocked[1], c='black', s=30, marker='s')
    pl.scatter(full[0], full[1], c='red', s=30, marker='s')
    pl.scatter(vacant[0], vacant[1], c='white', s=30, marker='s')
    pl.axis('equal')
    pl.show()

Answer

If you’re not allowed to use other libraries, you’re stuck with ASCII art. For example:

>>> import numpy as np
>>> a = np.random.randint(-1, 2, (5,5))
>>> a
array([[ 0,  0,  1,  1,  0],
       [ 1, -1,  0, -1,  0],
       [-1,  0,  0,  1, -1],
       [ 0,  0, -1, -1,  1],
       [ 1,  0,  1, -1,  0]])
>>> 
>>> 
>>> b[a<0] = '.'
>>> b = np.empty(a.shape, dtype='S1')
>>> b[a<0] = '_'
>>> b[a==0] = ''
>>> b[a>0] = '#'
>>> b
array([['', '', '#', '#', ''],
       ['#', '_', '', '_', ''],
       ['_', '', '', '#', '_'],
       ['', '', '_', '_', '#'],
       ['#', '', '#', '_', '']], 
      dtype='|S1')
>>> 
>>> for row in b:
...     for elm in row:
...         print elm,
...     print
... 
  # # 
# _  _ 
_   # _
  _ _ #
#  # _ 
>>>     

But, if you can use matplotlib, you can just use imshow to plot your matrix:

>>> import matplotlib.pyplot as plt
>>> plt.ion()
>>> plt.imshow(a, cmap='gray', interpolation='none')

enter image description here

Leave a Reply

Your email address will not be published. Required fields are marked *