import numpy as np

import matplotlib.pyplot as plt

def show_array(Z, name):

Z = np.atleast_2d(Z)

rows,cols = Z.shape

fig = plt.figure(figsize=(cols/4.,rows/4.), dpi=72)

ax = plt.subplot(111)

#plt.imshow(Z, cmap='Purples', extent=[0,cols,0,rows],

# vmin=-.2, vmax=1, interpolation='nearest', origin='upper')

plt.imshow(Z, cmap='Purples', extent=[0,cols,0,rows],

interpolation='nearest', origin='upper')

plt.xticks([]), plt.yticks([])

for pos in ['top', 'bottom', 'right', 'left']:

ax.spines[pos].set_edgecolor('k')

ax.spines[pos].set_alpha(.25)

plt.savefig('../figures/%s' % name, dpi=72)

#plt.show()

rows,cols = 5, 9

Z1 = np.linspace(0,1,rows*cols).reshape(rows,cols)

show_array(Z1, 'ops-where-before.png')

Z2 = np.where(Z1 > 0.5, 0, 1)

show_array(Z2, 'ops-where-after.png')

Z1 = np.linspace(0,1,rows*cols).reshape(rows,cols)

show_array(Z1, 'ops-maximum-before.png')

Z2 = np.maximum(Z1, 0.5)

show_array(Z2, 'ops-maximum-after.png')

Z1 = np.linspace(0,1,rows*cols).reshape(rows,cols)

show_array(Z1, 'ops-minimum-before.png')

Z2 = np.minimum(Z1, 0.5)

show_array(Z2, 'ops-minimum-after.png')

Z1 = np.linspace(0,1,rows*cols).reshape(rows,cols)

show_array(Z1, 'ops-sum-before.png')

Z2 = Z1.sum(axis=0)

show_array(Z2, 'ops-sum-after.png')