Replacing one element with another in Python

Question

I would like to replace inf with 0 in the matrix, P . The desired output is attached.

import numpy as np
P = np.array([-1.54511316e+12-inf, -1.54511316e+12-inf,-inf,inf,inf])

The desired output is:

array([-1.54511316e+12, 0, -1.54511316e+12, 0, 0, 0, 0])

Answer 1

You can combine numpy.where and numpy.isfinite :

P2 = np.where(np.isfinite(P), P, 0)

output:

array([-1.54511316e+12,  0.00000000e+00, -1.54511316e+12,  0.00000000e+00,
        0.00000000e+00,  0.00000000e+00,  0.00000000e+00])

Or, for in place modification:

P[~np.isfinite(P)] = 0

Answer 2

As an alternative way, if being sure that we have only inf s (not nan s), there is another NumPy tool np.isinf :

P[np.isinf(P)] = 0

# [-1.54511316e+12  0.00000000e+00 -1.54511316e+12  0.00000000e+00 
#                   0.00000000e+00  0.00000000e+00  0.00000000e+00]

which is straight forward (exactly for np.inf s) and don't need to use logical not ~ .