Assign to masked numpy array without removing mask?

Question

I would like to assign to (a slice of) a masked numpy array but not modify the mask. (Assignment normally clears the mask (unless it is "hard"), which seems completely contrary to the point of masking, but that's what we've got to work with.) I would also like this routine to work for plain unmasked arrays.

Is there a better way to do this than saving and restoring the mask?

a = np.ma.array([0, 1, 2], mask=[0, 1, 0])
mask = a.mask.copy() if np.ma.is_masked(a) else None        # Have to copy because it might be shared
a[a < 2] = -1
if mask is not None:
    a.mask = mask
print(a, a.data)
# [-1 -- 2] [-1 -1 2]

This is Python 2, numpy 1.11.1.

Alok Singhal · Accepted Answer · 2016-07-27 22:20:12Z

3

I think what you want can be done by:

a.data[a < 2] = -1

answered Jul 27, 2016 at 22:20

Alok Singhal

96.8k21 gold badges130 silver badges158 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Doctor J · Accepted Answer · 2016-07-27 22:14:46Z

1

In researching the question, I found an answer:

np.copyto(a, -1, where=a < 2)

answered Jul 27, 2016 at 22:14

Doctor J

6,3726 gold badges48 silver badges41 bronze badges

Collectives™ on Stack Overflow

Assign to masked numpy array without removing mask?

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related