Setting specific values to 0 in a NumPy array

Question

I have two arrays (gt and pred) with values ranging from 0 to 4.

The shape of these two arrays is (1, 1, 93, 349, 219). My target is to create a mask to ignore and multiplying it into gt and pred to ignore the value in two arrays. However, I am facing an issue

ignore_value=4
if ignore_value is not None:
    mask[gt!=ignore_value]=1  # ignore value mask
    gt=mask*gt
    pred=mask*pred  # ignore value mask for pred


print "after removing ignore value: ", np.unique(gt),np.unique(pred)
output: after removing ignore value:  [0 1 2 3] [0 1 2 3 4]

why it is not removing the ignore value in pred?

Other thing, is this code being called inside a function that you passed these arrays to? — cs95
– cs95, Commented Dec 23, 2018 at 8:58
@coldspeed yes, gt, pred, and ignore_value are input arguments passed to the function — S.EB
– S.EB, Commented Dec 23, 2018 at 9:00
Ah, okay. I get it. You initialised a mask based on gt, but pred may not necessarily have 4s in the same cells as gt does. Bottom line, you need two separate masks. — cs95
– cs95, Commented Dec 23, 2018 at 9:02

cs95 · Accepted Answer · 2018-12-23 09:03:47Z

1

IMO, a better solution would be to use boolean indexing and explicitly set cells to 0.

gt[gt == ignore_value] = 0
pred[pred == ignore_value] = 0

In general, it is not guaranteed that gt and pred will have ignore_value in the same cells, so using a single mask for both of them is not appropriate.

However, the code above works and is more efficient because it is only operating on a small portion of the array, not all of it (as multiplying them would do). The is the output

answered Dec 23, 2018 at 9:03

cs95

406k106 gold badges744 silver badges797 bronze badges

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

2 Comments

S.EB Over a year ago

thank you for your help, what if I write like this: gt=np.where(gt==ignore_value,0,gt) pred=np.where(pred==ignore_value,0,pred)

cs95 Over a year ago

@S.EB It will work, but will generate a new array. It is not as efficient.

Collectives™ on Stack Overflow

Setting specific values to 0 in a NumPy array

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related