replace values in an array

Question

as a replacement value for another within an operation with arrays, or how to search within an array and replace a value by another

for example:

array ([[NaN, 1., 1., 1., 1., 1., 1.]
       [1., NaN, 1., 1., 1., 1., 1.]
       [1., 1., NaN, 1., 1., 1., 1.]
       [1., 1., 1., NaN, 1., 1., 1.]
       [1., 1., 1., 1., NaN, 1., 1.]
       [1., 1., 1., 1., 1., NaN, 1.]
       [1., 1., 1., 1., 1., 1., NaN]])

where it can replace NaN by 0. thanks for any response

Glorfindel · Accepted Answer · 2022-12-09 21:00:52Z

31

You could do this:

import numpy as np
x=np.array([[np.NaN, 1., 1., 1., 1., 1., 1.],[1., np.NaN, 1., 1., 1., 1., 1.],[1., 1., np.NaN, 1., 1., 1., 1.], [1., 1., 1., np.NaN, 1., 1., 1.], [1., 1., 1., 1., np.NaN, 1., 1.],[1., 1., 1., 1., 1., np.NaN, 1.], [1., 1., 1., 1., 1., 1., np.NaN]])
x[np.isnan(x)]=0

np.isnan(x) returns a boolean array which is True wherever x is NaN. x[ boolean_array ] = 0 employs fancy indexing to assign the value 0 wherever the boolean array is True.

For a great introduction to fancy indexing and much more, see also the numpybook.

edited Dec 9, 2022 at 21:00

Glorfindel

22.8k13 gold badges97 silver badges124 bronze badges

answered Nov 25, 2009 at 21:53

unutbu

886k197 gold badges1.9k silver badges1.7k bronze badges

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

1 Comment

unutbu Over a year ago

@ricardo: Let x be your numpy array.

clh · Accepted Answer · 2011-10-17 16:21:49Z

15

these days there is the special function:

a = numpy.nan_to_num(a)

answered Oct 17, 2011 at 16:21

clh

8298 silver badges11 bronze badges

4 Comments

David Poole Over a year ago

Just saved my bacon while doing in Inverse Filter. [image-processing]

dashesy Over a year ago

But this will involve a temporary variable with same type and shape as a, it will matter on large matrices.

Brian O'Donnell Over a year ago

@dashesy As of version 1.13 you can do in place: a = numpy.nan_to_num(a, copy=False) ... More information on the parameter 'copy': Whether to create a copy of x (True) or to replace values in-place (False). The in-place operation only occurs if casting to an array does not require a copy. Default is True. New in version 1.13.

BallpointBen Over a year ago

Any way to choose a fill value other than 0?

Mike T · Accepted Answer · 2017-02-28 03:21:28Z

1

Here is the example array in the question:

import numpy as np
a = np.where(np.eye(7), np.nan, 1)

You can either use numpy.where and numpy.isnan functions to create a new array b:

b = np.where(np.isnan(a), 0, a)

Or use an in-place function to directly modify the a array:

np.place(a, np.isnan(a), 0)  # returns None

answered Feb 28, 2017 at 3:21

Mike T

44.3k18 gold badges166 silver badges213 bronze badges

Collectives™ on Stack Overflow

replace values in an array

3 Answers 3

1 Comment

4 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

1 Comment

4 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related