Mask specific columns of a numpy array

Question

I have a 2D numpy array A of (60,1000) dimensions.
Say, I have a variable idx=array([3,72,403, 512, 698]).

Now, I want to mask all the elements in the columns specified in idx. The values in these columns might appear in other columns but they shouldn't be masked.

Any help would be appriciated.

Would creating a new array that is a copy of the original with the idx columns excluded something that would solve your problem? — JonathanV
– JonathanV, Commented Jan 30, 2013 at 18:37
@JonathanV, Not exactly, because I need to keep track of indices in subsequent calculations. If I modify the array indexing, it would be difficult to track. — jitendra
– jitendra, Commented Jan 30, 2013 at 18:38
@NPE, I am asking about masking in numpy.ma library. What I basically need is a Null value. — jitendra
– jitendra, Commented Jan 30, 2013 at 18:41

NPE · Accepted Answer · 2013-01-30 18:48:07Z

In [22]: A = np.random.rand(5, 10)

In [23]: idx = np.array([1, 3, 5])

In [24]: m = np.zeros_like(A)

In [25]: m[:,idx] = 1

In [26]: Am = np.ma.masked_array(A, m)

In [27]: Am
Out[27]: 
masked_array(data =
 [[0.680447483547 -- 0.290757600047 -- 0.0718559525615 -- 0.334352145502
  0.0861242618662 0.527068091963 0.136280743038]
 [0.729374999214 -- 0.76026650048 -- 0.656082247985 -- 0.492464543871
  0.903026937193 0.0792660503403 0.892132409419]
 [0.0845266821684 -- 0.838838594048 -- 0.396344231382 -- 0.703748703373
  0.380441396691 0.010521007806 0.344945867845]
 [0.7501401585 -- 0.0685427000113 -- 0.587100320511 -- 0.780160645327
  0.276328587928 0.0665949459004 0.604174142611]
 [0.599926798275 -- 0.686378805503 -- 0.776940069716 -- 0.0452833614622
  0.598622591094 0.942843765543 0.528082379918]],
             mask =
 [[False  True False  True False  True False False False False]
 [False  True False  True False  True False False False False]
 [False  True False  True False  True False False False False]
 [False  True False  True False  True False False False False]
 [False  True False  True False  True False False False False]],
       fill_value = 1e+20)

Collectives™ on Stack Overflow

Mask specific columns of a numpy array

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related