filtering lines in a numpy array according to values in a range

问题

Let an array:

 a =np.array([[1,2],[3,-5],[6,-15],[10,7]])

to get lines with elements of the second column above -6 it' s possible to do

>>> a[a[:,1]>-6]
array([[ 1,  2],
   [ 3, -5],
   [10,  7]])

but how to get lines with the second element between -6;3? I tried:

>>> a[3>a[:,1]>-6]

and also (which raises an error):

>>> np.ma.masked_inside(a,-6,3)

which gives:

 masked_array(data =
 [[-- --]
  [-- --]
  [6 -15]
  [10 7]],
         mask =
  [[ True  True]
  [ True  True]
  [False False]
  [False False]],
   fill_value = 999999)

but the result is not too clear

Thanks jp

回答1:

>>> a[ (3>a[:,1]) & (a[:,1]>-6) ]

array([[ 1,  2],
      [ 3, -5]])

回答2:

The np.ma.masked_inside(a, -6, 3) will create a MaskedArray object, where the values between -6 and 3 are masked (that is, flagged as invalid). In other terms, you're filtering out the values between -6 and 3. Instead, you should use np.ma.masked_outside(a, -6, 3):

>>> a = np.array([[1,2],[3,-5],[6,-15],[10,2]])
>>> np.ma.masked_outside(a,-6,3)
>>> masked_array(data =
 [[1 2]
 [3 -5]
 [-- --]
 [-- 2]],
             mask =
 [[False False]
 [False False]
 [ True  True]
 [ True False]],
       fill_value = 999999)

Note that with this function, you are filtering out the whole array, element by element, which is not what you want.

The indexing approach given in another solution is by far the most straightforward and understandable.

来源：https://stackoverflow.com/questions/13066249/filtering-lines-in-a-numpy-array-according-to-values-in-a-range