I want to change the number according to the value of a specific axis.

I would like to use python's numpy to change the process according to the specific axis value.

For example, if there is a a[3][3][3] ndarray like the following,

[[0 12]
  [ 3  4  5]
  [ 6  7  8]]

 [[ 9 10 11]
  [12 13 14]
  [15 16 17]]

 [[18 19 20]
  [21 22 23]
  [24 25 26]]]

For a[:,:,:1], we would like to overwrite 10 or more to 0, for a[:,:,1:2], for a[:,:,2:], we would like to overwrite 20 or more to 0.

After examining the numpy function, I implemented it as follows and now I can handle it as intended, but I'm thinking about whether I can handle the slice smartly (without slices).

I think there are probably some appropriate functions, but is there a function that is appropriate for such a process?

import numpy as np

defmain():
    a=np.range(27).reshape(3,3,3)
    print(a)

    a0 = a [:, :, :, :1]
    a1 = a [:, :, 1:2]
    a2 = a [ : , : , 2: ]

    a0_update=np.where(a0<10,a0,0)
    a1_update=np.where(a1<15,a1,0)
    a2_update=np.where(a2<20,a2,0)

    a_update=np.dstack((a0_update, a1_update, a2_update))

    print(a_update.shape)
    print(a_update)

if__name__=="__main__":
    main()