问题
I have a 3d array, a, of shape say a.shape = (10, 10, 10)
When slicing, the dimensions are squeezed automatically i.e.
a[:,:,5].shape = (10, 10)
I'd like to preserve the number of dimensions but also ensure that the dimension that was squeezed is the one that shows 1 i.e.
a[:,:,5].shape = (10, 10, 1)
I have thought of re-casting the array and passing ndmin but that just adds the extra dimensions to the start of the shape tuple regardless of where the slice came from in the array a.
回答1:
a[:,:,[5]].shape
# (10,10,1)
a[:,:,5] is an example of basic slicing.
a[:,:,[5]] is an example of integer array indexing -- combined with basic slicing. When using integer array indexing the resultant shape is always "identical to the (broadcast) indexing array shapes". Since [5] (as an array) has shape (1,),
a[:,:,[5]] ends up having shape (10,10,1).
来源：https://stackoverflow.com/questions/2640147/preserving-the-dimensions-of-a-slice-from-a-numpy-3d-array