I was using KernelDensity and the underlying tree_ attribute to check the shape when querying. It seems that the shape is a ndarray of shape (X.shape[:-1], k), isn't it?

Thanks for the comment and the title change.

Sorry, I don't know that (X.shape[:-1], k) is equal to X.shape[:-1] + (k,) as notation.
To be precise, (X.shape[:-1], k) is a tuple of tuple and int.
And X.shape[:-1] + (k,) is a tuple of int.
(I'm sorry if that's not what you intended.)

Example is as follows.

>>> import numpy as np
>>> from sklearn.neighbors import KDTree
>>> rng = np.random.RandomState(0)
>>> X = rng.random_sample((10, 3))
>>> tree = KDTree(X, leaf_size=2)
>>> k = 5
>>> dist, ind = tree.query(X, k=k)
>>> X.shape
(10, 3)
>>> X.shape[:-1] + k
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: can only concatenate tuple (not "int") to tuple
>>> X.shape[:-1] + (k,)
(10, 5)
>>> (X.shape[:-1], k)
((10,), 5)
>>> dist.shape
(10, 5)
>>> ind.shape
(10, 5)

My bad, my brain misread X.shape[-1] and X.shape[:-1]. Thus, it makes sense.

Thanks @Kot271828