Is there a head and tail method for Numpy array?

For a head-like function you can just slice the array using dataset[:10].

For a tail-like function you can just slice the array using dataset[-10:].

You can do this for any python iterable.

PEP-3132 which is in python 3.x (https://www.python.org/dev/peps/pep-3132/) can use the * symbol for the 'rest' of the iterable.

To do what you want:

>>> import numpy as np
>>> np.array((1,2,3))
array([1, 2, 3])
>>> head, *tail = np.array((1,2,3))
>>> head
1
>>> tail
[2, 3]

This works well:

def nparray_tail(x: np.array, n:int):
    """
    Returns tail N elements of array.
    :param x: Numpy array.
    :param n: N elements to return on end.
    :return: Last N elements of array.
    """
    if n == 0:
        return x[0:0]  # Corner case: x[-0:] will return the entire array but tail(0) should return an empty array.
    else:
        return x[-n:]  # Normal case: last N elements of array.

Discussion

As a bonus, this fixes a non-intuitive corner case in the answer from @feedMe: dataset[-0:] returns the entire array, not an empty array as one would expect when requesting the last 0 elements on the tail end of the array. This is consistent with the .tail() function in Pandas.