harold.matrix_slice

harold.matrix_slice(M, corner_shape, corner='nw')

Takes a two dimensional array M and slices into four parts dictated by the corner_shape and the corner string corner.

    m   n
p [ A | B ]
  [-------]
q [ C | D ]

If the given corner and the shape is the whole array then the remaining arrays are returned as empty arrays, numpy.array([]).

Parameters:

• M (ndarray) – 2D input matrix
• corner_shape (tuple) – An integer valued 2-tuple for the shape of the corner
• corner (str) – Defines which corner should be used to start slicing. Possible options are the compass abbreviations: 'nw', 'ne', 'sw', 'se'. The default is the north-west corner.

Returns:

• A (ndarray) – Upper left corner slice
• B (ndarray) – Upper right corner slice
• C (ndarray) – Lower left corner slice
• D (ndarray) – Lower right corner slice

Examples

>>> A = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
>>> matrix_slice(A,(1,1))
(array([[1]]),
 array([[2, 3]]),
 array([[4],
        [7]]),
 array([[5, 6],
        [8, 9]])
)
>>> matrix_slice(A, (2,2), 'sw')
(array([[1, 2]]),
 array([[3]]),
 array([[4, 5],
        [7, 8]]),
 array([[6],
        [9]])
 )
>>> matrix_slice(A, (0, 0))  % empty A
(array([], shape=(0, 0), dtype=int32),
 array([], shape=(0, 3), dtype=int32),
 array([], shape=(3, 0), dtype=int32),
 array([[1, 2, 3],
        [4, 5, 6],
        [7, 8, 9]]))