ITK filter methods that want a sequence of ints get confused when passed a sequence of np.int64s

[HastingsGreer]

Description

When you index into a numpy array and think you got an integer, you actually get a numpy.int64. python ints and numpy.int64 are almost indistinguishable, but If you pass a sequence of numpy.int64 s to an ITK method that expects a sequence of integers, it throws an inscrutable error. This is unpleasant to debug.

Steps to Reproduce

import itk
import numpy as np

crop_amounts = np.array([1, 2, 3])
crop_amounts = [i for i in crop_amounts]

print(crop_amounts)
# [1, 2, 3]

filter = itk.CropImageFilter[itk.Image[itk.F, 3], itk.Image[itk.F, 3]].New()
filter.SetLowerBoundaryCropSize(crop_amounts)

Expected behavior

sets the lower boundary

Actual behavior

Traceback (most recent call last):
  File "<stdin>", line 8, in <module>
ValueError: Expecting a sequence of int (or long)

Reproducibility

100%

Versions

any

Environment

any

Additional Information