I'm almost sure this is intended, but it caught me by surprise. We have that min(a.min(), b.min()) != np.min([a, b]) when (a, b) are masked arrays, but I expected them to be the same. 1.21.4 3.8.1 ...

There is a difference between numpy and numba. Numpy returns an array of np.int32 when np.cumsum is applied to an array of np.int32, whereas numba returns an array of np.int64. (Apologies, I'm not ...