Unexpected result with += on NumPy arrays

Question

I am creating symmetric matrices/arrays in Python with NumPy, using a standard method:

x = rand(500,500)
x = (x+x.T)
all(x==x.T)
> True

Answer 1

The implementation detail mentioned by others is called buffering. You can read more about it in the docs on array iteration.

If you look at your failing example in a little more detail:

>>> a = np.random.rand(91, 91)
>>> a += a.T
>>> a[:5, -1]
array([ 0.83818399,  1.06489316,  1.23675312,  0.00379798,  1.08967428])
>>> a[-1, :5]
array([ 0.83818399,  1.06489316,  1.75091827,  0.00416305,  1.76315071])

So the first wrong value is 90*91 + 2 = 8192, which, unsurprisingly, is what we get from:

>>> np.getbufsize()
8192

And we can also set it higher, and then:

>>> np.setbufsize(16384)  # Must be a multiple of 16
8192  # returns the previous buffer size
>>> a = np.random.rand(91, 91)
>>> a += a.T
>>> np.all(a == a.T)
True

Although now:

>>> a = np.random.rand(129, 129)
>>> a += a.T
>>> np.all(a == a.T)
False

This is of course a dangerous thing to rely on for correctness, as it is indeed an implementation detail subject to change.