Possible rounding issue in AveragePooling2DLayer

[serenegrace]

Upon attempting to write a test for AveragePooling2DLayer, I have encountered the following flaw:

inputVol(:, :, 1) = [0 0 2 0 0; 1 2 0 2 0; 0 0 2 2 0; 0 2 2 2 2; 2 2 2 1 1]; % channel 1 input matrix
L = AveragePooling2DLayer([3 3], [2 2], [0 0 0 0]);
y = L.evaluate(inputVol);
a=y(1, 1, 1);
b=sum(inputVol(1:3, 1:3, 1), 'all')/9;
The values are extremely close, but not actually the same. Notably, a9=6.9999999 (etc), while b9=7.

I suspect that this level of precision is not relevant usually, but I thought I should mention it all the same