Inconsistent results between MacOS and Linux

Hi, my framework uses the Numerics package. We developed a model using the SwiftRT API and a version using TensorFlow. On the Mac we get the exact same results. The results were recorded and put into the unit tests to assure a constant point of reference.

On Ubuntu and gLinux, TensorFlow produces the exact same results. SwiftRT using the Numerics package almost produces the same results up to 0.001 difference. I am wondering if the Numerics package is relying on system library functions which are inconsistent across platforms?

If so, this is a big problem. Any insights?

Thanks, Ed

Start with the existing unit tests that record the reference results, then reproduce the SwiftRT and TensorFlow comparisons on MacOS, Ubuntu, and gLinux. Investigate whether Numerics relies on platform-dependent system math functions; done means identifying the source of the 0.001 difference and documenting or correcting the cross-platform behavior.

Written by the indexing model from the issue text.