It’s 44 threads (22 cores). They also compare with tensorflow cpu compiled with SIMD instruction sets on the same hardware. What other optimizations would you

It’s 44 threads (22 cores). They also compare with tensorflow cpu compiled with SIMD instruction sets on the same hardware. What other optimizations would you like to see? I would expect the tensorflow team to already pay pretty close attention to performance in cpu and gpu implementations, not to mention CUDNN and such...