I’m David Shaw and
this is AI News. This week, we’ll take a quick
look at a deep learning code sample that you can use to
take advantage of the new Intel AVX-512 Deep Learning Boost. It demonstrates testing
the new functionality using intrinsic functions. The second generation Intel
Xeon scalable processors now include Intel
AVX512-DL Boost, which can improve the throughput
of integer linear algebra. This instruction
can accelerate loops in some convolutional
neural networks. The code sample uses
Intel AVX-512 intrinsics to illustrate the use of
both Intel AVX512-DL Boost fused instruction
and the 3 equivalent FEMA-based instructions. Follow the link
provided to access the code sample yourself. Also, you’ll find detailed
descriptions of Intel AVX-512 intrinsics in the
Intel Intrinsics Guide. Thanks for watching and
I’ll see you next week. [INTEL LOGO]

Intel® AVX512 Deep Learning Boost: Intrinsic Functions | AI News | Intel Software

Leave a Reply

Your email address will not be published. Required fields are marked *