Accelerating TensorFlow with oneDNN

TensorFlow is an end-to-end open-source machine learning framework by Google. This talk will highlight Google and Intel’s collaboration on accelerating TensorFlow on x86 CPUs with oneDNN framework. We will cover support for new neural-network-focused hardware features such as int8 and bfloat16 vector and matrix extensions, including demonstrating select workload speedups.


Learn about joining the UXL Foundation:

Join now