Following Google and Intel’s partnership, the open-source TensorFlow has announced that their machine learning library is being updated.
Intel’s open-source oneAPI Deep Neural Network Library (oneDNN) is now enabled by default in Tensor Flow, a Google-led effort. OneDNN is an open-source cross-platform performance collection of deep learning building blocks aimed at deep learning application and framework developers, such as TensorFlow.
According to Intel, oneDNN will provide the organizations with a considerable speed boost of up to 3 times for AI operations using TensorFlow.
Read More: https://venturebeat.com/2022/05/31/tensorflow-now-defaults-to-intel-onednn-ai-optimizations/
