Intel Opens Up Compute Library for Deep Neural Networks

Intel's 01.org has open-sourced a new Compute Library for Deep Neural Networks (clDNN).

This clDNN library is accelerated on Intel HD/Iris Graphics hardware and is a big step for Intel graphics in the area of deep learning. clDNN so far supports AlexNet, VGG, GoogleNet, and ResNet. The code is licensed under the Apache 2.0.

This compute library is designed for Intel Skylake GPUs and newer and targets OpenCL. Both Windows and Linux are currently supported.

Those wishing to learn more about Intel's Compute Library for Deep Neural Networks can find the details via GitHub.