Math Kernel Library for Deep Neural Networks

Intel® Math Kernel Library for Deep Neural Networks (Intel® MKL-DNN) is an open source performance library for Deep Learning (DL) applications intended for acceleration of DL frameworks on Intel® architecture. Intel® MKL-DNN includes highly vectorized and threaded building blocks to implement convolutional neural networks (CNN) with C and C++ interfaces.

For more information and documentation on Intel® Math Kernel Library see The sources can be downloaded from


Ref-OS-IoT has support for Intel® MKL-DNN pre-integrated, including pre-compiled versions of some of the examples.(

These examples can easily be compiled and run directly in Ref-OS-IoT using the following commands (names between <> should be replaced by actual name):

Copy the example file from

For C examples

gcc -Wall <example>.c -lmkldnn -o <cexample>

For C++ examples

g++ -std=c++11 <example>.cpp -lmkldnn -o <cppexample>


Para obter informações mais completas sobre otimizações do compilador, consulte nosso aviso de otimização.