What can be improved with ngraph funtion instead of .xml in openvino inference engine
Hi Maheshwari,
nGraph is a ‘framework-neutral’ Deep Neural Network (DNN) model compiler.
An nGraph function, once created, can wrap into a CNNNetwork, thus creating utility for data scientists or app developers to define a Deep Learning (DL) model in a neutral way that does not depend on existing DL frameworks.
In other words, nGraph compiler offers framework abstraction, which allows the creation of DL model without having to think about how that model must be adjusted across different frameworks.
Regards,
Munesh
Hi Maheshwari,
nGraph is a ‘framework-neutral’ Deep Neural Network (DNN) model compiler.
An nGraph function, once created, can wrap into a CNNNetwork, thus creating utility for data scientists or app developers to define a Deep Learning (DL) model in a neutral way that does not depend on existing DL frameworks.
In other words, nGraph compiler offers framework abstraction, which allows the creation of DL model without having to think about how that model must be adjusted across different frameworks.
Regards,
Munesh
For more complete information about compiler optimizations, see our Optimization Notice.