Microsoft and Amazon today announced a new deep learning interface called Gluon. The Seattle-based companies are join forces to make deep learning accessible to all developers with the new deep library. The library provides a “concise and easy-to-understand” programming interface that will allow developers to build machine learning models for use on the cloud, smartphones, and more.
Gluon will provide developers with a Python API which they can use to prototype, build, and train complex machine learning models. Microsoft and Amazon said in a joint press release today that building machine learning models can be very difficult, requires a lot of time and processing power. Gluon hopes to tackle that solution with its Python API And pre-build neural network components. Gluon currently works with Apache MXNet, but Microsoft and Amazon will be adding support for Microsoft’s Cognitive Toolkit later.
“We created the Gluon interface so building neural networks and training models can be as easy as building an app. We look forward to our collaboration with Microsoft on continuing to evolve the Gluon interface for developers interested in making machine learning easier to use,” said Swami Sivasubramanian, the VP of Amazon AI.
Using the Pyhton API provide by Gluon, you will be able to build machine learning models and train them without needing to write thousands of lines of code that you would normally need to build and train these models. Gluon also won’t affect the training performance at all, as you’ll be able to train your models at the same speed you’d get when using engines like Apache MXNet directly.
Gluon is available today, and you can check it out on GitHub here.