Microsoft Open Sources AI Toolkit

After waiting for a big open source toolkit, two come along at the same time. Microsoft follows Google in providing the world with a way to do advanced distributed AI.  Earlier this month Google open sourced TensorFlow, the tool that it uses internally for a wide range of parallel computations, including implementing neural networks and other AI learning methods. Shortly afterwards researchers at the Microsoft Asia research lab made the Microsoft Distributed Machine Learning Toolkit available on GitHub under an MIT licence. This toolkit includes three key components: • DMTK Framework: a flexible framework that supports unified interface for data parallelization, hybrid data structure for big model storage, model scheduling for big model training, and automatic pipelining for high training efficiency. • LightLDA, an extremely fast and scalable topic model algorithm,…


Link to Full Article: Microsoft Open Sources AI Toolkit