TensorFlow unveils MLIR for faster machine learning

Engineers working on Google's TensorFlow machine learning framework have revealed a subproject, MLIR, that is intended to be a common intermediate language for machine learning frameworks. MLIR, short for Multi-Level Intermediate Representation, will allow projects using TensorFlow and other machine learning libraries to be compiled to more efficient code that takes maximum advantage of underlying hardware. What's more, MLIR could in time be used by compilers generally, extending its optimization benefits beyond machine learning projects. It represents an intermediate compilation step between those higher-level languages and machine code. The compiler framework LLVM uses an intermediate representation, or IR, of its own.