DLVM: A modern compiler infrastructure for deep learning systems. Deep learning software demands reliability and performance. However, many of the existing deep learning frameworks are software libraries that act as an unsafe DSL in Python and a computation graph interpreter. We present DLVM, a design and implementation of a compiler infrastructure with a linear algebra intermediate representation, algorithmic differentiation by adjoint code generation, domain-specific optimizations and a code generator targeting GPU via LLVM. Designed as a modern compiler IR inspired by LLVM and the Swift Intermediate Language, DLVM IR is more modular and more generic than existing deep learning compiler IRs, and supports tensor DSLs with high expressivity. With our prototypical staged DSL embedded in Swift, we argue that the DLVM system enables a form of modular, safe and performant frameworks for deep learning.
Keywords for this software
References in zbMATH (referenced in 1 article )
Showing result 1 of 1.
- Richard Wei, Vikram Adve, Lane Schwartz: DLVM: A modern compiler infrastructure for deep learning systems (2017) arXiv