Structured Operations: Modular Design of Code Generators for Tensor Compilers
Abstract
The performance of machine learning systems heavily relies on code generators tailored to tensor computations.
We propose an approach to the design and implementation of such code generators leveraging the natural structure of tensor algebra and illustrating the progressive lowering of domain-specific abstractions in the MLIR infrastructure.
We propose an approach to the design and implementation of such code generators leveraging the natural structure of tensor algebra and illustrating the progressive lowering of domain-specific abstractions in the MLIR infrastructure.