dequantize softmax (#9337) See #8974. This is still a 20% end-to-end latency improvement on MobileBert-int8 on configs where matmuls are already reasonably fast, making other things like Softmax more important relatively. That is even after Softmax slowness was much improved recently as observed in #9170. Moreover, discussion around #8974 suggests that the path forward for non-dequantized Softmax is nontrivial, so putting our benchmarks on the dequantized path for now will help insulate them a bit from what we expect to be in-flux for the foreseeable future.
IREE (Intermediate Representation Execution Environment, pronounced as “eerie”) is an MLIR-based end-to-end compiler and runtime that lowers Machine Learning (ML) models to a unified IR that scales up to meet the needs of the datacenter and down to satisfy the constraints and special considerations of mobile and edge deployments.
See our website for project details, user guides, and instructions on building from source.
IREE is still in its early phase. We have settled down on the overarching infrastructure and are actively improving various software components as well as project logistics. It is still quite far from ready for everyday use and is made available without any support at the moment. With that said, we welcome any kind of feedback on any communication channels!
See our website for more information.
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.