commit | 6bad5662fbf4a017fa2a508775acabf55a4c5f5d | [log] [tgz] |
---|---|---|
author | mariecwhite <mariewhite@google.com> | Tue Jan 31 19:26:50 2023 -0800 |
committer | GitHub <noreply@github.com> | Wed Feb 01 03:26:50 2023 +0000 |
tree | a07399b1f97173d1111e4a770670dbcea2f6158d | |
parent | 8bedd4b249caa06ed346557214c016fe9b685dbf [diff] |
Add ability to run SHARK at presubmit (#11946) This change adds the ability to run SHARK benchmarks at presubmit. IREE is compiled to incorporate the changes in the presubmit and generates a summary: https://storage.googleapis.com/shark-benchmark-artifacts/2023-01-31.sha_8111.timestamp_1675131232.presubmit/summary.html. Note version info for IREE is now a git commit SHA instead of a nightly release. This is not intended to run frequently but gives developers the option to double-check whether their changes will impact SHARK. The presubmit workflow trigger will be removed before submission .
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.