commit | 5ba8a7a1465ef61ffc3bb08232659322162e5a8b | [log] [tgz] |
---|---|---|
author | Stella Laurenzo <stellaraccident@gmail.com> | Tue May 11 21:29:23 2021 -0700 |
committer | GitHub <noreply@github.com> | Tue May 11 21:29:23 2021 -0700 |
tree | 83c09ab3c101d0825fc8d4ffe05b0f40a54772ff | |
parent | 4d1a317b1cd5a13c780bde9d224822f4d3769954 [diff] |
Finish implementing native ABI support. (#5844) * Does not yet flip it on. * Adds a pass and a commented out usage to generate ABI metadata for MHLO, allowing JAX tests to pass. * I was just going to let this go down the untyped path, but it turns out people have been relying on implicit casting between scalar -> 0d tensor, so I just did it right. * Switches TF returns to emit tuples instead of lists. * Passes pytree_test.py, mobile bert (uses old kwarg style TF arg passing), and JAX tests. * Once landed, I will start a branch to flip everything for real, and there will likely be additional triage.
IREE (Intermediate Representation Execution Environment, pronounced as “eerie”) is an MLIR-based end-to-end compiler that lowers Machine Learning (ML) models to a unified IR optimized for real-time inference on mobile/edge devices against heterogeneous hardware accelerators. IREE also provides flexible deployment solutions for its compiled ML models.
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!
IREE is licensed under the terms of the Apache license. See LICENSE for more information.