| commit | b37566985c941d77a0df596fdd1234fbc339f7cc | [log] [tgz] |
|---|---|---|
| author | MaheshRavishankar <1663364+MaheshRavishankar@users.noreply.github.com> | Mon Jul 24 14:38:45 2023 -0700 |
| committer | GitHub <noreply@github.com> | Mon Jul 24 14:38:45 2023 -0700 |
| tree | 7a0d41f918477050278152a678be7140e1282060 | |
| parent | 44502480fe8e2376e8a1d36835fc65825f088fcb [diff] |
Enable multi-use fusion for cases where the shapes of the two ops are the same. (#13747) Previous implementation of multi-use fusion was too aggressive making it hard to enable by default. For example it would fuse a reduction producer with its consumer which generates outputs of different shapes making it hard for backends to handle this. Making the conditions for fusion less aggressive to restrict this to all parallel producers/consumers of the same shape, should make this handalable on all backends. Issue #13545
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.