commit | 4bcfab3c4e1ac9ca689de98a2e94badbf27080eb | [log] [tgz] |
---|---|---|
author | Stella Laurenzo <stellaraccident@gmail.com> | Wed Aug 04 20:28:24 2021 -0700 |
committer | GitHub <noreply@github.com> | Wed Aug 04 20:28:24 2021 -0700 |
tree | 215123f39bb12a0169b1f0bb0d6de6df120b3129 | |
parent | 27c6082ba584ea4e79c259b7e96d2ec82aad993a [diff] |
Flatten first structure level of tensorflow args and results. (#6636) * Flatten first structure level of tensorflow args and results. * Has the effect that V1 saved models just use regular args (vs vm.list) for their implicit keyword arguments. * Also "inlines" the first level result (tuple/list/dict) so that simple, one level nests (i.e. dict of keyword results) is compiled as just a normal multi-result function. * Breaking change: The python API will not be able to process structured invocations across this commit. * Add function unit test with mocks (needs more test cases in anger but tests new behavior).
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!
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.