commit | dacf2b19798d2251409d78b0faf88d07756faa26 | [log] [tgz] |
---|---|---|
author | Ben Vanik <benvanik@google.com> | Mon Nov 15 11:55:40 2021 -0800 |
committer | GitHub <noreply@github.com> | Mon Nov 15 11:55:40 2021 -0800 |
tree | 43022c2b6b695482b5dc82bc43183c37b485bc5f | |
parent | acb25c46cdb0ffacd20c4392511a6e22bf15ca04 [diff] |
Changing tflite binding generation to use hal.tensor.cast. (#7661) This allows for the buffers to be directly specified (just like we specify buffer_views in the normal IREE bindings) and lets us drop all of the shapex dialect ops from this layer. Previously this was two passes but it relied on some shady assumptions about when and how shapes were propagated. Now as a single pass all global expanded shape dimensions, the ops tying them to values, and the logic for query/update are explicitly specified and robust to further transformation.
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.