commit | 037497fdf79e96e344a63f5b6ab462820cecd318 | [log] [tgz] |
---|---|---|
author | Stella Laurenzo <stellaraccident@gmail.com> | Wed Jun 16 11:32:05 2021 -0700 |
committer | GitHub <noreply@github.com> | Wed Jun 16 11:32:05 2021 -0700 |
tree | 55bb07fd47f2f99dad128af6d3a1d06375bef9f7 | |
parent | 5d5bce260b32867978ee6f25b1b98591dc49fbb6 [diff] |
Add a facility so that the Python API will save off all temporaries. (#6229) * Add a facility so that the Python API will save off all temporaries. * This is done at a lower level than the TF test framework and happens automatically if env variables are set (can be customized). * Intended for debugging and getting dumps of foreign systems where the high level should not be altered to manually pass in artifact retention. * I'm sure folks will write scripts around this, but the intent is for humans to have a less miserable time when dealing with an arbitrary system. * To shortcut everything, just set the env var IREE_SAVE_TEMPS to something like "/tmp/ireedumps/{pid}/{id}". * Updates README (including fixing the package setup which was excluding it).
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.