Skip to content
This repository has been archived by the owner on Apr 8, 2022. It is now read-only.

Commit

Permalink
Add 1.10.0 package (#4)
Browse files Browse the repository at this point in the history
* update readme

* Update readme

* add 1.10 android package

* minor update

* Add official release link
  • Loading branch information
guoyu-wang authored Dec 8, 2021
1 parent 4854a2c commit 292147b
Show file tree
Hide file tree
Showing 2 changed files with 13 additions and 3 deletions.
7 changes: 6 additions & 1 deletion CHANGELOG.md
Original file line number Diff line number Diff line change
@@ -1,5 +1,10 @@
# Changelog

[1.9.1]
# [1.10.0]
[Official ONNX Runtime 1.10.0 release](https://github.com/microsoft/onnxruntime/releases/tag/v1.10.0)
- Added Android AAR package - 12/07/2021

# [1.9.1]
[Official ONNX Runtime 1.9.1 release](https://github.com/microsoft/onnxruntime/releases/tag/v1.9.1)
- Added Android AAR packgage - 10/27/2021

9 changes: 7 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,15 +1,17 @@
# Unofficial ONNX Runtime Mobile Releases

The repository hosts the Unofficial [ONNX Runtime](https://github.com/microsoft/onnxruntime) Mobile Release for
The repository hosts the **Unofficial** [ONNX Runtime](https://github.com/microsoft/onnxruntime) Mobile Release for
- Android
- iOS (work in progress)

## Disclaimer

The mobile packages published here are built from the official [ONNX Runtime](https://github.com/microsoft/onnxruntime) project. They are unofficial ONNX Runtime packages for Mobile devices.
The mobile packages published here are built from the official [ONNX Runtime](https://github.com/microsoft/onnxruntime) project. They are **Unofficial** ONNX Runtime packages for Mobile devices.

For any issues or questions, please visit the official [ONNX Runtime](https://github.com/microsoft/onnxruntime) repository.

As for now, there is no plan to publish these unofficial packages to public package managers such as Maven or CocoaPods.

## Motivation

The official ONNX Runtime Mobile packages are published at
Expand All @@ -19,6 +21,7 @@ The official ONNX Runtime Mobile packages are published at
These official packages target reduced disk footprint for mobile platforms, however they have certain limitations, such as
- Limited Operators, Opsets and Types support, see [ONNX Runtime Mobile Pre-Built Package](https://onnxruntime.ai/docs/reference/mobile/prebuilt-package/), if your model has unsupported operators, you may see error message like, `"AssignNodesToEpsFromHashesImpl Failed to find kernel def hash (8090321298879394920) in kernel registries for MaxPool(8) node with name 'layer3/pool_3/MaxPool'."`
- Support only ORT file format, but not the ONNX file format. You will need to convert your ONNX model, see [ONNX Model Conversion](https://onnxruntime.ai/docs/tutorials/mobile/model-conversion.html)
- For more information, see [ONNX Runtime Mobile Limitations](https://onnxruntime.ai/docs/tutorials/mobile/limitations.html)

The ONNX Runtime Mobile packages published here are full ONNX Runtime packages which do not have the above limitations. The drawback is that the packages publish here has large disk footprint, for example,

Expand Down Expand Up @@ -50,3 +53,5 @@ dependencies {

### iOS
Work in progress

## [Changelog](./CHANGELOG.md)

0 comments on commit 292147b

Please sign in to comment.