Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/webmachinelearning/webnn-native
🧠⚙️ Standalone native implementation of the Web Neural Network API
https://github.com/webmachinelearning/webnn-native
Last synced: about 2 months ago
JSON representation
🧠⚙️ Standalone native implementation of the Web Neural Network API
- Host: GitHub
- URL: https://github.com/webmachinelearning/webnn-native
- Owner: webmachinelearning
- License: apache-2.0
- Created: 2021-01-21T15:59:03.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2023-04-07T02:14:10.000Z (over 1 year ago)
- Last Synced: 2024-06-09T03:34:20.176Z (3 months ago)
- Language: C++
- Homepage:
- Size: 6.64 MB
- Stars: 48
- Watchers: 10
- Forks: 21
- Open Issues: 38
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-oneapi - webnn-native - WebNN Native is an implementation of the Web Neural Network API, providing building blocks, headers, and backends for ML platforms including DirectML, OpenVINO, and XNNPACK. (Table of Contents / AI - Frameworks and Toolkits)
- awesome-openvino - webnn-native - WebNN Native is an implementation of the Web Neural Network API, providing building blocks, headers, and backends for ML platforms including DirectML, OpenVINO, and XNNPACK. (Table of content / Miscellaneous)
README
| Backend \ OS | Windows | Linux |
|---|---|---|
| null (for unit test) | [](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_null.yml)| |
| DirectMLX | [](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_dmlx.yml)
[](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_node_dmlx.yml)
[](https://github.com/webmachinelearning/webnn-native/actions/workflows/memory_leak_check_dmlx.yml)| |
| OpenVINO | [](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_openvino_windows.yml)
[](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_node_openvino_windows.yml)|[](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_openvino_linux.yml)
[](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_node_openvino_linux.yml)|
| XNNPACK | [](https://github.com/BruceDai/webnn-native/actions/workflows/build_test_xnnpack_windows.yml)|[](https://github.com/BruceDai/webnn-native/actions/workflows/build_test_xnnpack_linux.yml)|
| oneDNN |[](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_onednn_windows.yml)|[](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_test_onednn_linux.yml)|
| MLAS | [](https://github.com/webmachinelearning/webnn-native/actions/workflows/build_mlas_windows.yml)| |
[](https://github.com/webmachinelearning/webnn-native/actions/workflows/clang_format_check.yml)
# WebNN-native
WebNN-native is a native implementation of the [Web Neural Network API](https://webmachinelearning.github.io/webnn/).
It provides several building blocks:
- **WebNN C/C++ headers** that applications and other building blocks use.
- The `webnn.h` that is an one-to-one mapping with the WebNN IDL.
- A C++ wrapper for the `webnn.h`
- **Backend implementations** that use platforms' ML APIs:
- **DirectML** on Windows 10
- **DirectMLX** on Windows 10
- **OpenVINO** on Windows 10 and Linux
- **oneDNN** on Windows 10 and Linux
- **XNNPACK** on Windows 10 and Linux
- **MLAS** on Windows 10 and Linux
- _Other backends are to be added_
WebNN-native uses the code of other open source projects:
* The code generator and infrastructure code of [Dawn](https://dawn.googlesource.com/dawn/) project.
* The DirectMLX and device wrapper of [DirectML](https://github.com/microsoft/DirectML) project.
* The [XNNPACK](https://github.com/google/XNNPACK) project.
* The [oneDNN](https://github.com/oneapi-src/oneDNN) project.
* The [MLAS](https://github.com/microsoft/onnxruntime/tree/master/onnxruntime/core/mlas) project.
## Build and Run
### Install `depot_tools`
WebNN-native uses the Chromium build system and dependency management so you need to [install depot_tools] and add it to the PATH.
[install depot_tools]: http://commondatastorage.googleapis.com/chrome-infra-docs/flat/depot_tools/docs/html/depot_tools_tutorial.html#_setting_up
**Notes**:
* On Windows, you'll need to set the environment variable `DEPOT_TOOLS_WIN_TOOLCHAIN=0`. This tells depot_tools to use your locally installed version of Visual Studio (by default, depot_tools will try to download a Google-internal version).
### Get the code
Get the source code as follows:
```sh
# Clone the repo as "webnn-native"
> git clone https://github.com/webmachinelearning/webnn-native.git webnn-native && cd webnn-native
# Bootstrap the gclient configuration
> cp scripts/standalone.gclient .gclient
# Fetch external dependencies and toolchains with gclient
> gclient sync
```
### Setting up the build
Generate build files using `gn args out/Debug` or `gn args out/Release`.
A text editor will appear asking build options, the most common option is `is_debug=true/false`; otherwise `gn args out/Release --list` shows all the possible options.
To build with a backend, please set the corresponding option from following table.
| Backend | Option |
|---------|--------------|
| DirectML | `webnn_enable_dml=true` |
| DirectMLX | `webnn_enable_dmlx=true` |
| OpenVINO | `webnn_enable_openvino=true` |
| XNNPACK | `webnn_enable_xnnpack=true` |
| oneDNN | `webnn_enable_onednn=true` |
| MLAS | `webnn_enable_mlas=true` |
### Build
Then use `ninja -C out/Release` or `ninja -C out/Debug` to build WebNN-native.
**Notes**
* To build with XNNPACK backend, please build XNNPACK first, e.g. by [`./scripts/build-local.sh`](https://github.com/google/XNNPACK/blob/master/scripts/build-local.sh). For Windows build, it requires supplying -DCMAKE_MSVC_RUNTIME_LIBRARY="MultiThreaded$<$:Debug>" to set MSVC static runtime library.
* To build with oneDNN backend, please build oneDNN first by following the [build from source instructions](https://oneapi-src.github.io/oneDNN/dev_guide_build.html).
* To build with MLAS backend, please build MLAS (part of ONNX Runtime) first by following the [Build ONNX Runtime for inferencing](https://onnxruntime.ai/docs/build/inferencing.html#build-onnx-runtime-for-inferencing), e.g., by `.\build.bat --config Release --parallel --enable_msvc_static_runtime` for Windows build.
### Run tests
Run unit tests:
```sh
> ./out/Release/webnn_unittests
```
Run end2end tests on a default device:
```sh
> ./out/Release/webnn_end2end_tests
```
You can also specify a device to run end2end tests using "-d" option, for example:
```sh
> ./out/Release/webnn_end2end_tests -d gpu
```
Currently "cpu", "gpu" and "default" are supported, more devices are to be supported in the future.
**Notes**:
* For OpenVINO backend, please [install 2021.4 version](https://docs.openvinotoolkit.org/2021.4/openvino_docs_install_guides_installing_openvino_linux.html#install-openvino) and [set the environment variables](https://docs.openvinotoolkit.org/2021.4/openvino_docs_install_guides_installing_openvino_linux.html#set-the-environment-variables) before running the end2end tests.
* The current implementation of oneDNN and MLAS backends is mainly for the investigation of WebNN [Operation Level Execution
](https://webmachinelearning.github.io/webnn/#usecase-op-level-exec) use case. So only a limited set of tests (such as of conv2d) is expected to pass.
### Run examples
* [LeNet](/examples/LeNet/README.md)
## License
Apache 2.0 Public License, please see [LICENSE](/LICENSE).