4dbaa73738
## Summary
* following inference's [set-up for end-to-end
tests](https://github.com/microsoft/onnxruntime/tree/main/js/web/test/e2e),
created an end-to-end test runner for training
* this test runner copies testdata from the [trainingapi
folder](https://github.com/microsoft/onnxruntime/tree/main/onnxruntime/test/testdata/training_api)
* then runs two tests (training session with evalModel & optimizer
model, and training session with the minimum options), and tests if the
ORT-web training package encompasses inference
* these tests check
* createTrainingSession
* runTrainStep
* runOptimizerStep if applicable
* the parameters methods (getParametersSize, loadParametersBuffer, and
getContiguousParameters)
## TL;DR
*
[`js/web/test/training/e2e/run.js`](https://github.com/microsoft/onnxruntime/compare/main...carzh:onnxruntime:carzh/training-e2e-runner?expand=1#diff-c1359c4d401f9ba69e937814219cefe5fd11b151a6ffd084c641af3c82e8216c)
is responsible for setting up and running the end to end tests
*
[`js/web/test/training/e2e/common.js`](https://github.com/microsoft/onnxruntime/compare/main...carzh:onnxruntime:carzh/training-e2e-runner?expand=1#diff-ee5452491b7b2563d175d13d81d10f2323b12b18589aa4c5798962a8b904a4a8)
contains the test function definitions (`testInferenceFunction`,
`testTrainingFunctionMin`, `testTrainingFunctionAll`)
## Flow
* entrypoint: user runs the following command in the terminal: `npm run
test:training:e2e`
*
[`js/web/package.json`](https://github.com/microsoft/onnxruntime/compare/main...carzh:onnxruntime:carzh/training-e2e-runner?expand=1#diff-79275844e75c3c410bb3a71c7f59b2b633e5a3e975c804ffc47220025084da28)
was modified to include an npm script that will run `run.js` which will
run the end to end tests
*
[`js/web/test/training/e2e/run.js`](https://github.com/microsoft/onnxruntime/compare/main...carzh:onnxruntime:carzh/training-e2e-runner?expand=1#diff-c1359c4d401f9ba69e937814219cefe5fd11b151a6ffd084c641af3c82e8216c)
is responsible for
* detecting and installing local tarball packages of ORT-web
* copying training data to the `js/web/training/e2e/data` folder
* starting two Karma processes. Karma is a test runner framework that
simulates testing in the browser.
* In this case, the tests happen in Chrome. We can configure the tests
to run in Edge and other browsers in the future.
* one of these karma processes is self-hosted, meaning it pulls the
ORT-web package from local
* the other karma process is not self-hosted, meaning it pulls the
ORT-web package from another source. In this case, we start an http
server that serves the ORT-web binaries.
*
[`js/web/test/training/e2e/simple-http-server.js`](https://github.com/microsoft/onnxruntime/compare/main...carzh:onnxruntime:carzh/training-e2e-runner?expand=1#diff-f798ab485f3ec26c299fe5b2923574c9e4b090200ba20d490bbf6c183286993c)
is responsible for starting the HTTP server and serving the ORT binary
files. This code almost identical to the same code in the inference E2E
tests.
*
[`js/web/test/training/e2e/karma.conf.js`](https://github.com/microsoft/onnxruntime/compare/main...carzh:onnxruntime:carzh/training-e2e-runner?expand=1#diff-436cfe8f670c768a04895bd4a1874a5e033f85e0e2d84941c62ff1f7c30a9f28)
Karma configuration file that specifies what happens when a karma
process is started. The config specifies Mocha as the testing framework,
which will go through all the loaded files and run any tests that exist
*
[`js/web/test/training/e2e/browser-test-wasm.js`](https://github.com/microsoft/onnxruntime/compare/main...carzh:onnxruntime:carzh/training-e2e-runner?expand=1#diff-13b6155e106dddc7b531ef671186e69b2aadb8a0f4b2f3001db0991567d78221)
File that contains the tests that Mocha will pick up on and run.
* The test functions (such as testInference and testTrainingFunctionAll)
are defined in
[`js/web/test/training/e2e/common.js`](https://github.com/microsoft/onnxruntime/compare/main...carzh:onnxruntime:carzh/training-e2e-runner?expand=1#diff-ee5452491b7b2563d175d13d81d10f2323b12b18589aa4c5798962a8b904a4a8).
## Notes
* I followed the [tests for training
core](
|
||
|---|---|---|
| .. | ||
| docs | ||
| lib | ||
| script | ||
| test | ||
| .gitignore | ||
| .npmignore | ||
| karma.conf.js | ||
| package-lock.json | ||
| package.json | ||
| README.md | ||
| tsconfig.json | ||
| types.d.ts | ||
ONNX Runtime Web
ONNX Runtime Web is a Javascript library for running ONNX models on browsers and on Node.js.
ONNX Runtime Web has adopted WebAssembly and WebGL technologies for providing an optimized ONNX model inference runtime for both CPUs and GPUs.
Why ONNX models
The Open Neural Network Exchange (ONNX) is an open standard for representing machine learning models. The biggest advantage of ONNX is that it allows interoperability across different open source AI frameworks, which itself offers more flexibility for AI frameworks adoption.
Why ONNX Runtime Web
With ONNX Runtime Web, web developers can score models directly on browsers with various benefits including reducing server-client communication and protecting user privacy, as well as offering install-free and cross-platform in-browser ML experience.
ONNX Runtime Web can run on both CPU and GPU. On CPU side, WebAssembly is adopted to execute the model at near-native speed. ONNX Runtime Web complies the native ONNX Runtime CPU engine into WebAssembly backend by using Emscripten, so it supports most functionalities native ONNX Runtime offers, including full ONNX operator coverage, multi-threading, ONNX Runtime Quantization as well as ONNX Runtime Mobile. For performance acceleration with GPUs, ONNX Runtime Web leverages WebGL, a popular standard for accessing GPU capabilities. We are keeping improving op coverage and optimizing performance in WebGL backend.
See Compatibility and Operators Supported for a list of platforms and operators ONNX Runtime Web currently supports.
Usage
Refer to ONNX Runtime JavaScript examples for samples and tutorials.
Documents
Developement
Refer to the following links for development information:
Compatibility
| OS/Browser | Chrome | Edge | Safari | Electron | Node.js |
|---|---|---|---|---|---|
| Windows 10 | wasm, webgl | wasm, webgl | - | wasm, webgl | wasm |
| macOS | wasm, webgl | wasm, webgl | wasm, webgl | wasm, webgl | wasm |
| Ubuntu LTS 18.04 | wasm, webgl | wasm, webgl | - | wasm, webgl | wasm |
| iOS | wasm, webgl | wasm, webgl | wasm, webgl | - | - |
| Android | wasm, webgl | wasm, webgl | - | - | - |
Operators
WebAssembly backend
ONNX Runtime Web currently support all operators in ai.onnx and ai.onnx.ml.
WebGL backend
ONNX Runtime Web currently supports a subset of operators in ai.onnx operator set. See webgl-operators.md for a complete, detailed list of which ONNX operators are supported by WebGL backend.
WebGPU backend
WebGPU backend is still an experimental feature. See webgpu-operators.md for a detailed list of which ONNX operators are supported by WebGPU backend.
License
License information can be found here.