📄 Tutorial - https://rekoil.io/blog/onnxruntime-web-tutorial/
This repository contains the code for the tutorial on building a browser-native machine learning app using ONNX Runtime Web. In the tutorial, we dive into onnxruntime-web
by deploying a pre-trained PyTorch model. The model is run natively in the browser using WebAssembly via onnxruntime-web
. The model that in use is AlexNet which, has been trained as an image classifier on the ImageNet dataset.
To launch a live server locally, simply use npm run serve
. Note that npm run serve
also called npm run build
meaning that any changes in main.js
or other package dependencies will be included in the deployed app bundle.
The app uses webpack
to bundle the package dependencies into dist/bundle.min.js
, to re-build the bundle: npm run build
.