nbc-lightweight-models
v1.0.2
Published
Pre-trained models for Browser by Oleksandr Lebedyev
Downloads
3
Readme
Pre-trained Models
This repository hosts a collection of pre-trained models that have been adapted for Browsers.
These models are available on NPM and unpkg, making them easily usable in any project. They can be utilized directly or integrated into transfer learning scenarios with TF.js
For comprehensive information about the APIs for each model, please refer to the README files in their respective directories. I strive to simplify the usage so that the API can be accessible to non-machine learning experts.
If you're interested in contributing a model, please open a GitHub issue to discuss and gauge interest. I aim to include models that complement the existing set and can serve as foundational blocks in various applications.
Models
Here's an overview of the models currently provided, along with links to live demos and sources for further exploration:
Development
You can run unit tests for any of the models by executing the following command within a directory:
yarn test
New models should have a test NPM script (see the package.json and run_tests.ts helper script for reference).
To run all tests, you can use the following command from the root of this repository:
yarn presubmit