https://github.com/mgechev/mk-tfjs
Play MK.js with TensorFlow.js
https://github.com/mgechev/mk-tfjs
cnn ml mobilenet tensorflow-js transfer-learning
Last synced: 6 months ago
JSON representation
Play MK.js with TensorFlow.js
- Host: GitHub
- URL: https://github.com/mgechev/mk-tfjs
- Owner: mgechev
- Created: 2018-10-23T19:50:54.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2023-01-07T02:33:26.000Z (over 2 years ago)
- Last Synced: 2024-12-15T21:34:58.133Z (6 months ago)
- Topics: cnn, ml, mobilenet, tensorflow-js, transfer-learning
- Language: JavaScript
- Homepage: https://blog.mgechev.com/2018/10/20/transfer-learning-tensorflow-js-data-augmentation-mobile-net/
- Size: 20 MB
- Stars: 157
- Watchers: 10
- Forks: 38
- Open Issues: 10
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Play MK.js with TensorFlow.js
Source code for my article ["Playing Mortal Kombat with TensorFlow.js. Transfer learning and data augmentation"](https://blog.mgechev.com/2018/10/20/transfer-learning-tensorflow-js-data-augmentation-mobile-net/).
You can find the post [here](https://blog.mgechev.com/2018/10/20/transfer-learning-tensorflow-js-data-augmentation/) and MK.js [here](https://github.com/mgechev/mk.js).
## Usage
To try the demo run the following commands:
```bash
npm i && npm i -g serve
npm start
cd model && serve -s .
```
Keep in mind that the model is trained with a small dataset. If it doesn't perform well for you, feel free to follow the instructions from the [blog post](https://blog.mgechev.com/2018/10/20/transfer-learning-tensorflow-js-data-augmentation-mobile-net/) and improve its accuracy.
## License
MIT