The goal of Janus is to be a launcher that is:
- Minimal
- Made with Kotlin
- Gesture based
- Open source
Building Janus requires:
- git
- Android Studio Canary ("Android 3" as of 2017-08-05T06:25:29.933Z)
- python3, pip3, virtualenv, virtualenvwrapper (Check NOTE)
-
The launcher module
- Download and setup Android Studio Canary: link
- Open the android studio project in
/launcher
-
The model module
- Setup
tensorhack
as described inNOTE
- Play with stuffs in
model
- Setup
Follow this guide: link
- Fork the repo into your remote
- Hack
- Make a pull request here or the community driven organization repo.
- 2017-08-15T05:41:05.891Z: App filtering added
- 2017-08-13T02:36:29.212Z: Drawing pad added
- 2017-08-07T05:16:44.862Z: Initialize tensorflow project with readme and notes
- 2017-08-06T23:37:54.234Z: Refactor into a monorepo project with two module, one for launcher and one for the model.
- 2017-08-05T05:41:49.454Z: Launcher with apps initialized
- 2017-08-03T06:48:30.239Z: Initialize the project
- @mari-linhares's mnist-android-tensorflow
- @miyosuda's TensorFlowAndroidMNIST
- Keras Pull request on exporting a model
- Keras API guide
- Generate alphabet codegolf
- Deploying a tensorflow model to android
- Basic Painting with Views
- Creating a- deep learning ios-app with keras and tensorflow by @stratospark
- EMNIST github repository
- EMNIST paper
- Build a custom launcher on Android by Ashraff Hathibelagal
- How to Write Custom Launcher App in Android by Arnab Chakraborty
- Cohen, G., Afshar, S., Tapson, J., & van Schaik, A. (2017). EMNIST: an extension of MNIST to handwritten letters. Retrieved from http://arxiv.org/abs/1702.05373 (link is external)
MIT