fortytwoapps / tfkt

Class based ML code in the Browser - A KotlinJS wrapper for the wonderful TensorFlow.js library.

TensorFlow.kt

Class based ML code in the Browser - A KotlinJS wrapper for the wonderful TensorFlow.js library.

Base library: https://www.tensorflow.org/js

Getting Started

Development

You can explore the TensorFlow-kt code either by browsing this repo or cloning from source:

git clone https://github.com/fortytwoapps/tfkt.git

Installing

To setup TensorFlow-kt and start using it in your project, add these lines to build.gradle:

repositories {
    maven { url = "https://dl.bintray.com/robert-cronin/fortytwoapps" }
}

dependencies {
    implementation "fortytwoapps:tfkt:1.5.1"
}

Usage

To start using TensorFlow-kt in your project, simply import like this:

import com.fortytwoapps.tfkt.tensors.*

Note: only tensors are available so far, methods for the rest of the library will be added soon.

Built With

• TensorFlow.js - The underlying JavaScript library used
• Travis CI - Continuous Integration
• Gradle - Dependency Management

Contributing

Feel free to contribute if you feel this toolkit is lacking for your application. Contributions are open.

Authors

• Robert Cronin - Initial work - robert-cronin

License

This project is licensed under the Apache 2.0 License - see the LICENSE.md file for details

Acknowledgments

• Hat tip goes to developers at Google for developing the JavaScript TensorFlow Library
• Template for this readme came from PurpleBooth