The TensorFlow ecosystem can only grow through the contributions of this community. Thanks so much for your enthusiasm and your work—we appreciate everything you do!
In the interest of fostering an open and welcoming environment, contributors and maintainers pledge to make participation in our project and our community a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, gender identity and expression, level of experience, nationality, personal appearance, race, religion, or sexual identity and orientation.
Examples of behaviors that contribute to creating a positive environment include:
- Use welcome and inclusive language.
- Be respectful of differing viewpoints and experiences.
- Gracefully accept constructive criticism.
- Foster what's best for the community.
- Show empathy for other community members.
Decisions are made based on technical merit and consensus. The TensorFlow community aspire to treat everyone equally, and to value all contributions. For more information on best practices in the TensorFlow community, please review our Code of Conduct.
Make your first contribution
There are many ways to contribute to TensorFlow and all are welcome!
Whether you want to contribute code, make improvements to the TensorFlow API documentation, or add your Jupyter notebooks to the tensorflow/examples repo, this guide provides everything you need to get started. Our most common contributions include code, documentation, and community support.
- Write code.
- Improve documentation.
- Answer questions on StackOverflow.
- Participate in the discussion on the TensorFlow forums.
- Contribute example notebooks.
- Investigate bugs and issues on GitHub.
- Review and comment pull requests from other developers.
- Report issues.
- Give a “thumbs up” on issues that are relevant to you.
- Reference TensorFlow in your blogs, papers, and articles.
- Talk about TensorFlow on social media.
- ... even just starring the repos you like on GitHub!
We're excited to have you here, and thank you for your interest in contributing to the project! This is an exciting time to be working on scientific computation and machine learning, and it's only just beginning.
The TensorFlow ecosystem
TensorFlow is an open source software library for high performance numerical computation. Its flexible architecture allows for easy deployment of algorithms and code across a variety of platforms (CPUs, GPUs, TPUs)—from desktops, to clusters of servers, to mobile and edge devices.
TensorFlow was originally developed by researchers and engineers from the Google Brain team within Google's AI organization. Google open sourced TensorFlow in the hope of sharing technology with the external community and encouraging collaboration between researchers and industry. Since then, TensorFlow has grown into a thriving ecosystem of products, on a wide range of platforms. But our goal is still to make machine learning accessible to anyone, anywhere.
Below is a brief list of products that the TensorFlow team is working on right now:
- TensorFlow Core
- Swift for TensorFlow
- Model Garden
- TensorFlow Serving
- TensorFlow Documentation
To see what else we're working on at any time, see the TensorFlow organization on GitHub.
Related community projects
An important part of the TensorFlow ecoystem is the network of related (but unaffiliated) projects. Each of these projects has its own sponsors, contributors, and community.
- OpenSeq2Seq — distributed and mixed-precision training of sequence-to-sequence models, created by the folks at NVIDIA.
- Sonnet — DeepMind's library for building complex neural networks.
- Lattice — implementation of monotonic calibrated interpolated look-up tables.
- TensorFrames — TensorFlow binding for Apache Spark.
- TensorForce — library for applied reinforcement learning.
- TensorFlowOnSpark — distributed TensorFlow with Apache Spark.
- Horovod — distributed training framework for Keras, TensorFlow, and PyTorch.
- R Interface to TensorFlow — Keras and Estimator APIs in R
- TensorFlow-Slim — high-level library for defining models.
- tensorpack — toolbox focused on training speed for large datasets.
- tf-encrypted — machine learning on encrypted data.
- TensorLayer — deep learning and reinforcement learning library.
If your project is not listed below, please contact firstname.lastname@example.org to have it included.