Research paper code contribution
We encourage researchers to publish new state-of-the-art machine learning models to the TensorFlow Model Garden.
To contribute a new research paper code, please provide your plans using GitHub issues in this repository before making any pull requests.
We want to ensure research code implementations from contributors are high-quality and well-documented.
Your contributions must meet the following requirements to be accepted to the TensorFlow Model Garden repository.
| Directory | Requirements |
|---|---|
| official | • Provide a model implemented in TensorFlow 2 • Use the modelling libraries provided by the Model Garden • Provide baseline results • Support distributed training on GPUs and TPUs • Reasonable performance on GPUs and TPUs • Need a SLA (Service Level Agreement) for community support • Pass the TensorFlow code usability review process |
| research | • Provide a model implemented in TensorFlow 2 • Use the modelling libraries provided by the Model Garden for supported ML tasks • Provide baseline results • Reasonable performance on GPUs or TPUs • Need a SLA (Service Level Agreement) for community support |
| community | • Models implemented in TensorFlow 2 by external contributors • Reproduce the paper results |
- A model from the paper accepted at top machine learning venues or
- A state-of-the-art model from a pre-publication available at arXiv
- Should be able to reproduce the same results in a published paper
- Should provide reasonable out-of-box performance
- Should have accuracy and performance test results on GPUs or TPUs
- Pre-trained models in TensorFlow SavedModel format should be published to TensorFlow Hub.
- Use the README template that describes the information required for publishing a new code implementation.
- We also recommend to use Read the Docs for hosting documentation.
- Documentation can be automatically generated from your repository.
- Please see Read the Docs Template.
Note: Exceptions can be made case by case basis.