Contributing
Each of us can make a difference ❤️
There are many ways you can contribute to the AutoML benchmark - and many don’t even require any coding! We value contributions equally. No matter which way you contribute, you make using AMLB a better experience for everyone. Here are some ways you can help out:
- Use it, and share it with others! We created this benchmark to make it simple for people to provide a high-quality evaluation of AutoML frameworks, and raise the standards for reproducibility in our field. Using the AutoML benchmark signals that you support these goals. By setting a higher standard, even evaluations that do not use AMLB will be required to justify their experimental setup and evaluation to a higher standard. When you use AMLB in a publication, please cite us (see below).
- Documentation! Write about your experience using the AutoML benchmark independently, or contribute to our documentation pages. Creating additional resources and improving existing ones makes using AMLB more accessible to everyone. See "contributing changes" for information on how to set up your development environment if you want to contribute to the documentation pages.
- Help others use the AutoML benchmark! We regularly get people asking questions about how to use the benchmark on our issue tracker and discussion board. They often only need a little bit of help troubleshooting their local setup, and in some cases with filing a bug report. It often does not take much more than some familiarity with the benchmark to help people out. So don't be hesitant to help people.
- Report bugs. See "reporting a bug" in the code contributors guide.
- Contributing code, see "contributing changes" in the code contributors guide. Small incremental changes are preferred, and there are a lot of tasks we could use your help with!
- And more... organize an event, suggest new datasets or benchmarks, take part in the design discussions, and many more. If you have an idea and want to discuss it with us, please let us know! The discussion board is the preferred place.
Citing AMLB
When referring to the benchmark please cite
@article{JMLR:v25:22-0493,
author = {Pieter Gijsbers and Marcos L. P. Bueno and Stefan Coors and Erin LeDell and S{{\'e}}bastien Poirier and Janek Thomas and Bernd Bischl and Joaquin Vanschoren},
title = {AMLB: an AutoML Benchmark},
journal = {Journal of Machine Learning Research},
year = {2024},
volume = {25},
number = {101},
pages = {1--65},
url = {http://jmlr.org/papers/v25/22-0493.html}
}
author = {Pieter Gijsbers and Marcos L. P. Bueno and Stefan Coors and Erin LeDell and S{{\'e}}bastien Poirier and Janek Thomas and Bernd Bischl and Joaquin Vanschoren},
title = {AMLB: an AutoML Benchmark},
journal = {Journal of Machine Learning Research},
year = {2024},
volume = {25},
number = {101},
pages = {1--65},
url = {http://jmlr.org/papers/v25/22-0493.html}
}