How to contribute to Vecto

Vecto is an open-source project which you are very welcome to contribute to.

If there's an issue you'd like us to fix, or question not covered in the tutorial, please raise an issue on Github. If there's an issue you'd like to fix, the most recent contribution guide can be found here. In short, we like code that is pep8-compliant and comes with unit tests.

If you would like to collaborate and/or contribute some new functionality, get in touch! The Vecto team currently includes 7 researchers from Japan, USA, Russia, and Canada. The ongoing work is discussed on Gitter.

Publishing your models and embeddings in Vecto will make it easier for other people to reproduce your work. All they will have to do is pip install, rather than spend weeks trying to decipher half-working code from random repositories. If you make it easy for other researchers to build on and compare with your work, they are more likely to use it and give you credit! The best part is that the bibtex for your published papers on all models and/or data will also be included in the Vecto metadata.

If you have a corpus or a task dataset that you can share and would like to spread the word about, that is also very welcome - please get in touch. By making your benchmark/corpus easily available with all the rest of our library you will increase the chances that more people will use it and cite you. Once again, all the datasets and corpora we share come with Vecto metadata, which will include the citation for your paper.