Tabnet with R

This repository uses the tidymodels framework to fit a Tabnet Neural Network on the NHS Stranded data. This data is provides a binary classification problem. Tabnet is a Neural Network model which is supposed to provide competitive performance on tabular data. In this repository, the aim is to compare Tabnet's performance with traditional ML methods.

References

1. https://arxiv.org/pdf/1908.07442.pdf
2. https://mlverse.github.io/torchbook_materials/simple-net-tensors.html
3. https://blogs.rstudio.com/ai/posts/2021-02-11-tabnet/
4. https://github.com/mlverse/tabnet
5. https://mlverse.shinyapps.io/torch-tour/#section-torch-components