[This article was first published on RBelgium open content
, and kindly contributed to R-bloggers
]. (You can report issue about the content on this page here
Want to share your content on R-bloggers? click here
if you have a blog, or here
if you don't.
You can see the recorded presentation below of the RBelgium session we held last week (on Thursday 2022-02-03).
Christophe Regouby (co-author of the tabnet R package) talked about
- the design of the the recent torch R package which interfaces with the libtorch C++ library for deep learning
- his torch for R cheat sheet available here
- the high-level setup of tabnet and the use of tabnet for supervised and unsupervised modelling
- how to run tabnet from R and how to visualise the model results
- recent additions he made to the package at https://github.com/mlverse/tabnet
And answered the numerous questions from the audience.
Enjoy the recordings below, download the presentation and if you have further questions, you can join the tabnet/torch discussion at the discord channel at https://discord.com/invite/s3D5cKhBkx
You can be kept up to date at upcoming meetups at https://www.meetup.com/RBelgium.
If you have topics to cover, feel free to contact one of us through Meetup and we will be happy to host it.