Practical deep learning with R

Sigrid Keydana (Rstudio, Münich, Germany)

Course description

This short course will introduce you to torch for R, an R-native port of PyTorch that requires no Python installation.
We start by a thorough introduction to the basics that make everything possible: tensors, automatic differentiation, and neural network modules.
Equipped with that knowledge, participants will explore two applications: time series forecasting and numerical optimization. While the former showcases renowned deep learning architectures, the latter demonstrates the usefulness of torch as a high-performance tensor-computation library, going beyond the deep learning context.
All modules will incorporate ample occasion for practice. This course does not presuppose familiarity with either deep learning concepts or frameworks; however, participants should have a basic knowledge of the R programming language, as well as of basic machine learning terminology.

Message to attendees

Hi, thanks for registering for Practical deep learning with R! I am looking forward to teaching the course.
Please install the following libraries before the course (all may be installed from CRAN):

• torch
• tidyverse
• tsibble
• tsibbledata
• lubridate
• fable
• zeallot
• nycflights13
• feasts

The single most important here is torch, but the others will be used in some of the exercises.
To check if torch was installed successfully, please run
torch_tensor(1)
and check for the expected result
torch_tensor
1
[ CPUFloatType{1} ]
If you have problems installing torch, please open an issue at https://github.com/mlverse/torch.
Finally, it would make sense to clone the tutorial repository https://github.com/skeydan/torch_tutorial_cnc2021 before we start.
Thanks!