STAN in the cloud: python/R

We’ve now ran a few STAN training course with @betanalpha and Sarah Heaps. Due to the usual issues of participants installing software on their own machines, we’ve become fairly adept at running Stan via RStudio and/or python notebooks in the cloud. Our typical setup is:

  • URL: XXX.jumpingrivers.training
  • Running in a docker container on Digital ocean
  • Usernames/passwords for each participant
  • Sudo user for the trainer
  • Each user has course material in their home area

If you run training courses and would like to have this as a backup (or the main event), feel free to contact me.

We’re are happy to provide (free) advice on how to set it up or provide it as a (paid) service.


At some point, we’ll write a blog post on the details, but there’s always something else that gets in the way.

3 Likes

These instances are great and provide seamless support for attendees who have their own machines but are not able to set up Stan Windows .

1 Like