Run in the cloud option for Stan

Interfaces
1

@s.maskell raised the issue of having an easy way to run a Stan model in the cloud at Thursday’s meeting. I thought it an excellent idea and didn’t want to let it drop.

There is an server interface: GitHub - stan-dev/httpstan: HTTP interface to Stan, a package for Bayesian inference. that was mentioned that covers half the problem (hosting). Client should be fairly straightforward.

@storopoli suggested an option at the cmdstanR level, something like:

library("cmdstanr")
model <- cmdstan_model("my_slow_model.stan", server = "https://monster_server_farm",...)
fit <- model$fit(data = stanData)

Someone, can’t remember who, mentioned that RStudio already has a launcher feature that may already offer full on hosting since they do run Stan in RStudio cloud. I have sent an email asking about pricing, what interfaces are run etc… This is a GUI integrated ‘run in the cloud’ service on first examination: Using Background Jobs in the RStudio IDE - RStudio :: Solutions

I’ll report back on the RStudio Launcher feature.

Please add ideas/elaborate as you see fit.

Breck

1 Like
2

@breckbaldwin: I posted but potentially with obscure/erroneous topics. See here: Seamlessly Running Stan Much Faster (using AWS/HPC etc) - #2 by s.maskell

Not sure if it matters that we have parallel threads.

3

We should merge the topics, unaware of how to do that. Whether it is ‘general’ or ‘interfaces’ is up for interpretation. I’ll add my post to your thread and see if I can shut this one down.

B

5

there are many ways to run Stan in the cloud, depending on what you mean by “running in the Cloud”.
for teaching purposes, Jupyter notebooks + Google colab as host are a possibilitiy. blog post here: Stan Notebooks in the Cloud

I think a lot of people are using docker containers on AWS, judging from the number of previous threads on this topic on the forums - cf Docker & Stan ?!