Feature Enhancement: auto save posterior summaries

[itsdfish]

Hi Rob-
Perhaps there is already a solution, but I could not figure out how to save either the Stan or Mamba posterior summaries. In either case, I was thinking it might be convenient to automatically save that along with the data and posterior sample files.

[goedman]

Hi Chris, That definitely makes sense for cases when reprocessing the Stan generated files takes a long time. Let me have a look at it. This weekend I'm still traveling but should be back in La Jolla next Tuesday. Best, Rob

…

Sent from my iPhone

On Jul 28, 2017, at 08:27, fisherc2 ***@***.***> wrote: Hi Rob- Perhaps there is already a solution, but I could not figure out how to save either the Stan or Mamba posterior summaries. In either case, I was thinking it might be convenient to automatically save that along with the data and posterior sample files. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[itsdfish]

Indeed, I have found myself in that very situation. Thanks for adding that feature.

Best regards,

Chris

[goedman]

Hi Chris, haven't forgotten your request.

Turns out Mamba has build-in features for writing the output to a file (http://mambajl.readthedocs.io/en/latest/tutorial.html?highlight=write#file-i-o).

In Stan.jl I receive the CmdStan summary as a String, I need to add a function to store that string to a file in the /tmp directory.

As a result of our trip this summer, we have decided to move (after 27 years CA!) to live part time in CO and part time in DC. This will definitely take many cycles, so I apologize in advance if I'm slow in responding until at least October.

[itsdfish]

Hi Rob-
Thanks for the update and no worries. It looks like the Mamba feature will hold me over.

Good luck with your new homes and have a safe trip!

[goedman]

I will close this issue for now in the hope the Mamba feature will suffice.

My plan is to refactor Stan.jl (over the next year) for Julia 1.0. During the refactor I will keep this request in mind.

[itsdfish]

Hi Rob-

I think Mamba suffices for the most part. The only downside is that it seems to compute some diagnostics differently from Stan. So a person may not be able to reproduce certain results from Python or R, but its not a huge deal. I would say consider including this feature if the solution is simple. Otherwise, no worries.

[goedman]

Hi Chris, Thanks for the update. I’ll keep this in mind. Hadn’t realized these different diagnostics! Best, Rob Sent from Rob Goedman's iPad Pro

…

On Apr 22, 2018, at 8:20 AM, fisherc2 ***@***.***> wrote: Hi Rob- I think Mamba suffices for the most part. The only downside is that it seems to compute some diagnostics differently from Stan. So a person may not be able to reproduce certain results from Python or R, but its not a huge deal. I would say consider including this feature if the solution is simple. Otherwise, no worries. — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub, or mute the thread.