How to use command line interface?

[SimonG03]

Hi everyone,

I am interested in the command line interface (CLI) of Stat-JR, and how I can use it for analysing my data. I use R for most of my computations; I would like to run Stat-JR from the command line rather than using the GUI, so I can stay within R if possible.
My Question(s): How can I use the CLI of Stat-JR? Is there documentation explaining this? Suppose I have already fit a model in Stat-JR, thus generating the code necessary to run it, can I use the CLI to fit the same model again using R?

So far, I've been using Stat-JR 1.0.2 via the browser-based GUI. As noted in [1] (see link below), there is a CLI for Stat-JR, which appears to be "under development", and which would allow Stat-JR to be run from external software. A neat feature, considering that most of the operations seem do be done by python and pre-compiled C++ code.
In the advanced documentation [2], I found a reference to a "Python command line interface". However, it is not discussed in that document how it can be used. In "~\StatJR\src\apps\scripting" there is an executable file called "runScript.exe", which triggers the same loading process as the main executable, but it doesn't start the TREE interface.

Is this the exetutable for the CLI? If so, how can I use it? Are there help files or man pages that explain the use of the CLI? If so, I would be happy if someone points me into the right direction; if not, I would be even more happy if someone could explain a few basic steps on "how to get started"

Thanks in advance! I'm looking forward to using this feature!

Best,
Simon

[1] http://www.bristol.ac.uk/cmm/software/s ... port/faqs/
[2] http://www.bristol.ac.uk/cmm/media/migr ... vanced.pdf

[ChrisCharlton]

Yes, it is possible to run Stat-JR templates from a command line, although as you have discovered the support for this is preliminary and the exact mechanism may change in the future (if you do attempt to use it we would be interested in your feedback). In another topic in this forum I have provided some syntax for running a Stat-JR model from R and then retrieving the MCMC chains:

https://www.cmm.bristol.ac.uk/forum/vie ... =879#p2201

Let us know if anything is unclear.

[SimonG03]

That already helps a lot. However, I am not sure yet if I stick with Stat-JR in this case because that would require me to write a new template to run in this manner.

Is there, perhaps, a good documentation on how to extend the current templates in Stat-JR? In the advanced documentation, there is a bit about extending templates and writing new ones, but it seems like great familiarity with Python and C++ is required to write them. Judging from that, the amount of work I would need to invest in that are outweighted (by far) by switching to a different program, even though I am genuinely interested in using Stat-JR.

I will have a look at the documentation and the example you provided. I'll let you know if there is any progress. I'd be happy to hear about a documentation on creating templates.

Best,
Simon

[richardparker]

Hi - thanks for your interest in the software. The forum message Chris linked to illustrates how to call Stat-JR from R and bring the results back. It's not actually necessary to write a Stat-JR template in order to do that (one can use the templates provided in the Stat-JR package), although it would be necessary to write a Stat-JR template if one wanted to extend Stat-JR's functionality.

Best wishes,

Richard