Just merged the last required pull request for 2.7.0

[Daniel Lee]

Hi all,

I just merged the last real-work pull request for 2.7.0 Stan. (Before it gets merged, there's a pull request for bumping version numbers and other administrative stuff, but that was the last required pull request before we release.)

I know this last pull request will break RStan and PyStan. I'm planning on helping out with that. But before I tagged, I wanted to make sure all the interfaces are compatible. Once we know that for a fact, I'll tag it. (I figured there's no point in tagging until we're sure we can release it.)

Daniel

[Daniel Lee]

I just checked RStan -- that looks ok. I was expecting compilation failures, but maybe it's all set.

[Jiqiang Guo]

No, it is not ok. I guess I can fix it.  

One question, what are the two parameters or streams of test_gradient.  I cannot figure it out from the following two lines of comments:

271      * @param o Output stream for messages.

272      * @param msgs Stream to which Stan programs write.

Jiqiang 

--
You received this message because you are subscribed to the Google Groups "stan development mailing list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to stan-dev+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Daniel Lee]

That should be Rcpp::Rcout and Rcpp::Rcerr from rstan.

How do you set it up? If I can build it on my side, I can help.

[Bob Carpenter]

> On Jun 21, 2015, at 10:01 PM, Jiqiang Guo <guo...@gmail.com> wrote:
>
> No, it is not ok. I guess I can fix it.
>
> One question, what are the two parameters or streams of test_gradient. I cannot figure it out from the following two lines of comments:
>
> 271 * @param o Output stream for messages.
> 272 * @param msgs Stream to which Stan programs write.

I can see how saying "o" is for messages and then calling
another stream "msgs" might be confusing! Sorry about the
confusing doc.

o: where the gradient program writes

msgs: where the Stan program that's running will write to
that is, target of the print() statement from a running
Stan program

- Bob

[Bob Carpenter]

> On Jun 21, 2015, at 5:18 PM, Daniel Lee <bea...@alum.mit.edu> wrote:
>
> Hi all,
>
> I just merged the last real-work pull request for 2.7.0 Stan.

I think the Wiener distribution should go in if it passes tests.

I also thought the variational inference branch would go in. Is
it waiting on something?

- Bob

[Daniel Lee]

Agreed. Just need to make sure the merge with develop is ok. I don't foresee any major problems.

[Ben Goodrich]

Joshua,

With R CMD check, I am seeing

omxMLFitFunction.cpp:19:30: fatal error: stan/agrad/rev.hpp: No such file or directory

Do you have this fixed locally? I think you can just do #include<stan/math.hpp> these days.

Ben

On Monday, June 22, 2015 at 1:36:11 AM UTC-4, Daniel Lee wrote:

Agreed. Just need to make sure the merge with develop is ok. I don't foresee any major problems.

On Jun 21, 2015 10:20 PM, "Bob Carpenter" <ca...@alias-i.com> wrote:

> On Jun 21, 2015, at 5:18 PM, Daniel Lee <bea...@alum.mit.edu> wrote:
>
> Hi all,
>
> I just merged the last real-work pull request for 2.7.0 Stan.

I think the Wiener distribution should go in if it passes tests.

I also thought the variational inference branch would go in.  Is
it waiting on something?

- Bob

--
You received this message because you are subscribed to the Google Groups "stan development mailing list" group.

To unsubscribe from this group and stop receiving emails from it, send an email to stan-dev+unsubscribe@googlegroups.com.

[Joshua N Pritikin]

On Mon, Jun 22, 2015 at 09:40:56AM -0700, Ben Goodrich wrote:
> With R CMD check, I am seeing
> omxMLFitFunction.cpp:19:30: fatal error: stan/agrad/rev.hpp: No such file or dir
> ectory
>
> Do you have this fixed locally?

Yeah, probably

> I think you can just do
> #include<stan/math.hpp> these days.

OK

--
Joshua N. Pritikin
Department of Psychology
University of Virginia
485 McCormick Rd, Gilmer Hall Room 102
Charlottesville, VA 22904
http://people.virginia.edu/~jnp3bc