Looping through a vector of paired indices (i,j) vs two for loops

[Daeil Kim]

I'm looking to do some link prediction on a graph using the Mixed Membership Stochastic Block model. To get AUC scores, I need to loop only through a subset of edges so that I could use my learned latent variables to predict the rest. I'm wondering what the best way of enforcing that would be. Is is possible to loop through a vector containing a list of valid edge indices? Most of the examples I'm seeing show a double for loop to loop through the generation of edges.

[Bob Carpenter]

There's an LDA example in the manual, but given your question,
you might find the chapter on sparse and ragged data structures
more helpful.

- Bob

> On Apr 1, 2016, at 4:56 PM, Daeil Kim <daei...@nytimes.com> wrote:
>
> I'm looking to do some link prediction on a graph using the Mixed Membership Stochastic Block model. To get AUC scores, I need to loop only through a subset of edges so that I could use my learned latent variables to predict the rest. I'm wondering what the best way of enforcing that would be. Is is possible to loop through a vector containing a list of valid edge indices? Most of the examples I'm seeing show a double for loop to loop through the generation of edges.
>
>

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