How to run R3 on command line

[rubenSaro]

Hi,

In the GUI to run the R3 algorithm, you need an undirected graph box (for example, from FAS) and a dataset box. How can a R3 search be done, if possible, in Tetrad Command Line?

Thanks

[jdramsey]

Ruben, was this ever made to work for you?

[rubenSaro]

no.

On Sun, Aug 29, 2021, 6:31 PM Joseph Ramsey @.***> wrote:

Ruben, was this ever made to work for you?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/cmu-phil/tetrad/issues/1326#issuecomment-907887219, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJUZSFLJ5TV3GYHIMRJ257DT7KYKXANCNFSM45XS3Z4Q . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[jdramsey]

Problem is, there's a skeleton crew working on that, but I may be able to create a "funnel". You want to run FGES and then R3? I could make an algorithm that just does that. Wouldn't involve any new syntax.

[jdramsey]

Yeah tell me exactly what you need and I'll make it.

[rubenSaro]

that is a good idea...it was actually a user asking about how to run R3 in command line, a while ago...and I remember I couldn't do it when command line tetrad first was published. Maybe wait for the crew to make it available in the future? or make available FGES + R3 as an option as you are suggesting?

On Sun, Aug 29, 2021, 6:49 PM Joseph Ramsey @.***> wrote:

Problem is, there's a skeleton crew working on that, but I may be able to create a "funnel". You want to run FGES and then R3? I could make an algorithm that just does that. Wouldn't involve any new syntax.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/cmu-phil/tetrad/issues/1326#issuecomment-907889962, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJUZSFN5DXYHXUN6AMZKRXLT7K2RPANCNFSM45XS3Z4Q . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[jdramsey]

Either is fine, although the second is easier for me to do on my own.

[jdramsey]

@kvb2univpitt I'm not sure whether this issue is yours or mine, since there are two ways to do it:

1. Add parameters to the pattern algorithms to do a subsequent pairwise orientation step (my issue).
2. Add a way in the command line to give an external graph as input to the pairwise algorithms (your issue).

Which way should we go?

[jdramsey]

I'm more and more thinking that there should be a flag in causal-cmd for this, that makes causal-cmd read in a graph from a file as an “external graph” and pass it to the algorithm, if the algorithm takes and external graph, like --externalgraph [path_to_external_graph.txt].

The implementation should be straightforward I think.