search

cmu-phil / tetrad

Repository for the Tetrad Project, www.phil.cmu.edu/tetrad.
GNU General Public License v2.0
404 stars 111 forks source link

Clark - New algorithms that are needed #596

Closed espinoj closed 2 years ago

espinoj commented 6 years ago

A. Kun's algorithm for data from multiple structures. Madelyn has coded it and given it preliminary testing. Need to test further and put description on Arxiv.

B. Fask Ramsey algorithm for fMRI time series.

C. Two-Step Kun algorithm for fmri time series.

D. FGES without one-edge faithfulness

espinoj commented 6 years ago

FGES has an option for one-edge faithfulness

image

jdramsey commented 6 years ago

@espinoj FASK is done.

espinoj commented 6 years ago

Two step is "done" in matlab

espinoj commented 6 years ago

@cg09 Is Kun's algorithm for data from multiple structures somewhere?

cg09 commented 6 years ago

Probably in matlab on his computer.

On Tue, Aug 21, 2018 at 2:25 PM, Jeremy Espino notifications@github.com wrote:

@cg09 https://github.com/cg09 Is Kun's algorithm for data from multiple structures somewhere?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/cmu-phil/tetrad/issues/596#issuecomment-414774536, or mute the thread https://github.com/notifications/unsubscribe-auth/APmNuSBHiCmdbxzNWYJwTSyafAa8wYEYks5uTFCBgaJpZM4QdfmE .

cg09 commented 6 years ago

Hi Jeremy (CC: Clark),

Exactly as Clark said, I have MATLAB source code on my computer. Do you need it? If so, please let me know exactly what function you need so that I can update the interface for you.

Best,

Kun

From: Clark Glymour cg09@andrew.cmu.edu Sent: Tuesday, August 21, 2018 2:35 PM To: cmu-phil/tetrad; Kun Zhang Subject: Re: [cmu-phil/tetrad] Clark - New algorithms that are needed (#596)

Probably in matlab on his computer.

On Tue, Aug 21, 2018 at 2:25 PM, Jeremy Espino notifications@github.com<mailto:notifications@github.com> wrote:

@cg09https://github.com/cg09 Is Kun's algorithm for data from multiple structures somewhere?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/cmu-phil/tetrad/issues/596#issuecomment-414774536, or mute the threadhttps://github.com/notifications/unsubscribe-auth/APmNuSBHiCmdbxzNWYJwTSyafAa8wYEYks5uTFCBgaJpZM4QdfmE.

jdramsey commented 5 years ago

@espinoj @cg09 Regarding Two-Step, which we have a Matlab java jar for, I recall the problem we were having with it, putting in the interface, was that you couldn't stop it once you started it, and it took a very long time. the only way to stop it was to quit the program. However, in the spirit of realizing things, that doesn't matter at all if you put it in the command line version or the R or Pascal versions, where you'd want to top the program anyway if you want to stop it. Proposal: Don't put Two-Step in the interface; instead, put it in the command-line Tetrad (and R and Pascal).

jdramsey commented 2 years ago

This is a nonstarter. Accessing Matlab from Java is not stable. Closing.