error with BART 4 para model

[Kshitij Kumar]

Hello, I am currently using BART with 30 trials, single balloon task with no significant modifications, but I'm encountering failure when running the model. I attempted to address the issue by filtering the data using mean pumps and running the model again, but unfortunately, it did not resolve the problem. 

Could someone provide assistance? 

I have included a screenshot of the issue for reference.

[Jeung-Hyun Lee]

Hi Kshitij,

The error you are getting (Stan model does not contain samples) can happen for many reasons so it would be best if you could share the Stan model (if you have made any modifications) and small reproducible example. In this way, we can help you figure out what the problem is.

Best, 

JH

2023년 7월 7일 금요일 오전 6시 30분 6초 UTC+9에 kshtjku...@gmail.com님이 작성:

[Kshitij Kumar]

Thank you for such quick reply!

I am not sure what is going wrong in the model. the model i am using is this: output <- bart_par4( data = "example", niter = 2000, nwarmup = 1000, nchain = 4, ncore = 4) 

https://search.r-project.org/CRAN/refmans/hBayesDM/html/bart_par4.html

I am not sure about Stan models, sorry am very new to this.

total sample are 1200+,  here is an example csv.

[Jeung-Hyun Lee]

In your dataset, the necessary data columns (subjID, explosion) that hBayesDM look for are missing. Could you rename the column names and try fitting it again? 

Best,

JH

2023년 7월 7일 금요일 오후 4시 48분 31초 UTC+9에 kshtjku...@gmail.com님이 작성:

[Kshitij Kumar]

Hi sorry about the confusion, while running it i changed the names to the same columns as that of the model parameters, i uploaded the unedited file here, please find the correct file attached, just the column names have been changed. My sincere apologies for the same.  

[Kshitij Kumar]

hi, i am still getting the same error, can you please help!

[Kshitij Kumar]

hi, could someone please help with the issue ?

regards

[봉수현]

Your file is csv. It should be txt! this sample data works on my machine.

2023년 8월 2일 수요일 오전 4시 7분 51초 UTC+9에 Kshitij Kumar님이 작성:

[Kshitij Kumar]

hi, when i run it for the .txt file i get these errors:

[Jeung-Hyun Lee]

You can ignore the warning but also may consider changing the initial values by yourself. Currently, the default setting uses vb estimate as the initial value but users can change it to “fixed” or “random", as needed. Please check out this link for the details. 

Best, 

JH

Jeung-Hyun Lee

PhD candidate,

Computational Clinical Science Lab.

Department of Psychology

Seoul National University 

Building 16, Room A107

Email : jhleeangel@gmail.com / jhlee...@snu.ac.kr 

Website : https://jeunghyunlee.github.io/ 

2023년 8월 8일 (화) 오후 1:40, Kshitij Kumar <kshtjku...@gmail.com>님이 작성:

--
You received this message because you are subscribed to a topic in the Google Groups "hbayesdm-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/hbayesdm-users/cBqBn-pTgto/unsubscribe.
To unsubscribe from this group and all its topics, send an email to hbayesdm-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hbayesdm-users/5b7e4945-7c51-4de0-9f11-8cbc20cf6c07n%40googlegroups.com.