hBayesDM on Mac M2

[Anastasia16c]

Hi everyone,

I am using Mac and ARM64 R and I am totally stuck.
Could someone advise me what packages I need to run hBayesDM on Mac?
I have installed from Cran and I am using Rstudio 2025 and R 4.5.1.
bandit2arm_delta returns: Error in is.null(data) || is.na(data) :
'length = 7560' in coercion to 'logical(1)'

thanks,
A

[Anastasia16c]

Hi again, so the errror has been reported again.
'The bug stems from the fact that given a data frame, is.na() returns a matrix indicating for each element of the data frame if it is NA. However, if() works only with scalar values. Also, && and || are meant to only be used with scalars. In earlier versions, when an input was non-scalar, R would make them scalar by taking only the first value of an input vector. But using non-scalars with these functions is almost always unintentional, and indicative of not actually testing what was intended to be tested. Thus, newer version fail on non-scalar inputs to if(), && and ||.'
However, my data is exactly as expected. Any ideas how to fix this?
thank you in advance :-)
Ax

[Jeongyeonni]

Hi,

Have you tried running the model with the example data provided in the hBayesDM GitHub?
Example data: https://github.com/CCS-Lab/hBayesDM/blob/develop/commons/extdata/bandit2arm_exampleData.txt

You can download the file and use it directly as the input data.

I’m using RStudio 2025 with R 4.5.1, and the model runs successfully with the example data. If it works on your side as well, the issue is likely specific to your dataset.

If it still doesn’t work, please try installing the developer version of hBayesDM:
devtools::install_github("CCS-Lab/hBayesDM", subdir = "R")

The errors you reported should be resolved in this updated version.

Please let me know if you encounter the same issue again.

Best regards,
Jeongyeon Shin

[Anastasia16c]

@Jeongyeonni thank you so much!
I did install the github version and it does work now. Thanks!!!

all the best,
Anastasia