Jacki Buros Novik

I agree - some option to restart a failed job a max of X times would be really helpful. In the case of a node going down (say, because it's...

FWIW, this example works fine if these lines are removed: https://github.com/arviz-devs/arviz/blob/main/arviz/data/io_cmdstanpy.py#L594-L595 I don't know if it will break anything else, but I can make a quick PR ..

@bmorris3 This hasn't been integrated into the main branch due to lack of testing, but I made this change which resolved the problem for me: https://github.com/arviz-devs/arviz/compare/main...jburos:fix-cmdstanpy-length1?expand=1 I'd be curious if...

Noting here, in response to the current PR, that this bug persists with the latest version of cmdstanpy (`1.0.1`), and cmdstan-2.29.2. Error output: ```py In [6]: idata1 = az.from_cmdstanpy(fit1, coords...

Seems relevant to this conversation, that in our workflow we are using InferenceData to store output from cmdstanpy as netCDF files, & finding it's pretty easy to use them in...

Another option along this line is to expand support for the experimental [tbl_cube](https://rdrr.io/cran/cubelyr/man/tbl_cube.html) data structure in R. This is very similar to the `xarray.DataSet` structure in Python, and has potential...

@mike-lawrence This is a good question. There are two reasons for this: 1. We're storing the nc files in the cloud, and the netcdf format doesn't support read-access over network...

Yep. I just need to find the one that yielded the error.

[Here](https://storage.googleapis.com/empirical-jm/example_bad_log_lik-1.Rds) is an Rds containing the log-lik values from one of the fits yielding that error. I have found several examples -- not sure which is the one I was...

It's been a while since this issue was addressed, but posting my work-around here in case it's helpful to others. My process now is: 1. read in posterior-draws from CSV...