R hat too big

[TomasPapantos]

Hi All!

I've trained multiple LMMM with different datasets and I was always able to get an r_hat near to 1 (from running the mm.fit() and then mmm.print_summary()), sometimes with more chains or more warmups, but I'm now using a new dataset and r_hat is very high despite using extremelly high warmups/chains (upto the point that the machine ran out of capacity).

By the way, the data manipulation code is 95% the same between previous LMMMs and the new one, and the LMMM portion of the code is 100% the same.

Does someone have any clues on what could be happening and how I might be able to get the r_hat to get to 1 or close to 1?

Thanks in advanced!
Tom