Open biglinn opened 1 month ago
You can use the official code, this is already a very old code, and it feels like the interface of mkl is not suitable. In fact, mkl was not merged into oneapi when rewriting this code. So I suggest you try the latest official code.
I'll take a try. Thanks!
Hi, @wxmwy
I have met a error during the iteration of ShengBTE, like this:
So in the
BTE.KappaTensorVsT_CONV
, there are all 'NaN'. What is the problem? dalao, qiu dai!