Fu Yin

Thanks for your help! I have tried, same results. Actually when I use conda build . in my m1-max, I got same hint but it's a warning. In azure pipeline...

I don't think so, conda-forge doesn't have the 'libopenblas'

Yaa so weird it has one, I will try! Thanks

Yes it's weird, but the document of xtensor-blas said it use openblas, also provide an example using find_package in cmake file. I literally don't know how it works

It works! Add the `libopenblas` is right, thanks @tdegeus. BTW I have learned a lot from some of your packages and code style, so much thanks！

We can close this

Thanks for your help!!! Maybe I can publish my first package in `conda-forge` in a few days. Thanks!

Hey, Sorry to bother you again @tdegeus and I am faced with a very weird problem. You know that I want to publish the package to conda-forge. I have successfully...

BTW I want to add OpenMP to for loop, that's the reason why I need a loop.

Thanks for your reply, actually when I just test 1-d array by xt::mean and np.mean, if the 1-d array is a little large, xt::mean will also be very slow than...