tatyana-perlova

I still don't know what the problem was, but it is solved by not importing Seurat in the same R session.

Thank you for getting back to me! I'm using version 0.3.3. It's not using gpu though (model.gpu is False).

Hi, thanks a lot! Got it. I though about analyzing it by cell type/cluster. The problem is that then the number of resulting metacells will be determined by original number...

Great, thanks a lot! I'll try it out.

I think you will have to reinitialize the model, but then you can just assign precomputed kernel matrix and archetypes from saved model.