1st:
Is there a way to lauch a single flow file on multiple processors (like 8 processors being used to denoise 1 .flow)
2nd:
I know that shhh.flow is modeling the error when denoising, so the more information there is, the better is the denoising, but I want to know what is the minimum sequences in a flow file to be sure to have enough error modeling and have the best denoising
Clearly, is there a time where the error doesn’t change anymore ?
Is there a way to lauch a single flow file on multiple processors (like 8 processors being used to denoise 1 .flow)
The fact is that it doesn’t really matter. We’ve run a single file on many processors and there’s no improvement. This turns out to be a product of the big compute steps being in the steps that are only run on a single processor.
Clearly, is there a time where the error doesn’t change anymore ?
Sounds like a great experiment that you could run Sorry, we don’t know.