Völgyes, David

Hi, Thanks for the prompt feedback. :) It is not urgent at all, but it is nice to be fixed. I would rather avoid poking CI/CD. I am also quite...

The primary question is about features. The parallel programming part is not easy, but also not too hard, if the expected behavior is defined. Basic assumptions: - loading one volume...

Hi, So more or less i have a working implementation, but there are a few issues to sort out. First, there is a new queue class, let's call it somehow....

> Hi > > What happens if in the end less than V volumes are present? E.g. 4 volume long queue, 11volumes in tota. > > ``` > > *...

That is a good point. First, in the current implementation there is no way to set number of processes, it will be equal to number of cores at max, or...

Hi, Number of visible CPUs: i use the pure python multiprocessing tools, so pytorch don't limit (and cannot limit) the number of cores/processes. Also i got pretty decent results even...

There are a few questions, especially about the last graph: do you have the same number of subjects for the two queues? I mean: i am pretty sure that still...

Hi, I added a new implementation, let's call it double buffering. The idea is more or less the same as in the pytorch dataloader: start loading the new subject as...

Hi, Memory: since it is also present (somewhat) in the classic queue, i would guess it is somewhere in your code, but every refactoring of mine also means some extra...

I need a few day break (i have some other pressing issues), but i will be back soon with a new variant. Besides memory consumption, there are two major concerns...