more pythonic file organization about pypulseq HOT 3 CLOSED

@grlee77 Thank you for your inputs. The 1:1 mapping was purely intended to provide familiarity to users jumping from Matlab. What do you think about rf and grad sub-packages instead of modules? I can imagine that a single rf.py module would be not so pretty to look at.

from pypulseq.

I am not saying you have to make any change here, it was just a suggestion. In your case, it may be beneficial to continue to mirror the structure of the Matlab code if you anticipate most users are migrating from Matlab and are more familiar with that structure. In that case, just go ahead and close this issue.

I can imagine that a single rf.py module would be not so pretty to look at.

I don't really agree. For Python projects, it is very common to group several related functions together in a single file. As a concrete 3rd party example, see the imports at the top of init.py for the filters submodule of scikit-image. Here there are several files within the filters folder, some of which have several functions internally, but they are grouped in a logical manner (e.g. edges vs. thresholding). To me, that seems better than if every individual function in that module were stored in its own file.

from pypulseq.

Thanks for sharing the scikit-image example, I suppose we will stick to the MATLAB convention to ensure familiarity. Once again, thanks for your insights, I learnt something new.

from pypulseq.