Hi - thanks for looking into this. I don't think we'll want to make this change, because the current implementation does precisely what is documented in the package README, which follow the convention in the referenced paper: https://www-user.tu-chemnitz.de/~potts/paper/nfft3.pdf
There may be good arguments for using a different normalization, but I'd prefer to stick to those conventions here.
Hi - thanks for looking into this. I don't think we'll want to make this change, because the current implementation does precisely what is documented in the package README, which follow the convention in the referenced paper: https://www-user.tu-chemnitz.de/~potts/paper/nfft3.pdf
There may be good arguments for using a different normalization, but I'd prefer to stick to those conventions here.