The template.fits (*.tpl.fits) is a tabulation of the spline model with an oversampling by a factor of about four. It is a relict of previous coadding methods. This leads to confusion and should be removed. Some refactoring will be needed.
Just storing the knot values (or the B-spline amplitudes) is really sufficient.
Instructions or a script, how the get the spline model, should be given.
The storing the flux as float32 should be also sufficient and will further reduce the size of serval output.
The
template.fits
(*.tpl.fits
) is a tabulation of the spline model with an oversampling by a factor of about four. It is a relict of previous coadding methods. This leads to confusion and should be removed. Some refactoring will be needed. Just storing the knot values (or the B-spline amplitudes) is really sufficient.Instructions or a script, how the get the spline model, should be given.
The storing the flux as float32 should be also sufficient and will further reduce the size of serval output.