This will be the first official public release of the VItamin code base. VItamin is a python package for producing fast gravitational wave posterior samples.
Why is the reconstruction loss summed on line 581 over the 1st and 2nd elements of the 2nd dimension?
It looks like this a hard coded assumption that there are 2 von mises parameters. I think at present that is true with phi_jl and phi_12 but it isn't always the case.
Why is the reconstruction loss summed on line 581 over the 1st and 2nd elements of the 2nd dimension?
It looks like this a hard coded assumption that there are 2 von mises parameters. I think at present that is true with phi_jl and phi_12 but it isn't always the case.
Just sum it over the 2nd dimension.