To simplify the pipeline and limit the data products, it makes sense to combine these things. There isn't much need for the intermediate cuboid data products, and if someone wants to look at them, cuboid remapping is quite fast to re-run.
This also requires some I/O changes to apply_hod, to allow it to handle the raw rho_to_halo outputs.
To simplify the pipeline and limit the data products, it makes sense to combine these things. There isn't much need for the intermediate cuboid data products, and if someone wants to look at them, cuboid remapping is quite fast to re-run.
This also requires some I/O changes to apply_hod, to allow it to handle the raw rho_to_halo outputs.