This will be used when a user does not specify over sampling and thus ensures all normal use cases use appropriate over sampling.
In PyAutoLens, this adaptive sub-grid is only used for the foreground lens light model, as it requires that a uniform grid is used in order to set up. To ensure lensed sources are resolved correctly, the workspace has been updated to use cored Sersic profiles by default, which do not require high levels of over sampling.
The following guides illustrate how a user can perform fits with appropriate over sampling for lensed sources which are not cored:
Extends adaptive over-sampling to by default use high levels of over sampling in the centre of a light profile and lower values in the outskirts.
The default settings use a 32 x 32 subgrid for the very central pixels, a 4 x 4 subgrid further out and 2 x 2 subgrid for all other pixels.
The following guide illustrates and describes this feature:
https://github.com/Jammy2211/autogalaxy_workspace/blob/main/notebooks/guides/over_sampling.ipynb
This will be used when a user does not specify over sampling and thus ensures all normal use cases use appropriate over sampling.
In PyAutoLens, this adaptive sub-grid is only used for the foreground lens light model, as it requires that a uniform grid is used in order to set up. To ensure lensed sources are resolved correctly, the workspace has been updated to use cored Sersic profiles by default, which do not require high levels of over sampling.
The following guides illustrate how a user can perform fits with appropriate over sampling for lensed sources which are not cored:
https://github.com/Jammy2211/autolens_workspace/blob/release/notebooks/guides/over_sampling.ipynb