Oversampling evaluates a light or mass profile with high precision, where an iterative grid may be used which increases the sub_size until a threshold precision is met.
This was previously implemented using a Grid2DIterate object. However, this led to a lot of code repetition with the Grid2D object and an unclear API in terms of how over sampling is performed.
This PR removes the Grid2DIterate object and moves the functionality into an OverSampleIterate object, which can be passed into a Grid2D in order to perform iterative oversampling.
This also led me to refactor the dataset package, removing Settings objects which wrapped the Grid2DIterate object in awkward ways, cleaning up the code even more!
A follow up PR will add a second OverSample object, which performs the normal over sampling which currently occur if the Grid2D does not have an over_sample object. This has the bold goal of removing the sub_size from all Mask and data structure objects, would has the potential to make the source code way cleaner. Wish me luck!
Oversampling evaluates a light or mass profile with high precision, where an iterative grid may be used which increases the
sub_sizeuntil a threshold precision is met.This was previously implemented using a
Grid2DIterateobject. However, this led to a lot of code repetition with theGrid2Dobject and an unclear API in terms of how over sampling is performed.This PR removes the
Grid2DIterateobject and moves the functionality into anOverSampleIterateobject, which can be passed into aGrid2Din order to perform iterative oversampling.This also led me to refactor the
datasetpackage, removingSettingsobjects which wrapped theGrid2DIterateobject in awkward ways, cleaning up the code even more!A follow up PR will add a second
OverSampleobject, which performs the normal over sampling which currently occur if theGrid2Ddoes not have anover_sampleobject. This has the bold goal of removing thesub_sizefrom allMaskand data structure objects, would has the potential to make the source code way cleaner. Wish me luck!