smops's Introduction

smops

smops (aka Smooth Operator) is a python package for interpolating channelised FITS model images onto user-specified higher resolution frequency grid. For example, you can give smops a set of 4 per-subband model FITS images, and ask it to return 16 model images, which can then be fed back into e.g. wsclean for a predict operation. Usage:

smops --ms /ms/used/togen/images.ms -ip prefix-used-for-those-images -co 16 -order 4

Its options are:

usage: smops [-h] [-v] [-op] [-j] [-s] [-mem] -ms  -ip  -co  [-order]

Refine model images in frequency

optional arguments:
-h, --help            show this help message and exit
-v, --version         show program's version number and exit
-op , --output-prefix
                        What to prefix the new interpolated model name with.
-j , --num-threads    Number of threads to use while writing out output images
-s , --stokes         Which stokes model to extrapolate. Write as single string e.g IQUV. Required when there are multiple Stokes
                        images in a directory. Default 'I'.
-mem , --max-mem      Approximate memory cap in GB

Required arguments:
-ms , --ms            Input MS. Used for getting reference frequency
-ip , --input-prefix
                        The input image prefix. The same as the one used for wsclean
-co , --channels-out
                        Number of channels to generate out
-order , --polynomial-order
                        Order of the spectral polynomial