/p8-pipeline

Primary LanguageJupyter Notebook

p8-pipeline

contains the proof-of-concept reconstruction pipeline used in this project-8. It uses Coordinate Descent to reconstruct an image of a Radio Interferometer. It reads CASA MS files and outputs a numpy array as image.

The actual code is located in ./pipeline

dependencies

The pipeline needs two python dependencies, which in turn require linux-only binaries.

pydata

pynfft

pipeline folder structure

./pipeline/single_run.py is the main project file, it contains the routines for loading simulated MS files and reconstructing an image with Coordinate Descent

  • ./pipeline/algorithms --> contains the coordinate descent implementation
  • ./pipeline/benchmark --> contains the simulated measurement sets
  • ./pipeline/debug_data --> contains matlab files from a different observation
  • ./pipeline/msinput.py --> contains modified pydata code for loading simulated MeerKAT MS
  • ./pipeline/nufftwrapper.py --> contains the wrapper code of the pynfft code