A Radio Astronomy Software post-processing tool for RIMS Dynamic Spectra output. The tool will produce a noise estimate and smooth the target dynamic spectra according to the provided RIMS output and specified parameters. In essence, it serves as a companion tool to RIMS, where it will improve and neatly plot (png) the dynamic spectra (fits) from RIMS.

git clone https://github.com/ratt-ru/inspect-dynspec.git
pip install ./inspect-dynspec

View help:

inspect-dynspec --help

Run job:

inspect-dynspec <path-to-RIMS-output> --kernel 20,5 --kernel 10,10 --output <path-to-desired-output-folder> --stokes IV --std-scale 2.5

Notes:

• You may specify multiple smoothing kernels as shown above.
• You may specify which Stokes dynamic spectra you want processed and plotted.
• The specified output directory will be created if it is not.
• The input path need not point exactly to the output of RIMS folder, as the tool will traverse down until it finds a RIMS-like output directory (to make sure it gets the right one though, you should probably be exact).

Depending on your parameters, the output folder will contain an assortment of png plots. Suffix identification:

• *_a_e_denoise.png : A beam/Jy fully denoised (analytically and excess) and smoothed target dynamic spectra - usually the product you want.
• *_W.png : A plot of the Weights produced by RIMS (used in analytical denoising).
• *_W2.png : A plot of the Weights^2 produced by RIMS (used in analytical denoising).
• _flagged_regions.png : A mask plot of scan breaks and flagged frequencies.
• _a_denoise.png : Analytically denoised and smoothed target dynamic spectra (beam/Jy).
• _rawdata.png : The smoothed raw target dynamic spectra (beam/Jy).
• _SNR.png : A fully denoised and smoothed target dynamic spectra with the smoothed calculated standard deviation divided out to give an SNR plot.
• _var_a.png : The analytical variance calculated from the W and W2 weights. It is applied to the rawdata to produce the analytically denoised product.
• _var_e.png : The excess variance calculated from the RIMS off target dynamic spectra. It is applied to the analytically denoised product to produce the fully denoised product.

MIT License

Copyright (c) 2025, Talon Myburgh, Cyril Tasse, Oleg Smirnov, Landman Bester, Rhodes University Centre for Radio Astronomy Techniques & Technologies (RATT), Observatoire de Paris