8000 GitHub - tronsgaard/songpipe: Pipeline for SONG
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

tronsgaard/songpipe

Repository files navigation

SONG pipeline

songpipe is the new default extraction pipeline for SONG (Stellar Observations Network Group), currently supporting the Tenerife and Australia nodes.

songpipe runs in Python (version 3.8 or higher) and works as a wrapper around the core routines of PyReduce (Piskunov & Valenti (2002), Piskunov, Wehrhahn & Marquart (2020)). As PyReduce was not being actively maintained for a period, and some adjustments of the code were necessary, songpipe uses a fork of the PyReduce repository, located here. Most of our changes have since been implemented in the official PyReduce repository, except some changes in header keywords that are difficult to resolve without breaking compatibility (see pull request).

songpipe is still being developed (see below).

Installation

  • Create and enter a virtual environment (venv /path/to/new/virtual/environment)
  • Download the modified version of PyReduce (https://github.com/tronsgaard/PyReduce) and install it using pip install --editable . from within the folder (the --editable flag can be omitted, but it makes it easier to update the code).
  • Download the latest release of songpipe or simply checkout this repository (git checkout https://github.com/tronsgaard/songpipe.git).
  • Install required packages listed in requirements.txt.
    • Note: Packages dill and colorlog are optional and only required for FITS header caching and colored terminal output, respectively.

Running

songpipe runs from the terminal, just execute run_australia.py or run_tenerife.py located in the root of the repository. For full syntax, call the script with --help.

The scripts must be called with a date string (YYYYMMDD) as the first argument. The keywords --rawdir DIRPATH and --outdir DIRPATH speficiy the location of the input and output directories. If called with --confirm, the script will pause and ask the user to confirm the settings and directory structure before proceeding.

Known issues

  • astropy.io.fits regularly complains about truncated FITS files, but no actual issues with the files have been found.
  • Extraction of single files using --extract keyword is currently not working as expected.

Planned development

  • Include subtraction of scattered light
  • Include order curvature for Tenerife
  • Include bad pixel mask
  • Support pre-CMOS Tenerife data
  • Multiprocessing support
  • Automatic deletion of intermediate files (prep/ directory)
  • Easier installation with pip (create setup.py)

Further reading

About

Pipeline for SONG

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

0