rfpipe
A fast radio interferometric transient search library. Extends on rtpipe.
This library primarily supports offline analysis of VLA data on a single workstation. Integration with the real-time VLA and cluster processing is provided by realfast.
Planned future development include:
- Supporting other search algorithms.
- Extending support for GPUs.
- Supporting other interferometers by adding data and metadata reading functions.
- Remove dependence on rtpipe.
- Python 3 support.
Installation
rfpipe requires the anaconda installer on Linux and OSX. The most reliable install process adds two custom channels and a new build environment.
conda config --add channels pkgw-forge
conda config --add channels conda-forge
conda create -n realfast numpy scipy cython matplotlib pwkit casa-tools casa-python casa-data numba pyfftw
source activate realfast
pip install -e git+git://github.com/realfastvla/rfpipe#egg=rfpipe
Dependencies
- numpy/scipy/matplotlib
- pwkit casa environment (for quanta and measures)
- numba (for multi-core and gpu acceleration)
- rtpipe (for flagging; will be removed soon)
- astropy (<3.0; for Python2 and 3 compatibility)
- sdmpy
- pyfftw
- pyyaml
- attrs
- rfgpu (optional; for GPU FFTs)
- vys/vysmaw and vysmaw_reader (optional; to read vys data from VLA correlator)
Citation
If you use rfpipe, please support open software by citing the record on the Astrophysics Source Code Library at http://ascl.net/1710.002. In AASTeX, you can do this like so:
\software{..., rfpipe \citep{2017ascl.soft10002L}, ...}
