Routines and procedures for calibrating and reducing raw COMAP spectroscopic single-dish radio data.
Prerequisites are listed in an approximate order in which they should be installed:
- Python 3.X.X
- SLALIB - Astronomical libraries for any spherical trig.
- OPENMPI/MPICH - The pipeline requires MPI to run, either backend will work fine.
- MPI4PY - MPI4Py must be compiled against the mpi library above.
- HDF5 --parallel-enabled - An mpi ready version of the HDF5 libraries (see link for instructions)
- H5PY - Must be compiled against above HDF5 library (not the one included with anaconda )
- HEALPY - Either against your local version of HEALPix or a pre-compiled version from pip.
Other standard libraries such as NumPy, SciPy, matplotlib, etc... are assumed to be installed already.
After all prerequistites are installed first install the library by
cd /path/to/COMAPreduce/
python setup.py install
You will then need to setup a working directory that will contain:
- run.py
- All the .ini files
- COMAP_FEEDS.dat
mkdir /path/to/working/directory/
cp /path/to/COMAPreduce/run.py /path/to/working/directory/
cp /path/to/COMAPreduce/*.ini /path/to/working/directory/
cp /path/to/COMAPreduce/COMAP_FEEDS.dat /path/to/working/directory/
Three example parameter files (the .ini files) have been provided.
To run the pipeline you will need to both choose a parameter file and generate a list of files. In
COMAPreduce/comancpipeline/scripts/io/
there is script called createFileList.py that can help to generate a filelist, it is executed as
python createFileList -D /path/to/level1/files -o "string describing observation you need (e.g. TauA)" -F output_filelist_name.dat
Finally, to run the pipeline you invoke in your working directory
python run.py -P parameterfile.ini -F filelist.dat
and for MPI executions
mpiexec -n X python run.py -P parameterfile.ini -F filelist.dat
where X is the number of cores you want to use.