This is a Python and pyCUDA-accelerated implementation of the MORESANE deconvolution algorithm (see Dabbech et al. 2014 for a full description of the algorithm, and Dabbech et al. 2012 for earlier work).

Resaonbly up-to-date binary packages of PyMORESANE are available via the radio-astro repository. Look for the python-pymoresane package. If you must install from source, read on.

Requirements:

• numpy
• scipy
• pyfits
• CUDA
• pycuda
• scikits.cuda

You can use pip to install these:

$ pip install -r requirements.txt

Now you can use the setup script to install pymoresane:

$ python setup.py install

Running PyMORESANE after setup is as simple as typing runsane followed by the inputs you desire. runsane --help should provide some assistance, otherwise the docstrings/comments in the code are very detailed.

Example usage:

• Simple: runsane dirty.fits psf.fits output_name
• Long options: runsane dirty.fits psf.fits output_name --enforcepositivity
• Short options: runsane dirty.fits psf.fits output_name -ep

Feel free to contact me at [email protected] with comments, questions, criticism and bug-reports.