Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/computing/software/miriad/doc/rmclean.html
Дата изменения: Unknown
Дата индексирования: Sun Apr 10 22:39:44 2016
Кодировка:

Поисковые слова: universe
rmclean
[ Basic Info | User Guide ]

Basic Information on rmclean


Task: rmclean
Purpose: Run RM-CLEAN on a set of 'dirty' rm-cubes
Categories: analysis.

        RMCLEAN reads in dirty Q and U cubes, generates rmtf based on
        the frequencies given in an ascii file, and cleans the rm
        spectra following the algorithm given by Brentjens (2007).
        The output cubes contain the clean model components, the
        CLEANed rm spectra and the residuals.
        The input cubes must be reordered with mode=312, and the
        output cubes will have the same ordering and thus must be
        reordered after being written to disk. A script (runrmclean.py)
        is provided to put everything together into one step.

Key: inq
        Cube containing dirty Q spectra. No default.

Key: inu
        Cube containing dirty U spectra. No default.

Key: freq
        Ascii file containing the observed frequencies. No default.
        The file should contain one frequency per line, in Hertz.
        Comment lines aren't allowed.

Key: weight
        Ascii file containing the weights. Default is equal weights.
        The file should contain one weight per line, in the same
        order as is given in the frequency file, and the same length.
        Comment lines aren't allowed.

Key: nmax
        Maximum number of iterations per pixel. Default 1000.

Key: gain
        CLEAN loop gain. Default 0.1.

Key: cutoff
        CLEAN cutoff (in P, without bias correction). Default 0.

Key: fwhm
        FWHM of restoring RM beam, in rad/m^2. Default 0==auto.

Key: method
        Method for determining clean component locations. Choices are:
        'peakp': Search P spectrum for largest value
        'xcorr': Perform the cross-correlation of the RMTF and
                 the P spectrum at each iteration, and find the
                 location of the largest value in the amplitude
                 of the cross-correlation spectrum.
        The cross-correlation option is slower, but should be better,
        especially in low signal-to-noise situations, and is default.

Key: outq
        Output (CLEANed) Q cube. Default 'Qclean'.

Key: outu
        Output (CLEANed) U cube. Default 'Uclean'.

Key: modq
        Model (CLEANed) Q cube. Default 'Qmodel'.

Key: modu
        Model (CLEANed) U cube. Default 'Umodel'.

Key: resq
        Residual Q cube. Default 'Qresid'.

Key: resu
        Residual U cube. Default 'Uresid'.

Key: ni
        Number of CLEAN iterations per pixel. Default 'niters'.

Key: derot
        Derotate the spectrum? 0=no, 1=yes. Default 1.

Key: debug
        Debug mode? 0=no, 1=yes. Default 0.
 
  $Id: rmclean.for,v 1.3 2013/05/30 23:07:02 wie017 Exp $

Generated by miriad@atnf.csiro.au on 14 Sep 2015