solve_gaintable

ska_sdp_func_python.calibration.solvers.solve_gaintable(vis: Visibility, modelvis: Visibility = None, gain_table=None, phase_only=True, niter=30, tol=1e-06, crosspol=False, normalise_gains='mean', solver='gain_substitution', jones_type='T', timeslice=None, refant=0) → GainTable

Solve a gain table by fitting an observed visibility to a model visibility.

If modelvis is None, a point source model is assumed.

Parameters:

  • vis – Visibility containing the observed data_models

  • modelvis – Visibility containing the visibility predicted by a model

  • gain_table – Existing gaintable (default=None)

  • phase_only – Solve only for the phases. default=True when solver=”gain_substitution”, otherwise it must be False.

  • niter – Maximum number of iterations (default=30)

  • tol – Iteration stops when the fractional change in the gain solution is below this tolerance (default=1e-6)

  • crosspol – Do solutions including cross polarisations i.e. XY, YX or RL, LR. Only used by the gain_substitution solver.

  • normalise_gains – Normalises the gains (default=”mean”) options are None, “mean”, “median”. None means no normalization. Only available with gain_substitution.

  • solver – Calibration algorithm to use (default=”gain_substitution”) options are: “gain_substitution” - original substitution algorithm with separate solutions for each polarisation term. “jones_substitution” - solve antenna-based Jones matrices as a whole, with independent updates within each iteration. “normal_equations” - solve normal equations within each iteration formed from linearisation with respect to antenna-based gain and leakage terms. “normal_equations_presum” - the same as the normal_equations option but with an initial accumulation of visibility products over time and frequency for each solution interval. This can be much faster for large datasets and solution intervals.

  • jones_type – Type of calibration matrix T or G or B

  • timeslice – Time interval between solutions (s)

  • refant – Reference antenna (default 0). Currently only activated for the gain_substitution solver.

Returns:

GainTable containing solution