baseline_name
Determine and apply a baseline based amplitude and phase correction.
EXAMPLE
-------

To correct for correlator offsets on baseline BONN-WSRT of the first
subarray.

  0> resoff "BON-WSRT"
  1:BONN-WSRT: 0.945151 amplitude change, -0.399408 degrees phase offset
  0>

To correct for correlator offsets on baseline HSTK-OVRO of the 3rd
sub-array.

  0> resoff 3:HSTK-OVRO
  3:HSTK-OVRO: 0.988347 amplitude change, -3.50419 degrees phase offset
  0>

To correct all baselines of sub-array 2.

  0> resoff 2
  2:WSRT-JBNK: 0 amplitude change, -5.63816 degrees phase offset
  2:WSRT-BOLOGNA: 0 amplitude change, 13.0693 degrees phase offset
  2:WSRT-HSTK: 0 amplitude change, -0.687232 degrees phase offset
  2:JBNK-BOLOGNA: 0 amplitude change, 12.97 degrees phase offset
  2:JBNK-HSTK: 0 amplitude change, 2.05668 degrees phase offset
  2:BOLOGNA-HSTK: 0 amplitude change, 2.33325 degrees phase offset

To correct all baselines.

  0> resoff
  1:BONN-WSRT: 0.945151 amplitude change, -0.399408 degrees phase offset
  .....etc....

PARAMETER
---------
baseline_name  -  Default=""
                  The name of the baseline(s) to be corrected. See the
                  antenna_names help topic for a discussion of the
                  syntax and interpretation of baseline specification
                  arguments.

CONTEXT
-------
The use of this command is NOT recommended except in exceptional
circumstances. The only recommended use of this command is for
inspection to determine whether a data-set contains closure offsets
with respect to a given model - in which case it should be followed by
clroff to remove the closure offsets instated. In very high dynamic
range mapping it can be used to fix baseline dependent amplitude and
phase "non-closing" errors. On other data-sets it is just as likely to
introduce new non-closing errors and should not be used.

resoff should only be used after all possible cleaning and
self-calibration has been performed and then only when the data appear
to fit very well on all but one or two baselines. Visual inspection of
the data is essential before embarking on this coarse, so use the
vplot and radplot commands first. If the miscreant baselines are short
baselines, then the real cause of the errors is undoubtedly an
incomplete model, possibly due to insufficient sampling at short UV
spacings. In such cases use of resoff is not justified. 

resoff determines and applies a single amplitude and phase correction
for the requested baseline. The amplitude correction is determined as
the ratio of the weighted means of the model and observed amplitudes
on the requested baseline, over the whole observation. The phase
correction is the weighted mean phase difference between the model and
observed phases over the same range.

Be warned, unlike self-calibration, resoff is not constrained by any
closure or other constraints. It will always make a map look more like
the model but at the expense of reality if the model isn't correct.

SIDE EFFECTS
------------
Resoff establishes the latest clean model when necessary. It also
invalidates the current residual map, which will be re-calculated by
the next command that needs it.

RELATED COMMANDS
----------------
clroff  -  Undo the effects of all resoff commands in this difmap session.