Error in regularize.values(x, y, ties, missing(ties)) : (converted from warning) collapsing to unique 'x' values

[rgiordan]

The new GetInferenceSignals sometimes throws the following harmless error due to our use of approx with possibly very small influence scores:

  (converted from warning) collapsing to unique 'x' values

It would be better to roll our own and avoid the use of approx entirely.