TEPS TEPS Type-based Epsilon Calculation

Usage

Returns eps for double precision arguments and
feps for single precision arguments.  The syntax for
teps is

   y = teps(x)

The teps function is most useful if you need to
compute epsilon based on the type of the array.