time-to-botec

Benchmark sampling in different programming languages
Log | Files | Refs | README

repl.txt (1421B)


      1 
      2 {{alias}}( W )
      3     Returns an accumulator function which incrementally computes a moving
      4     mean arctangent absolute percentage error (MAAPE).
      5 
      6     The `W` parameter defines the number of values over which to compute the
      7     moving mean arctangent absolute percentage error.
      8 
      9     If provided input values, the accumulator function returns an updated moving
     10     mean arctangent absolute percentage error. If not provided input values, the
     11     accumulator function returns the current moving mean arctangent absolute
     12     percentage error.
     13 
     14     Note that, unlike the mean absolute percentage error (MAPE), the mean
     15     arctangent absolute percentage error is expressed in radians on the interval
     16     [0,π/2].
     17 
     18     As `W` (f,a) pairs are needed to fill the window buffer, the first `W-1`
     19     returned values are calculated from smaller sample sizes. Until the window
     20     is full, each returned value is calculated from all provided values.
     21 
     22     Parameters
     23     ----------
     24     W: integer
     25         Window size.
     26 
     27     Returns
     28     -------
     29     acc: Function
     30         Accumulator function.
     31 
     32     Examples
     33     --------
     34     > var accumulator = {{alias}}( 3 );
     35     > var m = accumulator()
     36     null
     37     > m = accumulator( 2.0, 3.0 )
     38     ~0.32
     39     > m = accumulator( 5.0, 2.0 )
     40     ~0.65
     41     > m = accumulator( 3.0, 2.0 )
     42     ~0.59
     43     > m = accumulator( 2.0, 5.0 )
     44     ~0.66
     45     > m = accumulator()
     46     ~0.66
     47 
     48     See Also
     49     --------
     50