On 22 Mag, 01:10, "Skybuck Flying" <BloodySh...@[EMAIL PROTECTED]
> wrote:
> Ok,
>
> So does that mean you thinking about taking the average of each "piece"
? =
;)
>
> Bye,
> =A0 Skybuck.
No, I was thinking something like the following.
Assume for simplicity the bound for my list is 10.
Infinite values arrive, one at a time, and each one is inserted at
the beginning of the list.
I was thinking to proceed as follows (use monospaced font to see
alignment, which matters, copy into notepad):
=3D=3D Original =3D=3D 0,65 0,22 0,55 0,63 0,33 0,71 0,55 0,94
0,89 0,61
=3D=3D Resampled 1=3D=3D 0,65 0,55 0,33 0,55 0,89
at second iteration the piece processed already remains untouched:
=3D=3D Original 1=3D=3D 0,93 0,36 0,06 0,66 0,78 | 0,65 0,55
0,33=
0,55 0,89
=3D=3D Resampled 2=3D=3D 0,93 0,06 0,78 0,65 0,55
0,33=
0,55 0,89
at third iteration the 2 piece processed already remain untouched:
=3D=3D Original 2=3D=3D 0,3 0,17 | 0,93 0,06 0,78 0,65 0,55 0,33
0,55 0,89
=3D=3D Resampled 3=3D=3D 0,3 0,93 0,06 0,78 0,65 0,55 0,33
0,55 0,89
at 4th iteration the 3 piece processed already remain untouched:
=3D=3D Original 3=3D=3D 0,12 | 0,3 0,93 0,06 0,78 0,65 0,55 0,33
0,55 0,89
=3D=3D Resampled 4=3D=3D 0,12 0,3 0,93 0,06 0,78 0,65 0,55 0,33
0,55 0,89
0k finished pieces, restart from beginning:
=3D=3D Original 4=3D=3D 0,37 0,12 0,3 0,93 0,06 0,78 0,65 0,55
0,3=
3
0,55 0,89
=3D=3D Resampled 5=3D=3D 0,37 0,3 0,06 0,65
0,33 0,89
=3D=3D Original 5=3D=3D 0,09 0,82 0,02 0,86 | 0,37 0,3 0,06 0,65
0,33 0,89
=3D=3D Resampled 6=3D=3D 0,09 0,02 0,37 0,3 0,06 0,65
0,33 0,89
and so forever ...
The advantage of this approach seems the incredible simplicity, but
on the other hand
does not keep into account precisely the arrival time. In my case
arrival time
are almost equispaced, so I am not much concerned about that.
Do you think it could work ? Any better idea ?
-P
|
