[seqfan] Re: Run Superseeker on existing sequence?

Marc LeBrun mlb at well.com
Fri Aug 13 22:30:12 CEST 2010


>="N. J. A. Sloane" <njas at research.att.com>

>> Can superseeker be used to look for formulas for an existing sequence, rather
>> than just reporting that the sequence already exists?
> 
> Yes, it does that automatically. If the sequence is
> in the OEIS it tells you; independently, if it
> finds a formula it tells you that too
>
> (I think! Haven't looked at the source code for years)
> 

========

Easier to just try it!  Using the "odd numbers" example at
http://www.research.att.com/~njas/sequences/ol.html
I do get the [much snipped] output further below.

Some comments:

  *  Perhaps the above page should emphasize "...it will ALSO apply a large
number of algorithms..."?

  *  Does it try ALL the algorithms?  I would have expected that many
transforms of the odd numbers would be in the OEIS. (This exemplifies why
systematically augmenting the OEIS with at least the main transforms of the
169 core sequences seems a worthwhile priority).

  *  Also, a longstanding feature request: have Superseeker parrot back
whatever is in the Subject line of the query as the Subject of its reply, to
make them easier to manage in eMail folders.

  *  A while back some seqfans looked into specifying an XML schema for OEIS
entries, as an alternative to the classic "human readable" format.  It would
be relatively trivial to convert between them, and would enable then folks
to implement custom renderings, web services (eg other seekers) etc.  Maybe
this would be more timely to pursue nowadays?

--------

Report on [ 1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31]:
Many tests are carried out, but only potentially useful information
(if any) is reported here.


TEST: IS THE SEQUENCE OF ABSOLUTE VALUES IN THE ENCYCLOPEDIA?
Matches (up to a limit of 50) found for
"1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31"

%I A005408 M2400
...

%I A006257 M2216
...
%I A176271
...
%I A060747
...
%I A004273
...
%I A113648
...
%I A053229
...
%I A089684
...
%I A033041
...
%I A105356
...
%I A109846
...
%I A172049
...
%I A157142
...
%I A165747
...

        SUCCESS: the sequence is in the OEIS.


TEST: IS THE NTH TERM A POLYNOMIAL IN N?
        SUCCESS: nth term is nontrivial polynomial in n of
degree 1
Polynomial is:
1+2*n
memory used=1.4MB, alloc=1.7MB, time=0.01

Sequence is a polynomial of degree at most 2.  Bye!

...






More information about the SeqFan mailing list