|
|
A309363
|
|
Van Eck's sequence (cf. A181391), but outputting 2 for a new number, not 0.
|
|
2
|
|
|
0, 2, 2, 1, 2, 2, 1, 3, 2, 3, 2, 2, 1, 6, 2, 3, 6, 3, 2, 4, 2, 2, 1, 10, 2, 3, 8, 2, 3, 3, 1, 8, 5, 2, 6, 18, 2, 3, 8, 7, 2, 4, 22, 2, 3, 7, 6, 12, 2, 5, 17, 2, 3, 8, 15, 2, 4, 15, 3, 6, 13, 2, 6, 3, 5, 15, 8, 13, 7, 23, 2, 9, 2, 2, 1, 44
(list;
graph;
refs;
listen;
history;
text;
internal format)
|
|
|
OFFSET
|
1,2
|
|
COMMENTS
|
After the initial value, the sequence is extended by a(n+1) = min { k > 0: a(n-k) = a(n) } or 2 if no such k exists, i.e., if a(n) did not appear earlier.
Although the sequence has properties that are superficially similar to the original A181391, there is an important difference. Using a positive number m instead of 0 to mark a new value means there is no 1-to-1 correspondence between the occurrence of a new value and the occurrence of m. - Jan Ritsema van Eck, Aug 14 2019
|
|
LINKS
|
|
|
PROG
|
(Python)
from itertools import count, islice
def A309363gen(): # generator of terms
b, bdict = 0, {0:(1, )}
for n in count(2):
yield b
if len(l := bdict[b]) > 1:
b = n-1-l[-2]
else:
b = 2
if b in bdict:
bdict[b] = (bdict[b][-1], n)
else:
bdict[b] = (n, )
|
|
CROSSREFS
|
Cf. A181391, A171911, ..., A171918 (same but starting with 0, 1, 2, ..., 8 and returning 0 when a new term appears).
|
|
KEYWORD
|
nonn
|
|
AUTHOR
|
|
|
STATUS
|
approved
|
|
|
|