# Compute Bell and Uppuluri-Carpenter numbers¶

Compute Bell and Uppuluri-Carpenter numbers

AUTHORS:

• Nick Alexander
sage.combinat.expnums.expnums(n, aa)

Compute the first $$n$$ exponential numbers around $$aa$$, starting with the zero-th.

INPUT:

• n - C machine int
• aa - C machine int

OUTPUT: A list of length $$n$$.

ALGORITHM: We use the same integer addition algorithm as GAP. This is an extension of Bell’s triangle to the general case of exponential numbers. The recursion performs $$O(n^2)$$ additions, but the running time is dominated by the cost of the last integer addition, because the growth of the integer results of partial computations is exponential in $$n$$. The algorithm stores $$O(n)$$ integers, but each is exponential in $$n$$.

EXAMPLES:

sage: expnums(10, 1)
[1, 1, 2, 5, 15, 52, 203, 877, 4140, 21147]

sage: expnums(10, -1)
[1, -1, 0, 1, 1, -2, -9, -9, 50, 267]

sage: expnums(1, 1)
[1]
sage: expnums(0, 1)
[]
sage: expnums(-1, 0)
[]


AUTHORS:

• Nick Alexander
sage.combinat.expnums.expnums2(n, aa)

A vanilla python (but compiled via Cython) implementation of expnums.

We Compute the first $$n$$ exponential numbers around $$aa$$, starting with the zero-th.

EXAMPLES:

sage: from sage.combinat.expnums import expnums2
sage: expnums2(10, 1)
[1, 1, 2, 5, 15, 52, 203, 877, 4140, 21147]


#### Previous topic

The On-Line Encyclopedia of Integer Sequences (OEIS)

#### Next topic

Combinatorial Functions