1 | """ |
||
2 | Project Euler Problem 23: Non-Abundant Sums |
||
3 | =========================================== |
||
4 | |||
5 | .. module:: solutions.problem23 |
||
6 | :synopsis: My solution to problem #23. |
||
7 | |||
8 | The source code for this problem can be |
||
9 | `found here <https://bitbucket.org/nekedome/project-euler/src/master/solutions/problem23.py>`_. |
||
10 | |||
11 | Problem Statement |
||
12 | ################# |
||
13 | |||
14 | A perfect number is a number for which the sum of its proper divisors is exactly equal to the number. For example, the |
||
0 ignored issues
–
show
|
|||
15 | sum of the proper divisors of :math:`28` would be :math:`1 + 2 + 4 + 7 + 14 = 28`, which means that :math:`28` is a |
||
0 ignored issues
–
show
|
|||
16 | perfect number. |
||
17 | |||
18 | A number :math:`n` is called deficient if the sum of its proper divisors is less than :math:`n` and it is called |
||
0 ignored issues
–
show
|
|||
19 | abundant if this sum exceeds :math:`n`. |
||
20 | |||
21 | As :math:`12` is the smallest abundant number, :math:`1 + 2 + 3 + 4 + 6 = 16`, the smallest number that can be written |
||
0 ignored issues
–
show
|
|||
22 | as the sum of two abundant numbers is :math:`24`. By mathematical analysis, it can be shown that all integers greater |
||
0 ignored issues
–
show
|
|||
23 | than :math:`28123` can be written as the sum of two abundant numbers. However, this upper limit cannot be reduced any |
||
0 ignored issues
–
show
|
|||
24 | further by analysis even though it is known that the greatest number that cannot be expressed as the sum of two abundant |
||
0 ignored issues
–
show
|
|||
25 | numbers is less than this limit. |
||
26 | |||
27 | Find the sum of all the positive integers which cannot be written as the sum of two abundant numbers. |
||
0 ignored issues
–
show
|
|||
28 | |||
29 | Solution Discussion |
||
30 | ################### |
||
31 | |||
32 | The solution can be found with the following algorithm: |
||
33 | |||
34 | 1. Identify all abundant numbers |
||
35 | 2. Iterate over all integers in :math:`[2, 28123]` and determine if it is a sum of two abundant numbers, accumulate the |
||
0 ignored issues
–
show
|
|||
36 | sum of all integers that cannot be expressed as such a sum |
||
37 | |||
38 | Both steps can be optimised with the following approaches. |
||
39 | |||
40 | First, build a divisor sum sieve which can be enumerated to determine abundant numbers (i.e. :math:`n` s.t. |
||
0 ignored issues
–
show
|
|||
41 | :math:`s(n) \\gt n` where :math:`s(n)` is the sum of the proper divisors of :math:`n`). |
||
42 | |||
43 | .. note:: the use of a divisor sieve will be much quicker than identifying the divisors of each individual :math:`n`. |
||
0 ignored issues
–
show
|
|||
44 | |||
45 | Then, iterate through :math:`[2, 28123]` and check whether there are two abundant numbers that can sum to the current |
||
0 ignored issues
–
show
|
|||
46 | candidate. |
||
47 | |||
48 | * Let :math:`n` be the current candidate |
||
49 | * Let :math:`a_i` be the :math:`i^{th}` abundant number in :math:`[2, 28123]` |
||
50 | * For all :math:`a_i`, check whether there exists an :math:`a_j` s.t. :math:`a_i + a_j = n` |
||
51 | * If no :math:`a_i, a_j` exist, accumulate :math:`n` into the final answer |
||
52 | |||
53 | Importantly, by ordering :math:`\\lbrace a_i \\rbrace`, we don't need to check all :math:`a_j` at each step of this |
||
0 ignored issues
–
show
|
|||
54 | algorithm. |
||
55 | |||
56 | .. math:: |
||
57 | |||
58 | a_i + a_j &= n \\\\ |
||
59 | \\Rightarrow n - a_i &= a_j |
||
60 | |||
61 | Therefore, we only need to check :math:`a_i \\lt n` (otherwise :math:`n - a_i` would be negative) and since :math:`a_i` |
||
0 ignored issues
–
show
|
|||
62 | is positive, we only need to check :math:`0 \\lt a_j \\lt n`. This means, that we may consider the subset of |
||
0 ignored issues
–
show
|
|||
63 | :math:`\\lbrace a_k \\rbrace` s.t. :math:`a_k \\le n` for each candidate :math:`n`. |
||
64 | |||
65 | Solution Implementation |
||
66 | ####################### |
||
67 | |||
68 | .. literalinclude:: ../../solutions/problem23.py |
||
69 | :language: python |
||
70 | :lines: 73- |
||
71 | """ |
||
72 | |||
73 | from lib.numbertheory import divisors_sieve |
||
74 | |||
75 | |||
76 | def solve(): |
||
77 | """ Compute the answer to Project Euler's problem #23 """ |
||
78 | |||
79 | upper_bound = 28123 |
||
80 | |||
81 | # Step1: Sieve the sum of proper divisors in [2, upper_bound] |
||
82 | sum_divisors = list(divisors_sieve(upper_bound + 1, proper=True, aggregate="sum")) |
||
83 | |||
84 | # Step2: Identify integers not expressible as the sum of two abundant numbers, accumulate their sum |
||
0 ignored issues
–
show
|
|||
85 | abundant_subspace = set() # will contain only those a_i that need be considered for each n |
||
86 | answer = 0 |
||
87 | for n in range(1, upper_bound + 1): |
||
0 ignored issues
–
show
The name
n does not conform to the variable naming conventions ((([a-z][a-z0-9_]{2,30})|(_[a-z0-9_]*))$ ).
This check looks for invalid names for a range of different identifiers. You can set regular expressions to which the identifiers must conform if the defaults do not match your requirements. If your project includes a Pylint configuration file, the settings contained in that file take precedence. To find out more about Pylint, please refer to their site. ![]() |
|||
88 | if sum_divisors[n - 1] > n: |
||
89 | abundant_subspace.add(n) |
||
90 | if not any((n - a in abundant_subspace) for a in abundant_subspace): |
||
91 | # The any operator provides lazy evaluation, terminating on the first satisfied condition |
||
0 ignored issues
–
show
|
|||
92 | answer += n # accumulate sum |
||
93 | |||
94 | return answer |
||
95 | |||
96 | |||
97 | expected_answer = 4179871 |
||
0 ignored issues
–
show
The name
expected_answer does not conform to the constant naming conventions ((([A-Z_][A-Z0-9_]*)|(__.*__))$ ).
This check looks for invalid names for a range of different identifiers. You can set regular expressions to which the identifiers must conform if the defaults do not match your requirements. If your project includes a Pylint configuration file, the settings contained in that file take precedence. To find out more about Pylint, please refer to their site. ![]() |
|||
98 |
This check looks for lines that are too long. You can specify the maximum line length.