Maximum Element-Sum of a Complete Subset of Indices
HardUpdated: Aug 2, 2025
Practice on:
Problem
You are given a 1**-indexed** array nums. Your task is to select a
complete subset from nums where every pair of selected indices multiplied is a perfect square,. i. e. if you select ai and aj, i * j
must be a perfect square.
Return the sum of the complete subset with the maximum sum.
Examples
Example 1
Input: nums = [8,7,3,5,7,2,4,9]
Output: 16
Explanation:
We select elements at indices 2 and 8 and `2 * 8` is a perfect square.
Example 2
Input: nums = [8,10,3,8,1,13,7,9,4]
Output: 20
Explanation:
We select elements at indices 1, 4, and 9. `1 * 4`, `1 * 9`, `4 * 9` are
perfect squares.
Constraints
1 <= n == nums.length <= 10^41 <= nums[i] <= 10^9
Solution
Method 1 – Group by Square-Free Part
Intuition
For indices i and j, i * j is a perfect square if and only if their square-free parts are equal. So, group indices by their square-free part and sum the maximum group.
Approach
- For each index
i(1-based), compute its square-free part (product of primes with odd exponent in its factorization). - Group all indices with the same square-free part.
- For each group, sum the corresponding
numsvalues. - The answer is the maximum sum among all groups.
Code
C++
class Solution {
public:
int maxSum(vector<int>& nums) {
int n = nums.size();
unordered_map<int, int> group;
for (int i = 1; i <= n; ++i) {
int x = i, sqf = 1;
for (int d = 2; d * d <= x; ++d) {
int cnt = 0;
while (x % d == 0) x /= d, cnt++;
if (cnt % 2) sqf *= d;
}
if (x > 1) sqf *= x;
group[sqf] += nums[i-1];
}
int ans = 0;
for (auto& [_, s] : group) ans = max(ans, s);
return ans;
}
};
Go
func maxSum(nums []int) int {
n := len(nums)
group := map[int]int{}
for i := 1; i <= n; i++ {
x, sqf := i, 1
d := 2
for d*d <= x {
cnt := 0
for x%d == 0 {
x /= d
cnt++
}
if cnt%2 == 1 {
sqf *= d
}
d++
}
if x > 1 {
sqf *= x
}
group[sqf] += nums[i-1]
}
ans := 0
for _, s := range group {
if s > ans {
ans = s
}
}
return ans
}
Java
class Solution {
public int maximumSum(int[] nums) {
int n = nums.length;
Map<Integer, Integer> group = new HashMap<>();
for (int i = 1; i <= n; i++) {
int x = i, sqf = 1;
for (int d = 2; d * d <= x; d++) {
int cnt = 0;
while (x % d == 0) {
x /= d;
cnt++;
}
if (cnt % 2 == 1) sqf *= d;
}
if (x > 1) sqf *= x;
group.put(sqf, group.getOrDefault(sqf, 0) + nums[i-1]);
}
int ans = 0;
for (int s : group.values()) ans = Math.max(ans, s);
return ans;
}
}
Kotlin
class Solution {
fun maximumSum(nums: IntArray): Int {
val n = nums.size
val group = mutableMapOf<Int, Int>()
for (i in 1..n) {
var x = i
var sqf = 1
var d = 2
while (d * d <= x) {
var cnt = 0
while (x % d == 0) {
x /= d
cnt++
}
if (cnt % 2 == 1) sqf *= d
d++
}
if (x > 1) sqf *= x
group[sqf] = group.getOrDefault(sqf, 0) + nums[i-1]
}
return group.values.maxOrNull() ?: 0
}
}
Python
class Solution:
def maximumSum(self, nums: list[int]) -> int:
from collections import defaultdict
def square_free(x: int) -> int:
d, sqf = 2, 1
while d * d <= x:
cnt = 0
while x % d == 0:
x //= d
cnt += 1
if cnt % 2:
sqf *= d
d += 1
if x > 1:
sqf *= x
return sqf
group = defaultdict(int)
for i in range(1, len(nums)+1):
group[square_free(i)] += nums[i-1]
return max(group.values())
Rust
impl Solution {
pub fn maximum_sum(nums: Vec<i32>) -> i32 {
use std::collections::HashMap;
fn square_free(mut x: usize) -> usize {
let mut sqf = 1;
let mut d = 2;
while d * d <= x {
let mut cnt = 0;
while x % d == 0 {
x /= d;
cnt += 1;
}
if cnt % 2 == 1 {
sqf *= d;
}
d += 1;
}
if x > 1 {
sqf *= x;
}
sqf
}
let mut group = HashMap::new();
for (i, &v) in nums.iter().enumerate() {
let sqf = square_free(i+1);
*group.entry(sqf).or_insert(0) += v;
}
*group.values().max().unwrap()
}
}
TypeScript
class Solution {
maximumSum(nums: number[]): number {
const n = nums.length;
const group = new Map<number, number>();
for (let i = 1; i <= n; i++) {
let x = i, sqf = 1, d = 2;
while (d * d <= x) {
let cnt = 0;
while (x % d === 0) {
x /= d;
cnt++;
}
if (cnt % 2 === 1) sqf *= d;
d++;
}
if (x > 1) sqf *= x;
group.set(sqf, (group.get(sqf) ?? 0) + nums[i-1]);
}
return Math.max(...group.values());
}
}
Complexity
- ⏰ Time complexity:
O(n√n), for each index up ton, we factorize up to√n. - 🧺 Space complexity:
O(n), for the group map.