3883. Count Non Decreasing Arrays With Given Digit Sums
Description
You are given an integer array digitSum of length n.
An array arr of length n is considered valid if:
0 <= arr[i] <= 5000- it is non-decreasing.
- the sum of the digits of
arr[i]equalsdigitSum[i].
Return an integer denoting the number of distinct valid arrays. Since the answer may be large, return it modulo 109 + 7.
An array is said to be non-decreasing if each element is greater than or equal to the previous element, if it exists.
Example 1:
Input: digitSum = [25,1]
Output: 6
Explanation:
Numbers whose sum of digits is 25 are 799, 889, 898, 979, 988, and 997.
The only number whose sum of digits is 1 that can appear after these values while keeping the array non-decreasing is 1000.
Thus, the valid arrays are [799, 1000], [889, 1000], [898, 1000], [979, 1000], [988, 1000], and [997, 1000].
Hence, the answer is 6.
Example 2:
Input: digitSum = [1]
Output: 4
Explanation:
The valid arrays are [1], [10], [100], and [1000].
Thus, the answer is 4.
Example 3:
Input: digitSum = [2,49,23]
Output: 0
Explanation:
There is no integer in the range [0, 5000] whose sum of digits is 49. Thus, the answer is 0.
Constraints:
1 <= digitSum.length <= 10000 <= digitSum[i] <= 50
Solutions
Solution 1
Thinking
Count nondecreasing arrays with \(0 \le arr[i] \le 5000\) whose digit sums equal \(\textit{digitSum}[i]\). \(n \le 1000\) and digit sums \(\le 50\).
Monotonicity reduces the task to choosing a value at each index no smaller than the previous. Each digit sum has finitely many candidates.
Precompute legal numbers per sum, then DP on index and last value, moving to a candidate at least as large.
Modulo \(10^9+7\). Values go up to \(5000\), so sort candidates and use prefix sums to speed transitions.
1 | |
1 | |
1 | |
1 | |