更改数组以使最大元素为数组的LCM的方法数量

给定数组arr [] ，任务是对唯一数组的数量进行计数，方法是将给定数组的元素更新为[1，arr [i]]范围内的任何元素，以使更新后的数组等于最大元素。

例子：

Input: arr[] = {6, 3}
Output: 13
Explanation:
Possible Arrays are –
{[1, 1], [1, 2], [2, 1], [2, 2], [1, 3],
[3, 1], [3, 3], [4, 1], [4, 2], [5, 1],
[6, 1], [6, 2], [6, 3]}

Input: arr[] = {1, 4, 3, 2}
Output: 15

方法：

• 为了使最大元素成为数组的LCM，我们需要修复数组的最大元素。
• 因为，我们已经确定了一些数字$i$最大，现在对于LCM $i$ ，我们需要确保数组中的每个元素都是的倍数$i$包括$i$
• 我们会找到影响数字的因素$i$并找到将它们放置在数组中的方法数量。
• 比方说， $i$是$x, y, z$ 。因素的数量是$3$ 。
• 让我们假设$x$是$a$意味着有$a$数组中具有大于等于的位置数$x$然后让$y$有$b$职位和$z$有$c$职位。
• 现在，有几种方式分配x [/ Tex] $a$职位， $y$在$b - a$职位和$z$在$c - b - a$职位是$3 ^ a * 2 ^ {(b - a)} * 1 ^ {(c - b - a)}$等等。
• 现在，我们必须减去具有LCM的那些方法$i$但$i$不在这里。
• 我们需要精简$2 ^ b * 1 ^ {(c - b)}$从方式上
• 我们将使用BIT(二叉索引树)来查找数量大于某个数量的头寸$x$ 。

下面是上述方法的实现：

C++

// C++ implementation to find the
// Number of ways to change the array
// such that maximum element of the
// array is the LCM of the array
 
#include 
using namespace std;
 
// Modulo
const int MOD = 1e9 + 7;
const int N = 1e5 + 5;
 
// Fenwick tree to find number
// of indexes greater than x
vector BIT(N, 0);
 
// Function to compute
// x ^ y % MOD
int power(int x, int y)
{
    if (x == 0)
        return 0;
 
    int ans = 1;
 
    // Loop to compute the
    // x^y % MOD
    while (y > 0) {
        if (y & 1)
            ans = (1LL * ans * x) % MOD;
 
        x = (1LL * x * x) % MOD;
        y >>= 1;
    }
    return ans;
}
 
// Function to update the binary
// indexed tree
void updateBIT(int idx, int val)
{
    assert(idx > 0);
    while (idx < N) {
        BIT[idx] += val;
        idx += idx & -idx;
    }
}
 
// Function to find the prefix sum
// upto the current index
int queryBIT(int idx)
{
    int ans = 0;
    while (idx > 0) {
        ans += BIT[idx];
        idx -= idx & -idx;
    }
    return ans;
}
 
// Function to find the number of
// ways to change the array such
// that the LCM of array is
// maximum element of the array
int numWays(int arr[], int n)
{
 
    int mx = 0;
    for (int i = 0; i < n; i++) {
 
        // Updating BIT with the
        // frequency of elements
        updateBIT(arr[i], 1);
 
        // Maximum element in the array
        mx = max(mx, arr[i]);
    }
 
    // 1 is for every element
    // is 1 in the array;
    int ans = 1;
    for (int i = 2; i <= mx; i++) {
 
        // Vector for storing the factors
        vector factors;
        for (int j = 1; j * j <= i; j++) {
 
            // finding factors of i
            if (i % j == 0) {
                factors.push_back(j);
                if (i / j != j)
                    factors.push_back(i / j);
            }
        }
        // Sorting in descending order
        sort(factors.rbegin(), factors.rend());
 
        int cnt = 1;
 
        // for storing number of indexex
        // greater than the i - 1 element
        int prev = 0;
        for (int j = 0; j < factors.size(); j++) {
 
            // Number of remaining factors
            int remFactors = int(factors.size()) - j;
 
            // Number of indexes in the array
            // with element factor[j] and above
            int indexes = n - queryBIT(factors[j] - 1);
 
            // Multiplying count with
            // remFcators ^ (indexes - prev)
            cnt = (1LL
                   * cnt
                   * power(remFactors,
                           indexes - prev))
                  % MOD;
            prev = max(prev, indexes);
        }
 
        // Remove those counts which have
        // lcm as i but i is not present
        factors.erase(factors.begin());
 
        int toSubtract = 1;
        prev = 0;
 
        // Loop to find the count which have
        // lcm as i  but i is not present
        for (int j = 0; j < factors.size(); j++) {
            int remFactors = int(factors.size()) - j;
            int indexes = n - queryBIT(factors[j] - 1);
 
            toSubtract = (1LL
                          * toSubtract
                          * power(remFactors,
                                  indexes - prev));
            prev = max(prev, indexes);
        }
 
        // Adding cnt - toSubtract to answer
        ans = (1LL * ans + cnt
               - toSubtract + MOD)
              % MOD;
    }
    return ans;
}
 
// Driver Code
int main()
{
    int arr[] = { 6, 3 };
    int n = sizeof(arr) / sizeof(arr[0]);
 
    int ans = numWays(arr, n);
    cout << ans << endl;
    return 0;
}

Python3

# Python implementation to find the
# Number of ways to change the array
# such that maximum element of the
# array is the LCM of the array
 
# Modulo
MOD = int(1e9) + 9
MAXN = int(1e5) + 5
 
# Fenwick tree to find number
# of indexes greater than x
BIT = [0 for _ in range(MAXN)]
 
# Function to compute
# x ^ y % MOD
def power(x, y):
    if x == 0:
        return 0
    ans = 1
     
    # Loop to compute the
    # x ^ y % MOD
    while y > 0:
        if y % 2 == 1:
            ans = (ans * x) % MOD
        x = (x * x) % MOD
        y = y // 2
    return ans
 
# Function to update the
# Binary Indexed Tree
def updateBIT(idx, val):
     
    # Loop to update the BIT
    while idx < MAXN:
        BIT[idx] += val
        idx += idx & (-idx)
 
# Function to find
# prefix sum upto idx
def queryBIT(idx):
    ans = 0
    while idx > 0:
        ans += BIT[idx]
        idx -= idx & (-idx)
    return ans
 
# Function to find number of ways
# to change the array such that
# MAX of array is same as LCM
def numWays(arr):
    mx = 0
     
    # Updating BIT with the
    # frequency of elements
    for i in arr:
        updateBIT(i, 1)
         
        # Maximum element
        # in the array
        mx = max(mx, i)
 
    ans = 1
    for i in range(2, mx + 1):
         
        # For storing factors of i
        factors = []
        for j in range(1, i + 1):
            if j * j > i:
                break
                 
            # Finding factors of i
            if i % j == 0:
                factors.append(j)
                if i // j != j:
                    factors.append(i // j)
 
        # Sorting in descending order
        factors.sort()
        factors.reverse()
         
        # For storing ans
        cnt = 1
         
        # For storing number of indexes
        # greater than the i - 1 element
        prev = 0
        for j in range(len(factors)):
             
            # Number of remaining factors
            remFactors = len(factors) - j
             
            # Number of indexes in the array
            # with element factor[j] and above
            indexes = len(arr) - queryBIT(factors[j] - 1)
             
            # Multiplying count with
            # remFcators ^ (indexes - prev)
            cnt = (cnt * power(remFactors, \
                     indexes - prev)) % MOD
            prev = max(prev, indexes)
 
        # Remove those counts which have
        # lcm as i but i is not present
        factors.remove(factors[0])
 
        toSubtract = 1
        prev = 0
        for j in range(len(factors)):
            remFactors = len(factors) - j
            indexes = len(arr) - queryBIT(factors[j] - 1)
 
            toSubtract = (toSubtract *\
              power(remFactors, indexes - prev))
            prev = max(prev, indexes)
 
        # Adding cnt - toSubtract to ans;
        ans = (ans + cnt - toSubtract + MOD) % MOD;
         
    return ans
 
# Driver Code
if __name__ == "__main__":
    arr = [1, 4, 3, 2]
     
    ans = numWays(arr);
    print(ans)

输出：

13

时间复杂度： $O(MAX * $\sqrt{MAX}$)$ ， 在哪里$$MAX$是数组中的最大元素。