Maximum of all subarrays of size K using Segment Tree

Last Updated : 28 Mar, 2023

Given an array arr[] and an integer K, the task is to find the maximum for each and every contiguous subarray of size K.

Examples:

Input: arr[] = {1, 2, 3, 1, 4, 5, 2, 3, 6}, K = 3
Output: 3 3 4 5 5 5 6
Explanation:
Maximum of 1, 2, 3 is 3
Maximum of 2, 3, 1 is 3
Maximum of 3, 1, 4 is 4
Maximum of 1, 4, 5 is 5
Maximum of 4, 5, 2 is 5
Maximum of 5, 2, 3 is 5
Maximum of 2, 3, 6 is 6

Input: arr[] = {8, 5, 10, 7, 9, 4, 15, 12, 90, 13}, K = 4
Output: 10 10 10 15 15 90 90
Explanation:
Maximum of first 4 elements is 10, similarly for next 4
elements (i.e from index 1 to 4) is 10, So the sequence
generated is 10 10 10 15 15 90 90

Approach:
The idea is to use the Segment tree to answer the maximum of all subarrays of size K.

Representation of Segment trees
- Leaf Nodes are the elements of the input array.
- Each internal node represents the maximum of all of its children.
Construction of Segment Tree from the given array:
- We start with a segment arr[0 . . . n-1], and every time we divide the current segment into two halves(if it has not yet become a segment of length 1), and then call the same procedure on both halves, and for each such segment, we store the maximum value in a segment tree node.
- All levels of the constructed segment tree will be completely filled except the last level. Also, the tree will be a full Binary Tree because we always divide segments into two halves at every level.
- Since the constructed tree is always a full binary tree with n leaves, there will be n - 1 internal nodes. So total nodes will be 2 * n – 1.
- The height of the segment tree will be log₂n.
- Since the tree is represented using an array and the relation between parent and child indexes must be maintained, the size of memory allocated for the segment tree will be 2 *(2^ceil(log_²ⁿ⁾)-1.

Below is the implementation of the above approach.

C++

// C++  program to answer Maximum
// of allsubarrays of size k
// using segment tree
#include <bits/stdc++.h>
using namespace std;

#define MAX 1000000

// Size of segment
// tree = 2^{log(MAX)+1}
int st[3 * MAX];

// A utility function to get the
// middle index of given range.
int getMid(int s, int e)
{
    return s + (e - s) / 2;
}
// A recursive function that
// constructs Segment Tree for
// array[s...e]. node is index
// of current node in segment
// tree st
void constructST(int node, int s,
                 int e, int arr[])
{
    // If there is one element in
    // array, store it in current
    // node of segment tree and return
    if (s == e) {
        st[node] = arr[s];
        return;
    }
    // If there are more than
    // one elements, then recur
    // for left and right subtrees
    // and store the max of
    // values in this node
    int mid = getMid(s, e);

    constructST(2 * node, s,
                mid, arr);
    constructST(2 * node + 1,
                mid + 1, e,
                arr);
    st[node] = max(st[2 * node],
                   st[2 * node + 1]);
}

/* A recursive function to get the 
   maximum of range[l, r] The
   following parameters for
   this function:

st     -> Pointer to segment tree
node   -> Index of current node in
          the segment tree .
s & e  -> Starting and ending indexes
          of the segment represented
          by current node, i.e., st[node]
l & r  -> Starting and ending indexes
          of range query
 */
int getMax(int node, int s,
           int e, int l,
           int r)
{
    // If segment of this node
    // does not belong to
    // given range
    if (s > r || e < l)
        return INT_MIN;

    // If segment of this node
    // is completely part of
    // given range, then return
    // the max of segment
    if (s >= l && e <= r)
        return st[node];

    // If segment of this node
    // is partially the part
    // of given range
    int mid = getMid(s, e);

    return max(getMax(2 * node,
                      s, mid,
                      l, r),
               getMax(2 * node + 1,
                      mid + 1, e,
                      l, r));
}

// Function to print the max
// of all subarrays of size k
void printKMax(int n, int k)
{
    for (int i = 0; i < n; i++) {
        if ((k - 1 + i) < n)
            cout << getMax(1, 0, n - 1,
                           i, k - 1 + i)
                 << " ";
        else
            break;
    }
}

// Driver code
int main()
{
    int k = 4;
    int arr[] = { 8, 5, 10, 7, 9, 4, 15,
                  12, 90, 13 };
    int n = sizeof(arr) / sizeof(arr[0]);

    // Function to construct the
    // segment tree
    constructST(1, 0, n - 1, arr);

    printKMax(n, k);

    return 0;
}

Java

// Java program to answer Maximum
// of allsubarrays of size k
// using segment tree
import java.io.*;
import java.util.*;

class GFG{

static int MAX = 1000000;

// Size of segment
// tree = 2^{log(MAX)+1}
static int[] st = new int[3 * MAX];

// A utility function to get the
// middle index of given range.
static int getMid(int s, int e)
{ 
    return s + (e - s) / 2;
}

// A recursive function that
// constructs Segment Tree for
// array[s...e]. node is index
// of current node in segment
// tree st
static void constructST(int node, int s,
                        int e, int[] arr)
{
    
    // If there is one element in
    // array, store it in current
    // node of segment tree and return
    if (s == e)
    {
        st[node] = arr[s];
        return;
    }
    
    // If there are more than
    // one elements, then recur
    // for left and right subtrees
    // and store the max of
    // values in this node
    int mid = getMid(s, e);

    constructST(2 * node, s, mid, arr);
    constructST(2 * node + 1, 
                mid + 1, e, arr);
    
    st[node] = Math.max(st[2 * node],
                        st[2 * node + 1]);
}

/* A recursive function to get the
   maximum of range[l, r] The
   following parameters for
   this function:

st     -> Pointer to segment tree
node   -> Index of current node in
          the segment tree .
s & e  -> Starting and ending indexes
          of the segment represented
          by current node, i.e., st[node]
l & r  -> Starting and ending indexes
          of range query
 */
static int getMax(int node, int s, int e,
                            int l, int r)
{
    
    // If segment of this node
    // does not belong to
    // given range
    if (s > r || e < l)
        return Integer.MIN_VALUE;

    // If segment of this node
    // is completely part of
    // given range, then return
    // the max of segment
    if (s >= l && e <= r)
        return st[node];

    // If segment of this node
    // is partially the part
    // of given range
    int mid = getMid(s, e);

    return Math.max(getMax(2 * node, s, 
                           mid, l, r),
                    getMax(2 * node + 1,
                           mid + 1, e, l, r));
}

// Function to print the max
// of all subarrays of size k
static void printKMax(int n, int k)
{
    for(int i = 0; i < n; i++)
    {
        if ((k - 1 + i) < n)
            System.out.print(getMax(1, 0, n - 1, 
                                    i, k - 1 + i) + " ");
        else
            break;
    }
}

// Driver code
public static void main(String[] args)
{
    int k = 4;
    int[] arr = { 8, 5, 10, 7, 9, 
                  4, 15, 12, 90, 13 };
    int n = arr.length;

    // Function to construct the
    // segment tree
    constructST(1, 0, n - 1, arr);

    printKMax(n, k);
}
}

// This code is contributed by akhilsaini

Python3

# Python3 program to answer maximum
# of all subarrays of size k
# using segment tree
import sys 

MAX = 1000000

# Size of segment
# tree = 2^{log(MAX)+1}
st = [0] * (3 * MAX)

# A utility function to get the
# middle index of given range.
def getMid(s, e):
    return s + (e - s) // 2
    
# A recursive function that
# constructs Segment Tree for
# array[s...e]. node is index
# of current node in segment
# tree st
def constructST(node, s, e, arr):

    # If there is one element in
    # array, store it in current
    # node of segment tree and return
    if (s == e):
        st[node] = arr[s]
        return

    # If there are more than
    # one elements, then recur
    # for left and right subtrees
    # and store the max of
    # values in this node
    mid = getMid(s, e)
    constructST(2 * node, s, mid, arr)
    constructST(2 * node + 1, mid + 1, e, arr)
    st[node] = max(st[2 * node], st[2 * node + 1])

''' A recursive function to get the 
maximum of range[l, r] The
following parameters for
this function:

st     -> Pointer to segment tree
node -> Index of current node in
        the segment tree .
s & e -> Starting and ending indexes
        of the segment represented
        by current node, i.e., st[node]
l & r -> Starting and ending indexes
        of range query
'''
def getMax(node, s, e, l, r):

    # If segment of this node
    # does not belong to
    # given range
    if (s > r or e < l):
        return (-sys.maxsize - 1)

    # If segment of this node
    # is completely part of
    # given range, then return
    # the max of segment
    if (s >= l and e <= r):
        return st[node]

    # If segment of this node
    # is partially the part
    # of given range
    mid = getMid(s, e)

    return max(getMax(2 * node, s, mid, l, r), 
               getMax(2 * node + 1, mid + 1, e, l, r))

# Function to print the max
# of all subarrays of size k
def printKMax(n, k):

    for i in range(n):
        if ((k - 1 + i) < n):
            print(getMax(1, 0, n - 1, i,
                               k - 1 + i), end = " ")
        else:
            break

# Driver code
if __name__ == "__main__":
    
    k = 4
    arr = [ 8, 5, 10, 7, 9, 4, 15, 12, 90, 13 ]
    n = len(arr)

    # Function to construct the
    # segment tree
    constructST(1, 0, n - 1, arr)
    
    printKMax(n, k)

# This code is contributed by chitranayal

// C# program to answer Maximum
// of allsubarrays of size k
// using segment tree
using System;

class GFG{

static int MAX = 1000000;

// Size of segment
// tree = 2^{log(MAX)+1}
static int[] st = new int[3 * MAX];

// A utility function to get the
// middle index of given range.
static int getMid(int s, int e) 
{
    return s + (e - s) / 2; 
}

// A recursive function that
// constructs Segment Tree for
// array[s...e]. node is index
// of current node in segment
// tree st
static void constructST(int node, int s, 
                        int e, int[] arr)
{
    
    // If there is one element in
    // array, store it in current
    // node of segment tree and return
    if (s == e)
    {
        st[node] = arr[s];
        return;
    }
    
    // If there are more than
    // one elements, then recur
    // for left and right subtrees
    // and store the max of
    // values in this node
    int mid = getMid(s, e);

    constructST(2 * node, s, mid, arr);
    constructST(2 * node + 1, mid + 1, e, arr);
    
    st[node] = Math.Max(st[2 * node], 
                        st[2 * node + 1]);
}

/* A recursive function to get the
   maximum of range[l, r] The
   following parameters for
   this function:

st     -> Pointer to segment tree
node   -> Index of current node in
          the segment tree .
s & e  -> Starting and ending indexes
          of the segment represented
          by current node, i.e., st[node]
l & r  -> Starting and ending indexes
          of range query
 */
static int getMax(int node, int s, int e,
                            int l, int r)
{
    
    // If segment of this node
    // does not belong to
    // given range
    if (s > r || e < l)
        return int.MinValue;

    // If segment of this node
    // is completely part of
    // given range, then return
    // the max of segment
    if (s >= l && e <= r)
        return st[node];

    // If segment of this node
    // is partially the part
    // of given range
    int mid = getMid(s, e);

    return Math.Max(getMax(2 * node, s, 
                           mid, l, r),
                    getMax(2 * node + 1, 
                           mid + 1, e, l, r));
}

// Function to print the max
// of all subarrays of size k
static void printKMax(int n, int k)
{
    for(int i = 0; i < n; i++)
    {
        if ((k - 1 + i) < n)
            Console.Write(getMax(1, 0, n - 1, 
                                 i, k - 1 + i) + " ");
        else
            break;
    }
}

// Driver code
public static void Main()
{
    int k = 4;
    int[] arr = { 8, 5, 10, 7, 9,
                  4, 15, 12, 90, 13 };
    int n = arr.Length;

    // Function to construct the
    // segment tree
    constructST(1, 0, n - 1, arr);

    printKMax(n, k);
}
}

// This code is contributed by akhilsaini

JavaScript

<script>

// Javascript program to answer Maximum
// of allsubarrays of size k
// using segment tree

var MAX = 1000000;

// Size of segment
// tree = 2^{log(MAX)+1}
var st = Array(3*MAX);

// A utility function to get the
// middle index of given range.
function getMid(s, e)
{
    return s + parseInt((e - s) / 2);
}
// A recursive function that
// constructs Segment Tree for
// array[s...e]. node is index
// of current node in segment
// tree st
function constructST(node, s, e, arr)
{
    // If there is one element in
    // array, store it in current
    // node of segment tree and return
    if (s == e) {
        st[node] = arr[s];
        return;
    }
    // If there are more than
    // one elements, then recur
    // for left and right subtrees
    // and store the max of
    // values in this node
    var mid = getMid(s, e);

    constructST(2 * node, s,
                mid, arr);
    constructST(2 * node + 1,
                mid + 1, e,
                arr);
    st[node] = Math.max(st[2 * node],
                   st[2 * node + 1]);
}

/* A recursive function to get the 
   maximum of range[l, r] The
   following parameters for
   this function:

st     -> Pointer to segment tree
node   -> Index of current node in
          the segment tree .
s & e  -> Starting and ending indexes
          of the segment represented
          by current node, i.e., st[node]
l & r  -> Starting and ending indexes
          of range query
 */
function getMax(node, s, e, l, r)
{
    // If segment of this node
    // does not belong to
    // given range
    if (s > r || e < l)
        return -1000000000;

    // If segment of this node
    // is completely part of
    // given range, then return
    // the max of segment
    if (s >= l && e <= r)
        return st[node];

    // If segment of this node
    // is partially the part
    // of given range
    var mid = getMid(s, e);

    return Math.max(getMax(2 * node,
                      s, mid,
                      l, r),
               getMax(2 * node + 1,
                      mid + 1, e,
                      l, r));
}

// Function to print the max
// of all subarrays of size k
function printKMax(n, k)
{
    for (var i = 0; i < n; i++) {
        if ((k - 1 + i) < n)
            document.write( getMax(1, 0, n - 1,
                           i, k - 1 + i)
                 + " ");
        else
            break;
    }
}

// Driver code
var k = 4;
var arr = [8, 5, 10, 7, 9, 4, 15,
              12, 90, 13];
var n = arr.length;
// Function to construct the
// segment tree
constructST(1, 0, n - 1, arr);
printKMax(n, k);


</script>

Output:

10 10 10 15 15 90 90

Time Complexity: O(N * log K)
Auxiliary Space: O(N * log K)

Related Article: Sliding Window Maximum (Maximum of all subarrays of size k)
Related Topic: Segment Tree

Build a segment tree for N-ary rooted tree

saurabhshadow

Improve

Article Tags :

Practice Tags :

Maximum of all subarrays of size K using Segment Tree

Similar Reads

Lazy Propagation

Thank You!

What kind of Experience do you want to share?