Big Data applications for heap sequencing and heap

This exercise is the big data application of heap sorting and heap
The time complexity of heap sorting is O (n)
Heap of big data applications should choose a small heap for processing
But when the data is more than 100000, the speed is significantly slower, it may be the time to build a small heap slow "" "There is no better way



#include <iostream>
#include <vector>
#include <time.h>
using namespace Std;


//........................ The following is a heap sort ....... ..... ...............
void Adjustdowngreater (vector<int>& h,size_t size)//Build Large heap
{
if (size <= 1)//note parameter detection ah ....
{
Return
}
for (int parent = (SIZE-1-1)/2; parent >= 0; parent--)//experience: You can write the inner loop first, then the outer loop
{
while (parent <= size-1)
{
int Leftcchild = 2 * parent + 1;

if (Leftcchild + 1 <= size-1 && H[leftcchild] < H[leftcchild + 1])
{
leftcchild++;
}

if (leftcchild <= size-1 && h[parent] < H[leftcchild])
{
Swap (H[parent], h[leftcchild]);
parent = Leftcchild;
}
Else
{
Break
}
}
}
}

void Heapsort (vector<int>& h,size_t size)
{
while (Size > 1)
{
Swap (h[0], h[size-1]);
size--;

Adjustdowngreater (h,size);
}

}

void Print (vector<int>& h,size_t size)
{
for (int i = 0; i < size; i++)
{
cout << h[i]<< "";
}
}

void Test1 ()
{
int arr[] = {10, 16, 18, 12, 11, 13, 15, 17, 14, 19};

Vector<int> h;
for (int i = 0; i <; i++)
{
H.push_back (Arr[i]);
}

Adjustdowngreater (H, 10);
Heapsort (H, 10);
Print (H, 10);
}


//.................................. The following is a heap big data application ....... .....


The largest number of the top k from n data is ******* "small heap" **********************
The reason for choosing a small heap: Use an array to store the number of K arr[k] If the first number you choose is the biggest one, then you won't be able to get arr in the back.


const int N = 100000; When the speed of more than 100000 is significantly slower "there is no better way" "
const int K = 10;

void GetData (vector<int>& h)
{
Srand (Time (0));
for (int i = 0; i < N; i++)
{
H.push_back (rand ()% N + 1);
}
}

void Adjustdownless (vector<int>& h, size_t size)//Build small heap
{
if (size <= 1)
{
Return
}
for (int parent = (SIZE-1-1)/2; parent >= 0; parent--)
{
while (parent <= size-1)
{
int Leftcchild = 2 * parent + 1;

if (Leftcchild + 1 <= size-1 && H[leftcchild] > h[leftcchild + 1])
{
leftcchild++;
}

if (leftcchild <= size-1 && h[parent] > H[leftcchild])
{
Swap (H[parent], h[leftcchild]);
parent = Leftcchild;
}
Else
{
Break
}
}
}
}

void Getgreatestk (vector<int>& H, vector<int>& greatest)
{
Adjustdownless (H, N);

for (int i = 0; i < K; i++)//Put the front K smaller push in
{
Greatest.push_back (H[i]);
}

for (int index = K; index < N; index++)
{
Adjustdownless (Greatest, K); In order to get the smallest number

if (H[index] > Greatest[0])//Replace the smallest number
{
Greatest[0] = H[index];
}
}
}

void Print (vector<int>& h)
{
for (int i = 0; i < K; i++)
{
cout << h[i]<< "";
}
}

void Test2 ()
{
Vector<int> h;
Vector<int> greatest;
GetData (h);
GETGREATESTK (H, greatest);
Print (greatest);
}


int main ()
{
Test1 ();
Test2 ();

return 0;
}650) this.width=650; "src=" Http://s3.51cto.com/wyfs02/M02/7E/D8/wKiom1cKHXyj8EqyAAA0CYgicTU561.png "title=" 2016-04-10 17:31:27 screen. png "alt=" Wkiom1ckhxyj8eqyaaa0cygictu561.png "/>

This article is from "Narcissus" blog, please make sure to keep this source http://10704527.blog.51cto.com/10694527/1762368

Big Data applications for heap sequencing and heap