Arithmetic Progressions

Given N integers A1, A2, …. AN, Dexter wants to know how many ways he can choose three numbers such that they are three consecutive terms of an arithmetic progression.

Meaning that, how many triplets (i, j, k) are there such that 1 ≤ i < j < k ≤ Nand Aj - Ai = Ak - Aj.

So the triplets (2, 5, 8), (10, 8, 6), (3, 3, 3) are valid as they are three consecutive terms of an arithmetic
progression. But the triplets (2, 5, 7), (10, 6, 8) are not.

Input

First line of the input contains an integer N (3 ≤ N ≤ 100000). Then the following line contains N space separated integers A1, A2, …, AN and they have values between 1 and 30000 (inclusive).

Output

Output the number of ways to choose a triplet such that they are three consecutive terms of an arithmetic progression.

Example

Input:
10
3 5 3 6 3 4 10 4 5 2 Output:
9

Explanation

The followings are all 9 ways to choose a triplet

1 : (i, j, k) = (1, 3, 5), (Ai, Aj, Ak) = (3, 3, 3)
2 : (i, j, k) = (1, 6, 9), (Ai, Aj, Ak) = (3, 4, 5)
3 : (i, j, k) = (1, 8, 9), (Ai, Aj, Ak) = (3, 4, 5)
4 : (i, j, k) = (3, 6, 9), (Ai, Aj, Ak) = (3, 4, 5)
5 : (i, j, k) = (3, 8, 9), (Ai, Aj, Ak) = (3, 4, 5)
6 : (i, j, k) = (4, 6, 10), (Ai, Aj, Ak) = (6, 4, 2)
7 : (i, j, k) = (4, 8, 10), (Ai, Aj, Ak) = (6, 4, 2)
8 : (i, j, k) = (5, 6, 9), (Ai, Aj, Ak) = (3, 4, 5)
9 : (i, j, k) = (5, 8, 9), (Ai, Aj, Ak) = (3, 4, 5)

题解:

    考虑分块,分成block块

    假设三个点都在同一块,那么我们就在一块内暴力,复杂度block * ( n/block)  * (n/block)

    假设其中两个点在同一块,那么枚举其中一块的两个点算答案,block * n/block * n/block

  ·  假设三个点都不在同一块,枚举中间点属于的那一块 剩下左边和右边进行 FFT, 复杂度block * (n*logn)

    

#include<bits/stdc++.h>
using namespace std;
#pragma comment(linker, "/STACK:102400000,102400000")
#define ls i<<1
#define rs ls | 1
#define mid ((ll+rr)>>1)
#define pii pair<int,int>
#define MP make_pair
typedef long long LL;
typedef unsigned long long ULL;
const long long INF = 1e18+1LL;
const double pi = acos(-1.0);
const int N = 3e5+, M = 1e6+, mod = 1e9+,inf = 2e9; struct Complex {
double r , i ;
Complex () {}
Complex ( double r , double i ) : r ( r ) , i ( i ) {}
Complex operator + ( const Complex& t ) const {
return Complex ( r + t.r , i + t.i ) ;
}
Complex operator - ( const Complex& t ) const {
return Complex ( r - t.r , i - t.i ) ;
}
Complex operator * ( const Complex& t ) const {
return Complex ( r * t.r - i * t.i , r * t.i + i * t.r ) ;
}
} ; void FFT ( Complex y[] , int n , int rev ) {
for ( int i = , j , t , k ; i < n ; ++ i ) {
for ( j = , t = i , k = n >> ; k ; k >>= , t >>= ) j = j << | t & ;
if ( i < j ) swap ( y[i] , y[j] ) ;
}
for ( int s = , ds = ; s <= n ; ds = s , s <<= ) {
Complex wn = Complex ( cos ( rev * * pi / s ) , sin ( rev * * pi / s ) ) , w ( , ) , t ;
for ( int k = ; k < ds ; ++ k , w = w * wn ) {
for ( int i = k ; i < n ; i += s ) {
y[i + ds] = y[i] - ( t = w * y[i + ds] ) ;
y[i] = y[i] + t ;
}
}
}
if ( rev == - ) for ( int i = ; i < n ; ++ i ) y[i].r /= n ;
}
Complex s[N],t[N]; LL cnt[][];
int a[N];
int n,block,pos[N];
LL vis[N];
int main() {
while(scanf("%d",&n)!=EOF) {
block = ;
for(int i = ; i <= n; ++i)
pos[i] = (i-)/block + ;
int mx = -;
for(int i = ; i <= pos[n]; ++i)
for(int j = ; j <= ; ++j) cnt[i][j] = ;
for(int i = ; i <= n; ++i) {
scanf("%d",&a[i]);
mx = max(mx,a[i]);
cnt[pos[i]][a[i]]++;
} for(int i = ; i <= mx; ++i) {
for(int j = ; j <= pos[n]; ++j) {
cnt[j][i] += cnt[j-][i];
}
}
int len = ;
while(len <= *mx) len<<=;
LL ans = ;
for(int k = ; k <= pos[n]; ++k) {
for(int i = (k-)*block + ; i <= min(k*block,n); ++i) {
for(int j = i + ; j <= min(k*block,n); ++j) {
if(*a[i] - a[j] >= && *a[i] - a[j] <= mx)
ans += cnt[k-][*a[i] - a[j]] + vis[*a[i]-a[j]];
if(*a[j] - a[i] >= && *a[j] - a[i] <= mx)
ans += cnt[pos[n]][*a[j] - a[i]] - cnt[k][*a[j] - a[i]];
}
vis[a[i]] += ;
}
for(int i = (k-)*block + ; i <= min(k*block,n); ++i) {
vis[a[i]] = ;
} for(int j = ; j <= mx; ++j)
s[j] = Complex(cnt[k-][j],);
for(int j = mx+; j < len; ++j) s[j] = Complex(,); for(int j = ; j <= mx; ++j)
t[j] = Complex(cnt[pos[n]][j] - cnt[k][j] , );
for(int j = mx+; j < len; ++j) t[j] = Complex(,); FFT(s,len,);FFT(t,len,);
for(int j = ; j < len; ++j) s[j] = s[j] * t[j];
FFT(s,len,-); for(int j = ; j <= mx; ++j) {
LL tmp = (LL)(s[*j].r + 0.5);
ans += tmp*(cnt[k][j] - cnt[k-][j]);
}
}
printf("%lld\n",ans); }
return ;
}

  

CodeChef - COUNTARI FTT+分块的更多相关文章

  1. [BZOJ 3509] [CodeChef] COUNTARI (FFT+分块)

    [BZOJ 3509] [CodeChef] COUNTARI (FFT+分块) 题面 给出一个长度为n的数组,问有多少三元组\((i,j,k)\)满足\(i<j<k,a_j-a_i=a_ ...

  2. BZOJ3509 [CodeChef] COUNTARI 【分块 + fft】

    题目链接 BZOJ3509 题解 化一下式子,就是 \[2A[j] = A[i] + A[k]\] 所以我们对一个位置两边的数构成的生成函数相乘即可 但是由于这样做是\(O(n^2logn)\)的,我 ...

  3. bzoj 3509: [CodeChef] COUNTARI] [分块 生成函数]

    3509: [CodeChef] COUNTARI 题意:统计满足\(i<j<k, 2*a[j] = a[i] + a[k]\)的个数 \(2*a[j]\)不太好处理,暴力fft不如直接暴 ...

  4. BZOJ 3509: [CodeChef] COUNTARI

    3509: [CodeChef] COUNTARI Time Limit: 40 Sec  Memory Limit: 128 MBSubmit: 883  Solved: 250[Submit][S ...

  5. BZOJ3509: [CodeChef] COUNTARI

    3509: [CodeChef] COUNTARI Time Limit: 40 Sec  Memory Limit: 128 MBSubmit: 339  Solved: 85[Submit][St ...

  6. CodeChef COUNTARI Arithmetic Progressions(分块 + FFT)

    题目 Source http://vjudge.net/problem/142058 Description Given N integers A1, A2, …. AN, Dexter wants ...

  7. BZOJ 3509 [CodeChef] COUNTARI ——分块 FFT

    分块大法好. 块内暴力,块外FFT. 弃疗了,抄SX队长$silvernebula$的代码 #include <map> #include <cmath> #include & ...

  8. CodeChef FNCS (分块+树状数组)

    题目:https://www.codechef.com/problems/FNCS 题解: 我们知道要求区间和的时候,我们用前缀和去优化.这里也是一样,我们要求第 l 个函数到第 r 个函数 [l, ...

  9. CodeChef - COUNTARI Arithmetic Progressions (FFT)

    题意:求一个序列中,有多少三元组$(i,j,k)i<j<k $ 满足\(A_i + A_k = 2*A_i\) 构成等差数列. https://www.cnblogs.com/xiuwen ...

随机推荐

  1. soa服务治理

    SOA服务治理 文章:SOA 治理简介 文章:中小型互联网公司微服务实践-经验和教训

  2. Python 频繁读取Mysql相关问题

    1.需要频繁select大量数据,时间长.消耗内存大,如何解决mysql性能问题? 如果对返回的结果数量没有要求,可以控制返回的数量: cursor.fetchmany(size=1000) 这样是只 ...

  3. 使用Apriori算法进行关联分析

    关联分析是一种在大规模数据集中寻找有趣关系的任务.这些关系可以有两种形式:频繁项集或者关联规则.频繁项集是指经常出现在一块的物品的集合,关联规则暗示两种物品之间可能存在很强的关系.一个项集的支持度被定 ...

  4. 【2018.10.18】noip模拟赛Day2 地球危机(2018年第九届蓝桥杯C/C++A组省赛 三体攻击)

    题目描述 三体人将对地球发起攻击.为了抵御攻击,地球人派出了 $A × B × C$ 艘战舰,在太 空中排成一个 $A$ 层 $B$ 行 $C$ 列的立方体.其中,第 $i$ 层第 $j$ 行第 $k ...

  5. ndarray:一种多维数组对象

    ndarray是一个通用的同构数据多维容器,也就是说,其中的所有元素必须是相同类型的.每个数组都有一个shape(一个表示各维度大小的元组)和一个dtype(一个用于说明数组数据类型的对象). In ...

  6. net3:Button的CommandName使用,AdRotator,BulletedList的使用

    原文发布时间为:2008-07-29 -- 来源于本人的百度文章 [由搬家工具导入] using System;using System.Data;using System.Configuration ...

  7. 定时任务-Quartz

    Quartz Quartz w3c教程 参考:https://blog.csdn.net/lkl_csdn/article/details/73613033 Quartz 的使用 https://ww ...

  8. HDU 4770 Lights Against Dudely 暴力枚举+dfs

    又一发吐血ac,,,再次明白了用函数(代码重用)和思路清晰的重要性. 11779687 2014-10-02 20:57:53 Accepted 4770 0MS 496K 2976 B G++ cz ...

  9. HDU 3001 三进制 状压dp

    Travelling Time Limit: 6000/3000 MS (Java/Others)    Memory Limit: 32768/32768 K (Java/Others) Total ...

  10. PatentTips - Write Combining Buffer for Sequentially Addressed Partial Line Operations

    SUMMARY OF THE INVENTION The present invention pertains to a write combining buffer for use in a mic ...