Iahub accidentally discovered a secret lab. He found there n devices ordered in a line, numbered from 1 to n from left to right. Each device i (1 ≤ i ≤ n) can create either ai units of matter or ai units of antimatter.
Iahub wants to choose some contiguous subarray of devices in the lab, specify the production mode for each of them (produce matter or antimatter) and finally take a photo of it. However he will be successful only if the amounts of matter and antimatter produced in the selected subarray will be the same (otherwise there would be overflowing matter or antimatter in the photo).
You are requested to compute the number of different ways Iahub can successful take a photo. A photo is different than another if it represents another subarray, or if at least one device of the subarray is set to produce matter in one of the photos and antimatter in the other one.
The first line contains an integer n (1 ≤ n ≤ 1000). The second line contains n integers a1, a2, ..., an (1 ≤ ai ≤ 1000).
The sum a1 + a2 + ... + an will be less than or equal to 10000.
Output a single integer, the number of ways Iahub can take a photo, modulo 1000000007 (109 + 7).
4 1 1 1 1
12
The possible photos are [1+, 2-], [1-, 2+], [2+, 3-], [2-, 3+], [3+, 4-], [3-, 4+], [1+, 2+, 3-, 4-], [1+, 2-, 3+, 4-], [1+, 2-, 3-, 4+], [1-, 2+, 3+, 4-], [1-, 2+, 3-, 4+] and [1-, 2-, 3+, 4+], where "i+" means that the i-th element produces matter, and "i-" means that the i-th element produces antimatter.
题意:物质为正,反物质为负,组合起来就是+或者 - ,然后题目求的是:找出有多少种组合情况为0的数目(加起来为0)。
感想:这题我也不知道如何下手,因为这段时间专攻DP,所以一看这题目也知道肯定是DP做的了,因为如果模拟组合情况的话,那么就是指数级别的了,肯定超时,所以只有DP了。但是也不知道这题如何DP,好难!根本就想不出来。本来我也不太会DP的状态方程,然后就看了一眼别人的答案,觉得如果代码太长就不做了,没想到代码就几句就搞定了,顿觉自己见识浅短啊!
然后重新看题目的类型,原来是分治的,不过以前也没有做过分治的题目,只知道分治的基本概念……所以还是研究别人的代码吧。即使自己不知道怎么做,但是刚开始总得先研究别人怎么做,以后自己慢慢的才会做……但是……情况不好……根本就看不懂别人这几行代码是什么意思啊????????!!!!!!!!!!
又问了琦神!然后也不太懂,最后想想,还是用最笨的方法来理解别人的代码吧……就是把数代进去,把中间的结果都写出来!因为是分治,所以直接画一条直线,中间的就是分治中的边界(在这里是11000),然后左边小于11000,右边大于110000,OK……因为题目给出的样例都是1,中间结果看起来没代表性,所以自己想出一个样例:
5
1 3 4 2 4
依次代入各个数的时候就发现了其中的奥秘了……j-a 与 j+a 其实宏观来说就是 -a 与 + a,只不过因为用DP解决,所以数组下标不能为负数,所以才要引入 j 这个数来把数组的下标往后移的,j 这个数也就是分治的边界了当然。 j 依次往上递增就可以模拟出了各种组合的情况了,太强大了……因为每种状态的组合情况都是在前一种的基础上拓展的,所以所以每次组合的结果都要记下来,当然如果DP数组是三维的话,就不用每次都有这句话了:ans = (ans + d[i][11000]) % INF; 但是依然可以把它简化为二维……仔细想想就明白了,因为状态只有2种嘛(前一种状态与当前状态)……
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include