hdu 1031 (partial sort problem, nth_element, stable_partition, lambda expression) ...

partial sort.
first use std::nth_element to find pivot,
then use std::stable_partition with the pivot to partition the largest k, whose indices are in acsending order, print them in reverse order.
p.s. lambda expression is also used.
STL is powerful.

#include <cstdio>
#include <vector>
#include <algorithm>

struct IRPair{
    int ind;
    double rate;
};

int main() {
    //freopen("input.txt","r",stdin);
    int num_people, num_candidate, num_final, i,j;
    double tmp;
    std::vector<IRPair> vec;
    std::vector<double> ratings;
    while(scanf("%d%d%d",&num_people,&num_candidate,&num_final)!=EOF) {
        vec.resize(num_candidate);
        ratings.clear(); ratings.resize(num_candidate);
        for(i=0;i<num_people;++i)
        for(j=0;j<num_candidate;++j) { scanf("%lf",&tmp); ratings[j]+=tmp; }
        for(i=0;i<num_candidate;++i) { vec[i].ind=i+1; vec[i].rate=ratings[i]; }
        std::nth_element(&ratings[0],&ratings[num_candidate-num_final],&ratings[num_candidate]);
        tmp=ratings[num_candidate-num_final];
        std::stable_partition(&vec[0],&vec[num_candidate],[tmp](const IRPair &x) { return x.rate>=tmp; });
        for(i=num_final-1;i>0;--i) { printf("%d ",vec[i].ind);}
        printf("%d\n",vec[0].ind);
    }
    return 0;
}

156 ms for double, vector
140 ms for float, vector
140 ms for double, array
124 ms for float, arry
time for array / time for vector = 90%

版权声明：本文为博主原创文章，未经博主允许不得转载。// p.s. If in any way improment can be achieved, better performance or whatever, it will be well-appreciated to let me know, thanks in advance.

转载于:https://www.cnblogs.com/qeatzy/p/4716233.html