답안 #865809

# 제출 시각 아이디 문제 언어 결과 실행 시간 메모리
865809 2023-10-24T17:26:23 Z green_gold_dog Seesaw (JOI22_seesaw) C++17
100 / 100
71 ms 14024 KB
//#pragma GCC optimize("Ofast")
//#pragma GCC target("avx,avx2,sse,sse2,sse3,ssse3,sse4,abm,popcnt,mmx")
#include <bits/stdc++.h>

using namespace std;

typedef long long ll;
typedef double db;
typedef long double ldb;
typedef complex<double> cd;

constexpr ll INF64 = 9'000'000'000'000'000'000, INF32 = 2'000'000'000, MOD = 1'000'000'007;
constexpr db PI = acos(-1);
constexpr bool IS_FILE = false, IS_TEST_CASES = false;

random_device rd;
mt19937 rnd32(rd());
mt19937_64 rnd64(rd());

template<typename T>
bool assign_max(T& a, T b) {
        if (b > a) {
                a = b;
                return true;
        }
        return false;
}

template<typename T>
bool assign_min(T& a, T b) {
        if (b < a) {
                a = b;
                return true;
        }
        return false;
}

template<typename T>
T square(T a) {
        return a * a;
}

template<>
struct std::hash<pair<ll, ll>> {
        ll operator() (pair<ll, ll> p) const {
                return ((__int128)p.first * MOD + p.second) % INF64;
        }
};

void solve() {
        ll n;
        cin >> n;
        vector<ll> arr(n);
        ll sum = 0;
        for (ll i = 0; i < n; i++) {
                cin >> arr[i];
                sum += arr[i];
        }
        vector<tuple<ll, ll, ll>> now(1, make_tuple(sum, 0, n - 1));
        long double st = sum / (long double)n;
        long double mi = st, ma = st;
        vector<pair<long double, long double>> a;
        for (ll j = 1; j < n; j++) {
                vector<tuple<ll, ll, ll>> nnow;
                for (auto[x, l, r] : now) {
                        nnow.emplace_back(x - arr[l], l + 1, r);
                        nnow.emplace_back(x - arr[r], l, r - 1);
                }
                now.clear();
                sort(nnow.begin(), nnow.end());
                for (ll i = 0; i <= nnow.size(); i++) {
                        if ((i == 0 || get<0>(nnow[i - 1]) / (long double)(n - j) <= st) && (i == nnow.size() || get<0>(nnow[i]) / (long double)(n - j) >= st)) {
                                if (i > 0) {
                                        now.push_back(nnow[i - 1]);
                                }
                                if (i < nnow.size()) {
                                        now.push_back(nnow[i]);
                                }
                                break;
                        }
                }
                if (now.size() == 1) {
                        assign_min(mi, get<0>(now[0]) / (long double)(n - j));
                        assign_max(ma, get<0>(now[0]) / (long double)(n - j));
                        continue;
                }
                a.emplace_back(get<0>(now[0]) / (long double)(n - j), get<0>(now[1]) / (long double)(n - j));
        }
        sort(a.begin(), a.end());
        long double ans = ma - min(mi, (a.empty() ? mi : a[0].first));
        for (ll i = 0; i < a.size(); i++) {
                assign_max(ma, a[i].second);
                assign_min(ans, ma - min(mi, (i + 1 < a.size() ? a[i + 1].first : mi)));
        }
        cout << setprecision(20) << ans << '\n';
}

int main() {
        if (IS_FILE) {
                freopen("", "r", stdin);
                freopen("", "w", stdout);
        }
        ios_base::sync_with_stdio(false);
        cin.tie(0);
        cout.tie(0);
        ll t = 1;
        if (IS_TEST_CASES) {
                cin >> t;
        }
        for (ll i = 0; i < t; i++) {
                solve();
        }
}

Compilation message

seesaw.cpp: In function 'void solve()':
seesaw.cpp:71:34: warning: comparison of integer expressions of different signedness: 'll' {aka 'long long int'} and 'std::vector<std::tuple<long long int, long long int, long long int> >::size_type' {aka 'long unsigned int'} [-Wsign-compare]
   71 |                 for (ll i = 0; i <= nnow.size(); i++) {
      |                                ~~^~~~~~~~~~~~~~
seesaw.cpp:72:96: warning: comparison of integer expressions of different signedness: 'll' {aka 'long long int'} and 'std::vector<std::tuple<long long int, long long int, long long int> >::size_type' {aka 'long unsigned int'} [-Wsign-compare]
   72 |                         if ((i == 0 || get<0>(nnow[i - 1]) / (long double)(n - j) <= st) && (i == nnow.size() || get<0>(nnow[i]) / (long double)(n - j) >= st)) {
      |                                                                                              ~~^~~~~~~~~~~~~~
seesaw.cpp:76:39: warning: comparison of integer expressions of different signedness: 'll' {aka 'long long int'} and 'std::vector<std::tuple<long long int, long long int, long long int> >::size_type' {aka 'long unsigned int'} [-Wsign-compare]
   76 |                                 if (i < nnow.size()) {
      |                                     ~~^~~~~~~~~~~~~
seesaw.cpp:91:26: warning: comparison of integer expressions of different signedness: 'll' {aka 'long long int'} and 'std::vector<std::pair<long double, long double> >::size_type' {aka 'long unsigned int'} [-Wsign-compare]
   91 |         for (ll i = 0; i < a.size(); i++) {
      |                        ~~^~~~~~~~~~
seesaw.cpp:93:53: warning: comparison of integer expressions of different signedness: 'll' {aka 'long long int'} and 'std::vector<std::pair<long double, long double> >::size_type' {aka 'long unsigned int'} [-Wsign-compare]
   93 |                 assign_min(ans, ma - min(mi, (i + 1 < a.size() ? a[i + 1].first : mi)));
      |                                               ~~~~~~^~~~~~~~~~
seesaw.cpp: In function 'int main()':
seesaw.cpp:100:24: warning: ignoring return value of 'FILE* freopen(const char*, const char*, FILE*)' declared with attribute 'warn_unused_result' [-Wunused-result]
  100 |                 freopen("", "r", stdin);
      |                 ~~~~~~~^~~~~~~~~~~~~~~~
seesaw.cpp:101:24: warning: ignoring return value of 'FILE* freopen(const char*, const char*, FILE*)' declared with attribute 'warn_unused_result' [-Wunused-result]
  101 |                 freopen("", "w", stdout);
      |                 ~~~~~~~^~~~~~~~~~~~~~~~~
# 결과 실행 시간 메모리 Grader output
1 Correct 0 ms 348 KB Output is correct
2 Correct 0 ms 348 KB Output is correct
3 Correct 0 ms 344 KB Output is correct
# 결과 실행 시간 메모리 Grader output
1 Correct 0 ms 348 KB Output is correct
2 Correct 0 ms 348 KB Output is correct
3 Correct 0 ms 344 KB Output is correct
4 Correct 0 ms 348 KB Output is correct
5 Correct 0 ms 348 KB Output is correct
6 Correct 0 ms 348 KB Output is correct
7 Correct 0 ms 348 KB Output is correct
# 결과 실행 시간 메모리 Grader output
1 Correct 0 ms 348 KB Output is correct
2 Correct 0 ms 348 KB Output is correct
3 Correct 0 ms 344 KB Output is correct
4 Correct 0 ms 348 KB Output is correct
5 Correct 0 ms 348 KB Output is correct
6 Correct 0 ms 348 KB Output is correct
7 Correct 0 ms 348 KB Output is correct
8 Correct 1 ms 604 KB Output is correct
9 Correct 1 ms 600 KB Output is correct
10 Correct 1 ms 604 KB Output is correct
11 Correct 1 ms 476 KB Output is correct
# 결과 실행 시간 메모리 Grader output
1 Correct 0 ms 348 KB Output is correct
2 Correct 0 ms 348 KB Output is correct
3 Correct 0 ms 344 KB Output is correct
4 Correct 0 ms 348 KB Output is correct
5 Correct 0 ms 348 KB Output is correct
6 Correct 0 ms 348 KB Output is correct
7 Correct 0 ms 348 KB Output is correct
8 Correct 1 ms 604 KB Output is correct
9 Correct 1 ms 600 KB Output is correct
10 Correct 1 ms 604 KB Output is correct
11 Correct 1 ms 476 KB Output is correct
12 Correct 69 ms 13036 KB Output is correct
13 Correct 71 ms 13768 KB Output is correct
14 Correct 69 ms 13000 KB Output is correct
15 Correct 70 ms 12752 KB Output is correct
16 Correct 69 ms 14024 KB Output is correct