結果
| 問題 | No.1117 数列分割 |
| コンテスト | |
| ユーザー |
theory_and_me
|
| 提出日時 | 2020-07-18 02:01:55 |
| 言語 | C++17 (gcc 13.3.0 + boost 1.89.0) |
| 結果 |
AC
|
| 実行時間 | 2,135 ms / 3,000 ms |
| コード長 | 3,736 bytes |
| 記録 | |
| コンパイル時間 | 2,191 ms |
| コンパイル使用メモリ | 206,652 KB |
| 最終ジャッジ日時 | 2025-01-11 23:44:43 |
|
ジャッジサーバーID (参考情報) |
judge3 / judge4 |
(要ログイン)
| ファイルパターン | 結果 |
|---|---|
| sample | AC * 3 |
| other | AC * 26 |
ソースコード
#include <bits/stdc++.h>
using namespace std;
typedef unsigned long long ull;
typedef long long ll;
typedef pair<int, int> pii;
typedef pair<ll, ll> pll;
typedef pair<double, double> pdd;
typedef vector<ll> vl;
typedef vector<vector<ll>> vvl;
//typedef vector<vector<ll>> Graph;
const ll mod = 1e9 + 7;
//const ll mod = 998244353;
#define REP(i,n) for(ll i=0;i<(ll)n;i++)
#define dump(x) cerr << #x << " = " << (x) << endl;
#define spa << " " <<
#define fi first
#define se second
template<class T> bool chmax(T &a, const T &b) { if (a<b) { a=b; return 1; } return 0; }
template<class T> bool chmin(T &a, const T &b) { if (a>b) { a=b; return 1; } return 0; }
template<class S, class T> ostream& operator << (ostream& os, const pair<S, T> v){
os << "(" << v.first << ", " << v.second << ")"; return os;
}
template<class T> ostream& operator << (ostream& os, const vector<T> v){
for(int i = 0; i < (int)v.size(); i++){if(i > 0){os << " ";} os << v[i];} return os;
}
template<class T> ostream& operator << (ostream& os, const vector<vector<T>> v){
for(int i = 0; i < (int)v.size(); i++){if(i > 0){os << endl;} os << v[i];} return os;
}
template<typename T> void debug(vector<vector<T>>&v,ll h,ll w){for(ll i=0;i<h;i++)
{cerr<<v[i][0];for(ll j=1;j<w;j++)cerr spa v[i][j];cerr<<endl;}};
template<typename T> void debug(vector<T>&v,ll n){if(n!=0)cerr<<v[0];
for(ll i=1;i<n;i++)cerr spa v[i];
cerr<<endl;};
template <class T, class Op> class queue_aggregation {
private:
class node {
public:
T val, sum;
node(const T &val, const T &sum) : val(val), sum(sum) {}
};
Op op;
std::stack<node> front_stack, back_stack;
public:
queue_aggregation(const Op &op = Op())
: op(op), front_stack(), back_stack() {}
bool empty() const { return front_stack.empty() && back_stack.empty(); }
std::size_t size() const { return front_stack.size() + back_stack.size(); }
T fold_all() const {
assert(!empty());
if (front_stack.empty()) {
return back_stack.top().sum;
} else if (back_stack.empty()) {
return front_stack.top().sum;
} else {
return op(front_stack.top().sum, back_stack.top().sum);
}
}
void push(const T &x) {
if (back_stack.empty()) {
back_stack.emplace(x, x);
} else {
T s{op(back_stack.top().sum, x)};
back_stack.emplace(x, s);
}
}
void pop() {
assert(!empty());
if (front_stack.empty()) {
front_stack.emplace(back_stack.top().val, back_stack.top().val);
back_stack.pop();
while (!back_stack.empty()) {
T s{op(back_stack.top().val, front_stack.top().sum)};
front_stack.emplace(back_stack.top().val, s);
back_stack.pop();
}
}
front_stack.pop();
}
};
int main(){
cin.tie(0);
ios::sync_with_stdio(false);
ll N, K, M;
cin >> N >> K >> M;
vl A(N);
REP(i, N) cin >> A[i];
const ll INF = (1ll<<60);
vvl dp(N+1, vl(K+1, -INF));
vvl tmp1(N+1, vl(K+1, -INF));
vvl tmp2(N+1, vl(K+1, -INF));
vl S(N+1, 0);
REP(i, N) S[i+1] = S[i] + A[i];
dp[0][0] = 0;
auto f = [](ll x, ll y){
return max(x, y);
};
for(int j=1;j<=K;j++){
REP(i, N+1) tmp1[i][j-1] = dp[i][j-1] + S[i];
REP(i, N+1) tmp2[i][j-1] = dp[i][j-1] - S[i];
queue_aggregation<ll, decltype(f)> qa1(f), qa2(f);
for(int i=1;i<=N;i++){
qa1.push(tmp1[i-1][j-1]);
qa2.push(tmp2[i-1][j-1]);
ll ma1 = qa1.fold_all();
ll ma2 = qa2.fold_all();
chmax(dp[i][j], ma1 - S[i]);
chmax(dp[i][j], ma2 + S[i]);
if(i>=M) qa1.pop();
if(i>=M) qa2.pop();
}
}
cout << dp[N][K] << endl;
return 0;
}
theory_and_me