#include<iostream>
#include<vector>
#include<iomanip>
#include<utility>
#include<algorithm>
#define ld long double
using namespace std;
vector<pair<ld,ld>> vec;
int n,k;
ld solve(int m){
vector<vector<ld>> dp(n+1,vector<ld>(k-m+1,1e9+7));
dp[1][0]=(m>0?vec[1].first:0);
if(k-m>0) dp[1][1]=vec[1].second;
for(int i=2;i<=n;i++){
for(int j=0;j<=min(k-m,i);j++){
dp[i][j]=min(dp[i][j],dp[i-1][j]+(i-j>m?0:vec[i].first/float(i-j)));
if(j>0) dp[i][j]=min(dp[i][j],dp[i-1][j-1]+vec[i].second/float(m+1));
}
}
/*for(int i=1;i<=n;i++){
for(int j=0;j<=k-m;j++) if(m==2) cout<<dp[i][j]<<" ";
cout<<"\n";
}*/
return dp[n][k-m];
}
int main(){
ios::sync_with_stdio(false);
cin.tie(0);
cin>>n>>k;
vec.resize(n+1);
for(int i=1;i<=n;i++) cin>>vec[i].second>>vec[i].first;
for(int i=1;i<=n;i++) if(vec[i].first==-1) vec[i].first=1e9;
sort(next(vec.begin()),vec.end());
int l=0,r=k;
ld ans=1e9+7;
while(l<r){
int mid=(l+r)>>1;
ld a=solve(max(mid,0)),b=solve(min(mid+1,r));
ans=min(ans,min(a,b));
if(a>b) l=min(mid+1,r);
else r=(max(mid,0));
}
for(int i=0;i<=k;i++) ans=min(solve(i),ans);
cout<<fixed<<setprecision(12)<<ans;
}
Compilation message
Main.cpp:43:2: error: extended character is not valid in an identifier
43 | for(int i=0;i<=k;i++) ans=min(solve(i),ans);
| ^
Main.cpp:44:2: error: extended character is not valid in an identifier
44 | cout<<fixed<<setprecision(12)<<ans;
| ^
Main.cpp:44:5: error: extended character is not valid in an identifier
44 | cout<<fixed<<setprecision(12)<<ans;
| ^
Main.cpp: In function 'int main()':
Main.cpp:43:8: error: expected primary-expression before 'int'
43 | for(int i=0;i<=k;i++) ans=min(solve(i),ans);
| ^~~
Main.cpp:43:16: error: 'i' was not declared in this scope
43 | for(int i=0;i<=k;i++) ans=min(solve(i),ans);
| ^
Main.cpp:44:2: error: '\U000000a0' was not declared in this scope
44 | cout<<fixed<<setprecision(12)<<ans;
| ^
