message stringlengths 2 433k | message_type stringclasses 2
values | message_id int64 0 1 | conversation_id int64 113 108k | cluster float64 12 12 | __index_level_0__ int64 226 217k |
|---|---|---|---|---|---|
Provide tags and a correct Python 3 solution for this coding contest problem.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a. | instruction | 0 | 56,630 | 12 | 113,260 |
Tags: data structures, greedy, implementation
Correct Solution:
```
import sys
input = sys.stdin.readline
t=int(input())
for tests in range(t):
n=int(input())
A=list(map(int,input().split()))
compression_dict={a: ind for ind, a in enumerate(sorted(set(A)))}
A=[compression_dict[a]+3 for a in A] # [2, 3, 0, 1, 4]
LEN=n+6
BIT=[0]*(LEN+1)
USE=[0]*(LEN+1)
def update(v,w):
while v<=LEN:
BIT[v]+=w
v+=(v&(-v))
def getvalue(v):
ANS=0
while v!=0:
ANS+=BIT[v]
v-=(v&(-v))
return ANS
def bisect_on_BIT(x):
if x<=0:
return 0
ANS=0
h=1<<(LEN.bit_length()-1)
while h>0:
if ANS+h<=LEN and BIT[ANS+h]<x:
x-=BIT[ANS+h]
ANS+=h
h//=2
return ANS+1
USE[1]=1
USE[A[0]]=1
USE[n+5]=1
update(1,1)
update(A[0],1)
update(n+5,1)
before=A[0]
for a in A[1:]:
#print(a)
#print([getvalue(x) for x in range(n+6)])
if a==before:
continue
elif a>before:
x=getvalue(before)
u=bisect_on_BIT(x+1)
if a<=u:
if USE[a]==0:
USE[a]=1
update(a,1)
else:
print("NO")
break
else:
x=getvalue(before)
u=bisect_on_BIT(x-1)
#print(before,x,u)
if a>=u:
if USE[a]==0:
USE[a]=1
update(a,1)
else:
print("NO")
break
before=a
else:
print("YES")
``` | output | 1 | 56,630 | 12 | 113,261 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a. | instruction | 0 | 56,631 | 12 | 113,262 |
Tags: data structures, greedy, implementation
Correct Solution:
```
INF=1e9+1
for ii in range(int(input())):
n=int(input())
b=list(map(int,input().split()))
m=b[0]
l=[-INF]
r=[INF]
ans=1
for i in range(1,len(b)):
if m>b[i]:
if b[i]>l[-1]:
r+=[m]
elif b[i]<l[-1]:
ans=0
break
else :
r+=[m]
l.pop()
elif m<b[i]:
if b[i]<r[-1]:
l+=[m]
elif b[i]>r[-1]:
ans=0
break
else :
l+=[m]
r.pop()
m=b[i]
if(ans):
print("YES")
else :
print("NO")
``` | output | 1 | 56,631 | 12 | 113,263 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a. | instruction | 0 | 56,632 | 12 | 113,264 |
Tags: data structures, greedy, implementation
Correct Solution:
```
import io
import os
from bisect import bisect_left, bisect_right
from collections import Counter, defaultdict, deque
from heapq import heappush, heappop, heapify
from math import gcd, inf
# https://raw.githubusercontent.com/cheran-senthil/PyRival/master/pyrival/data_structures/SortedList.py
class SortedList:
def __init__(self, iterable=[], _load=200):
"""Initialize sorted list instance."""
values = sorted(iterable)
self._len = _len = len(values)
self._load = _load
self._lists = _lists = [values[i : i + _load] for i in range(0, _len, _load)]
self._list_lens = [len(_list) for _list in _lists]
self._mins = [_list[0] for _list in _lists]
self._fen_tree = []
self._rebuild = True
def _fen_build(self):
"""Build a fenwick tree instance."""
self._fen_tree[:] = self._list_lens
_fen_tree = self._fen_tree
for i in range(len(_fen_tree)):
if i | i + 1 < len(_fen_tree):
_fen_tree[i | i + 1] += _fen_tree[i]
self._rebuild = False
def _fen_update(self, index, value):
"""Update `fen_tree[index] += value`."""
if not self._rebuild:
_fen_tree = self._fen_tree
while index < len(_fen_tree):
_fen_tree[index] += value
index |= index + 1
def _fen_query(self, end):
"""Return `sum(_fen_tree[:end])`."""
if self._rebuild:
self._fen_build()
_fen_tree = self._fen_tree
x = 0
while end:
x += _fen_tree[end - 1]
end &= end - 1
return x
def _fen_findkth(self, k):
"""Return a pair of (the largest `idx` such that `sum(_fen_tree[:idx]) <= k`, `k - sum(_fen_tree[:idx])`)."""
_list_lens = self._list_lens
if k < _list_lens[0]:
return 0, k
if k >= self._len - _list_lens[-1]:
return len(_list_lens) - 1, k + _list_lens[-1] - self._len
if self._rebuild:
self._fen_build()
_fen_tree = self._fen_tree
idx = -1
for d in reversed(range(len(_fen_tree).bit_length())):
right_idx = idx + (1 << d)
if right_idx < len(_fen_tree) and k >= _fen_tree[right_idx]:
idx = right_idx
k -= _fen_tree[idx]
return idx + 1, k
def _delete(self, pos, idx):
"""Delete value at the given `(pos, idx)`."""
_lists = self._lists
_mins = self._mins
_list_lens = self._list_lens
self._len -= 1
self._fen_update(pos, -1)
del _lists[pos][idx]
_list_lens[pos] -= 1
if _list_lens[pos]:
_mins[pos] = _lists[pos][0]
else:
del _lists[pos]
del _list_lens[pos]
del _mins[pos]
self._rebuild = True
def _loc_left(self, value):
"""Return an index pair that corresponds to the first position of `value` in the sorted list."""
if not self._len:
return 0, 0
_lists = self._lists
_mins = self._mins
lo, pos = -1, len(_lists) - 1
while lo + 1 < pos:
mi = (lo + pos) >> 1
if value <= _mins[mi]:
pos = mi
else:
lo = mi
if pos and value <= _lists[pos - 1][-1]:
pos -= 1
_list = _lists[pos]
lo, idx = -1, len(_list)
while lo + 1 < idx:
mi = (lo + idx) >> 1
if value <= _list[mi]:
idx = mi
else:
lo = mi
return pos, idx
def _loc_right(self, value):
"""Return an index pair that corresponds to the last position of `value` in the sorted list."""
if not self._len:
return 0, 0
_lists = self._lists
_mins = self._mins
pos, hi = 0, len(_lists)
while pos + 1 < hi:
mi = (pos + hi) >> 1
if value < _mins[mi]:
hi = mi
else:
pos = mi
_list = _lists[pos]
lo, idx = -1, len(_list)
while lo + 1 < idx:
mi = (lo + idx) >> 1
if value < _list[mi]:
idx = mi
else:
lo = mi
return pos, idx
def add(self, value):
"""Add `value` to sorted list."""
_load = self._load
_lists = self._lists
_mins = self._mins
_list_lens = self._list_lens
self._len += 1
if _lists:
pos, idx = self._loc_right(value)
self._fen_update(pos, 1)
_list = _lists[pos]
_list.insert(idx, value)
_list_lens[pos] += 1
_mins[pos] = _list[0]
if _load + _load < len(_list):
_lists.insert(pos + 1, _list[_load:])
_list_lens.insert(pos + 1, len(_list) - _load)
_mins.insert(pos + 1, _list[_load])
_list_lens[pos] = _load
del _list[_load:]
self._rebuild = True
else:
_lists.append([value])
_mins.append(value)
_list_lens.append(1)
self._rebuild = True
def discard(self, value):
"""Remove `value` from sorted list if it is a member."""
_lists = self._lists
if _lists:
pos, idx = self._loc_right(value)
if idx and _lists[pos][idx - 1] == value:
self._delete(pos, idx - 1)
def remove(self, value):
"""Remove `value` from sorted list; `value` must be a member."""
_len = self._len
self.discard(value)
if _len == self._len:
raise ValueError("{0!r} not in list".format(value))
def pop(self, index=-1):
"""Remove and return value at `index` in sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
value = self._lists[pos][idx]
self._delete(pos, idx)
return value
def bisect_left(self, value):
"""Return the first index to insert `value` in the sorted list."""
pos, idx = self._loc_left(value)
return self._fen_query(pos) + idx
def bisect_right(self, value):
"""Return the last index to insert `value` in the sorted list."""
pos, idx = self._loc_right(value)
return self._fen_query(pos) + idx
def count(self, value):
"""Return number of occurrences of `value` in the sorted list."""
return self.bisect_right(value) - self.bisect_left(value)
def __len__(self):
"""Return the size of the sorted list."""
return self._len
def __getitem__(self, index):
"""Lookup value at `index` in sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
return self._lists[pos][idx]
def __delitem__(self, index):
"""Remove value at `index` from sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
self._delete(pos, idx)
def __contains__(self, value):
"""Return true if `value` is an element of the sorted list."""
_lists = self._lists
if _lists:
pos, idx = self._loc_left(value)
return idx < len(_lists[pos]) and _lists[pos][idx] == value
return False
def __iter__(self):
"""Return an iterator over the sorted list."""
return (value for _list in self._lists for value in _list)
def __reversed__(self):
"""Return a reverse iterator over the sorted list."""
return (value for _list in reversed(self._lists) for value in reversed(_list))
def __repr__(self):
"""Return string representation of sorted list."""
return "SortedList({0})".format(list(self))
def solve(N, B):
seen = set()
sl = SortedList()
seen.add(B[0])
sl.add(B[0])
BIG = 10 ** 9 + 10
for i, x in enumerate(B):
if i == 0:
continue
# print(sl, "want med", x, "at", i)
added = 0
if x not in seen:
added += 1
sl.add(x)
seen.add(x)
for _ in range(2 - added):
index = sl.bisect_left(x)
if index - i < 0:
sl.add(-BIG)
else:
sl.add(BIG)
# print(" result", sl)
if sl.bisect_left(x) != i:
return "NO"
assert len(sl) == 2 * i + 1
return "YES"
if False:
print(solve(4, [2, 1, 2, 3]))
exit()
print(solve(7, [1, 1, 3, 1, 0, -2, -1]))
exit()
if __name__ == "__main__":
input = io.BytesIO(os.read(0, os.fstat(0).st_size)).readline
TC = int(input())
for tc in range(1, TC + 1):
(N,) = [int(x) for x in input().split()]
B = [int(x) for x in input().split()]
ans = solve(N, B)
print(ans)
``` | output | 1 | 56,632 | 12 | 113,265 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a. | instruction | 0 | 56,633 | 12 | 113,266 |
Tags: data structures, greedy, implementation
Correct Solution:
```
from sys import stdin, stdout
import math
import heapq
R = lambda : stdin.readline().strip()
RL = lambda f=None: list(map(f, R().split(' '))) if f else list(R().split(' '))
output = lambda x: stdout.write(str(x) + '\n')
output_list = lambda x: output(' '.join(map(str, x)))
for tc in range(int(R())):
n = int(R())
a = RL(int)
prev = a[0]
left, right = [-int(1e10)], [int(1e10)]
ans = True
for i in a[1:]:
if i < prev:
if i < left[-1]:
ans=False
break
right.append(prev)
prev=i
if i==left[-1]:
left.pop()
elif i>prev:
if i > right[-1]:
ans=False
break
left.append(prev)
prev=i
if i==right[-1]:
right.pop()
if ans:
print('YES')
continue
print('NO')
``` | output | 1 | 56,633 | 12 | 113,267 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a. | instruction | 0 | 56,634 | 12 | 113,268 |
Tags: data structures, greedy, implementation
Correct Solution:
```
for s in[*open(0)][2::2]:
a=*map(int,s.split()),;l=[-1e9];r=[1e9];o=0
for m,x in zip(a,a[1:]):
if x>m:
if x>r[-1]:o=1;break
l+=m,
if x==r[-1]:r.pop()
elif x<m:
if x<l[-1]:o=1;break
r+=m,
if x==l[-1]:l.pop()
print("yNeOs"[o::2])
``` | output | 1 | 56,634 | 12 | 113,269 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a. | instruction | 0 | 56,635 | 12 | 113,270 |
Tags: data structures, greedy, implementation
Correct Solution:
```
import bisect
def updateQuery(BIT,index,val):
while index < len(BIT):
BIT[index] += val
index += index&-index
def sumQuery(BIT,index):
total = 0
while index > 0:
total += BIT[index]
index -= index&-index
return total
def solve(arr,n,ans):
nums = sorted(list(set(arr)))
BIT = [0]*(len(nums)+1)
present = set()
left = 0
right = 0
for i in arr:
index = bisect.bisect(nums,i-1)
small = left + sumQuery(BIT,index)
big = right + len(present) - sumQuery(BIT,index+1)
#print(i,small,big,left,right)
if abs(small-big) > 2:
ans.append('NO')
return
if small < big:
left += big-small
elif big < small:
right += small-big
else:
left += 1
right += 1
if i not in present:
present.add(i)
updateQuery(BIT,index+1,1)
ans.append('YES')
def main():
t = int(input())
ans = []
for i in range(t):
n = int(input())
arr = list(map(int,input().split()))
solve(arr,n,ans)
print('\n'.join(ans))
main()
``` | output | 1 | 56,635 | 12 | 113,271 |
Provide tags and a correct Python 2 solution for this coding contest problem.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a. | instruction | 0 | 56,636 | 12 | 113,272 |
Tags: data structures, greedy, implementation
Correct Solution:
```
from __future__ import print_function
import sys
import traceback
from bisect import bisect_left, bisect_right, insort
from itertools import chain, repeat, starmap
from math import log
from operator import add, eq, ne, gt, ge, lt, le, iadd
from textwrap import dedent
try:
from collections.abc import Sequence, MutableSequence
except ImportError:
from collections import Sequence, MutableSequence
from functools import wraps
from sys import hexversion
if hexversion < 0x03000000:
from itertools import imap as map # pylint: disable=redefined-builtin
from itertools import izip as zip # pylint: disable=redefined-builtin
try:
from thread import get_ident
except ImportError:
from dummy_thread import get_ident
else:
from functools import reduce
try:
from _thread import get_ident
except ImportError:
from _dummy_thread import get_ident
def recursive_repr(fillvalue='...'):
"Decorator to make a repr function return fillvalue for a recursive call."
# pylint: disable=missing-docstring
# Copied from reprlib in Python 3
# https://hg.python.org/cpython/file/3.6/Lib/reprlib.py
def decorating_function(user_function):
repr_running = set()
@wraps(user_function)
def wrapper(self):
key = id(self), get_ident()
if key in repr_running:
return fillvalue
repr_running.add(key)
try:
result = user_function(self)
finally:
repr_running.discard(key)
return result
return wrapper
return decorating_function
class SortedList(MutableSequence):
DEFAULT_LOAD_FACTOR = 1000
def __init__(self, iterable=None, key=None):
assert key is None
self._len = 0
self._load = self.DEFAULT_LOAD_FACTOR
self._lists = []
self._maxes = []
self._index = []
self._offset = 0
if iterable is not None:
self._update(iterable)
def __new__(cls, iterable=None, key=None):
# pylint: disable=unused-argument
if key is None:
return object.__new__(cls)
else:
if cls is SortedList:
return object.__new__(SortedKeyList)
else:
raise TypeError('inherit SortedKeyList for key argument')
@property
def key(self): # pylint: disable=useless-return
return None
def _reset(self, load):
values = reduce(iadd, self._lists, [])
self._clear()
self._load = load
self._update(values)
def clear(self):
self._len = 0
del self._lists[:]
del self._maxes[:]
del self._index[:]
self._offset = 0
_clear = clear
def add(self, value):
_lists = self._lists
_maxes = self._maxes
if _maxes:
pos = bisect_right(_maxes, value)
if pos == len(_maxes):
pos -= 1
_lists[pos].append(value)
_maxes[pos] = value
else:
insort(_lists[pos], value)
self._expand(pos)
else:
_lists.append([value])
_maxes.append(value)
self._len += 1
def _expand(self, pos):
_load = self._load
_lists = self._lists
_index = self._index
if len(_lists[pos]) > (_load << 1):
_maxes = self._maxes
_lists_pos = _lists[pos]
half = _lists_pos[_load:]
del _lists_pos[_load:]
_maxes[pos] = _lists_pos[-1]
_lists.insert(pos + 1, half)
_maxes.insert(pos + 1, half[-1])
del _index[:]
else:
if _index:
child = self._offset + pos
while child:
_index[child] += 1
child = (child - 1) >> 1
_index[0] += 1
def update(self, iterable):
_lists = self._lists
_maxes = self._maxes
values = sorted(iterable)
if _maxes:
if len(values) * 4 >= self._len:
_lists.append(values)
values = reduce(iadd, _lists, [])
values.sort()
self._clear()
else:
_add = self.add
for val in values:
_add(val)
return
_load = self._load
_lists.extend(values[pos:(pos + _load)]
for pos in range(0, len(values), _load))
_maxes.extend(sublist[-1] for sublist in _lists)
self._len = len(values)
del self._index[:]
_update = update
def __contains__(self, value):
_maxes = self._maxes
if not _maxes:
return False
pos = bisect_left(_maxes, value)
if pos == len(_maxes):
return False
_lists = self._lists
idx = bisect_left(_lists[pos], value)
return _lists[pos][idx] == value
def discard(self, value):
_maxes = self._maxes
if not _maxes:
return
pos = bisect_left(_maxes, value)
if pos == len(_maxes):
return
_lists = self._lists
idx = bisect_left(_lists[pos], value)
if _lists[pos][idx] == value:
self._delete(pos, idx)
def remove(self, value):
_maxes = self._maxes
if not _maxes:
raise ValueError('{0!r} not in list'.format(value))
pos = bisect_left(_maxes, value)
if pos == len(_maxes):
raise ValueError('{0!r} not in list'.format(value))
_lists = self._lists
idx = bisect_left(_lists[pos], value)
if _lists[pos][idx] == value:
self._delete(pos, idx)
else:
raise ValueError('{0!r} not in list'.format(value))
def _delete(self, pos, idx):
_lists = self._lists
_maxes = self._maxes
_index = self._index
_lists_pos = _lists[pos]
del _lists_pos[idx]
self._len -= 1
len_lists_pos = len(_lists_pos)
if len_lists_pos > (self._load >> 1):
_maxes[pos] = _lists_pos[-1]
if _index:
child = self._offset + pos
while child > 0:
_index[child] -= 1
child = (child - 1) >> 1
_index[0] -= 1
elif len(_lists) > 1:
if not pos:
pos += 1
prev = pos - 1
_lists[prev].extend(_lists[pos])
_maxes[prev] = _lists[prev][-1]
del _lists[pos]
del _maxes[pos]
del _index[:]
self._expand(prev)
elif len_lists_pos:
_maxes[pos] = _lists_pos[-1]
else:
del _lists[pos]
del _maxes[pos]
del _index[:]
def _loc(self, pos, idx):
if not pos:
return idx
_index = self._index
if not _index:
self._build_index()
total = 0
# Increment pos to point in the index to len(self._lists[pos]).
pos += self._offset
# Iterate until reaching the root of the index tree at pos = 0.
while pos:
# Right-child nodes are at odd indices. At such indices
# account the total below the left child node.
if not pos & 1:
total += _index[pos - 1]
# Advance pos to the parent node.
pos = (pos - 1) >> 1
return total + idx
def _pos(self, idx):
if idx < 0:
last_len = len(self._lists[-1])
if (-idx) <= last_len:
return len(self._lists) - 1, last_len + idx
idx += self._len
if idx < 0:
raise IndexError('list index out of range')
elif idx >= self._len:
raise IndexError('list index out of range')
if idx < len(self._lists[0]):
return 0, idx
_index = self._index
if not _index:
self._build_index()
pos = 0
child = 1
len_index = len(_index)
while child < len_index:
index_child = _index[child]
if idx < index_child:
pos = child
else:
idx -= index_child
pos = child + 1
child = (pos << 1) + 1
return (pos - self._offset, idx)
def _build_index(self):
row0 = list(map(len, self._lists))
if len(row0) == 1:
self._index[:] = row0
self._offset = 0
return
head = iter(row0)
tail = iter(head)
row1 = list(starmap(add, zip(head, tail)))
if len(row0) & 1:
row1.append(row0[-1])
if len(row1) == 1:
self._index[:] = row1 + row0
self._offset = 1
return
size = 2 ** (int(log(len(row1) - 1, 2)) + 1)
row1.extend(repeat(0, size - len(row1)))
tree = [row0, row1]
while len(tree[-1]) > 1:
head = iter(tree[-1])
tail = iter(head)
row = list(starmap(add, zip(head, tail)))
tree.append(row)
reduce(iadd, reversed(tree), self._index)
self._offset = size * 2 - 1
def __delitem__(self, index):
if isinstance(index, slice):
start, stop, step = index.indices(self._len)
if step == 1 and start < stop:
if start == 0 and stop == self._len:
return self._clear()
elif self._len <= 8 * (stop - start):
values = self._getitem(slice(None, start))
if stop < self._len:
values += self._getitem(slice(stop, None))
self._clear()
return self._update(values)
indices = range(start, stop, step)
# Delete items from greatest index to least so
# that the indices remain valid throughout iteration.
if step > 0:
indices = reversed(indices)
_pos, _delete = self._pos, self._delete
for index in indices:
pos, idx = _pos(index)
_delete(pos, idx)
else:
pos, idx = self._pos(index)
self._delete(pos, idx)
def __getitem__(self, index):
_lists = self._lists
if isinstance(index, slice):
start, stop, step = index.indices(self._len)
if step == 1 and start < stop:
# Whole slice optimization: start to stop slices the whole
# sorted list.
if start == 0 and stop == self._len:
return reduce(iadd, self._lists, [])
start_pos, start_idx = self._pos(start)
start_list = _lists[start_pos]
stop_idx = start_idx + stop - start
# Small slice optimization: start index and stop index are
# within the start list.
if len(start_list) >= stop_idx:
return start_list[start_idx:stop_idx]
if stop == self._len:
stop_pos = len(_lists) - 1
stop_idx = len(_lists[stop_pos])
else:
stop_pos, stop_idx = self._pos(stop)
prefix = _lists[start_pos][start_idx:]
middle = _lists[(start_pos + 1):stop_pos]
result = reduce(iadd, middle, prefix)
result += _lists[stop_pos][:stop_idx]
return result
if step == -1 and start > stop:
result = self._getitem(slice(stop + 1, start + 1))
result.reverse()
return result
# Return a list because a negative step could
# reverse the order of the items and this could
# be the desired behavior.
indices = range(start, stop, step)
return list(self._getitem(index) for index in indices)
else:
if self._len:
if index == 0:
return _lists[0][0]
elif index == -1:
return _lists[-1][-1]
else:
raise IndexError('list index out of range')
if 0 <= index < len(_lists[0]):
return _lists[0][index]
len_last = len(_lists[-1])
if -len_last < index < 0:
return _lists[-1][len_last + index]
pos, idx = self._pos(index)
return _lists[pos][idx]
_getitem = __getitem__
def __setitem__(self, index, value):
message = 'use ``del sl[index]`` and ``sl.add(value)`` instead'
raise NotImplementedError(message)
def __iter__(self):
return chain.from_iterable(self._lists)
def __reversed__(self):
return chain.from_iterable(map(reversed, reversed(self._lists)))
def reverse(self):
raise NotImplementedError('use ``reversed(sl)`` instead')
def islice(self, start=None, stop=None, reverse=False):
_len = self._len
if not _len:
return iter(())
start, stop, _ = slice(start, stop).indices(self._len)
if start >= stop:
return iter(())
_pos = self._pos
min_pos, min_idx = _pos(start)
if stop == _len:
max_pos = len(self._lists) - 1
max_idx = len(self._lists[-1])
else:
max_pos, max_idx = _pos(stop)
return self._islice(min_pos, min_idx, max_pos, max_idx, reverse)
def _islice(self, min_pos, min_idx, max_pos, max_idx, reverse):
_lists = self._lists
if min_pos > max_pos:
return iter(())
if min_pos == max_pos:
if reverse:
indices = reversed(range(min_idx, max_idx))
return map(_lists[min_pos].__getitem__, indices)
indices = range(min_idx, max_idx)
return map(_lists[min_pos].__getitem__, indices)
next_pos = min_pos + 1
if next_pos == max_pos:
if reverse:
min_indices = range(min_idx, len(_lists[min_pos]))
max_indices = range(max_idx)
return chain(
map(_lists[max_pos].__getitem__, reversed(max_indices)),
map(_lists[min_pos].__getitem__, reversed(min_indices)),
)
min_indices = range(min_idx, len(_lists[min_pos]))
max_indices = range(max_idx)
return chain(
map(_lists[min_pos].__getitem__, min_indices),
map(_lists[max_pos].__getitem__, max_indices),
)
if reverse:
min_indices = range(min_idx, len(_lists[min_pos]))
sublist_indices = range(next_pos, max_pos)
sublists = map(_lists.__getitem__, reversed(sublist_indices))
max_indices = range(max_idx)
return chain(
map(_lists[max_pos].__getitem__, reversed(max_indices)),
chain.from_iterable(map(reversed, sublists)),
map(_lists[min_pos].__getitem__, reversed(min_indices)),
)
min_indices = range(min_idx, len(_lists[min_pos]))
sublist_indices = range(next_pos, max_pos)
sublists = map(_lists.__getitem__, sublist_indices)
max_indices = range(max_idx)
return chain(
map(_lists[min_pos].__getitem__, min_indices),
chain.from_iterable(sublists),
map(_lists[max_pos].__getitem__, max_indices),
)
def irange(self, minimum=None, maximum=None, inclusive=(True, True),
reverse=False):
_maxes = self._maxes
if not _maxes:
return iter(())
_lists = self._lists
# Calculate the minimum (pos, idx) pair. By default this location
# will be inclusive in our calculation.
if minimum is None:
min_pos = 0
min_idx = 0
else:
if inclusive[0]:
min_pos = bisect_left(_maxes, minimum)
if min_pos == len(_maxes):
return iter(())
min_idx = bisect_left(_lists[min_pos], minimum)
else:
min_pos = bisect_right(_maxes, minimum)
if min_pos == len(_maxes):
return iter(())
min_idx = bisect_right(_lists[min_pos], minimum)
# Calculate the maximum (pos, idx) pair. By default this location
# will be exclusive in our calculation.
if maximum is None:
max_pos = len(_maxes) - 1
max_idx = len(_lists[max_pos])
else:
if inclusive[1]:
max_pos = bisect_right(_maxes, maximum)
if max_pos == len(_maxes):
max_pos -= 1
max_idx = len(_lists[max_pos])
else:
max_idx = bisect_right(_lists[max_pos], maximum)
else:
max_pos = bisect_left(_maxes, maximum)
if max_pos == len(_maxes):
max_pos -= 1
max_idx = len(_lists[max_pos])
else:
max_idx = bisect_left(_lists[max_pos], maximum)
return self._islice(min_pos, min_idx, max_pos, max_idx, reverse)
def __len__(self):
"""Return the size of the sorted list.
``sl.__len__()`` <==> ``len(sl)``
:return: size of sorted list
"""
return self._len
def bisect_left(self, value):
"""Return an index to insert `value` in the sorted list.
If the `value` is already present, the insertion point will be before
(to the left of) any existing values.
Similar to the `bisect` module in the standard library.
Runtime complexity: `O(log(n))` -- approximate.
>>> sl = SortedList([10, 11, 12, 13, 14])
>>> sl.bisect_left(12)
2
:param value: insertion index of value in sorted list
:return: index
"""
_maxes = self._maxes
if not _maxes:
return 0
pos = bisect_left(_maxes, value)
if pos == len(_maxes):
return self._len
idx = bisect_left(self._lists[pos], value)
return self._loc(pos, idx)
def bisect_right(self, value):
"""Return an index to insert `value` in the sorted list.
Similar to `bisect_left`, but if `value` is already present, the
insertion point will be after (to the right of) any existing values.
Similar to the `bisect` module in the standard library.
Runtime complexity: `O(log(n))` -- approximate.
>>> sl = SortedList([10, 11, 12, 13, 14])
>>> sl.bisect_right(12)
3
:param value: insertion index of value in sorted list
:return: index
"""
_maxes = self._maxes
if not _maxes:
return 0
pos = bisect_right(_maxes, value)
if pos == len(_maxes):
return self._len
idx = bisect_right(self._lists[pos], value)
return self._loc(pos, idx)
bisect = bisect_right
_bisect_right = bisect_right
def count(self, value):
"""Return number of occurrences of `value` in the sorted list.
Runtime complexity: `O(log(n))` -- approximate.
>>> sl = SortedList([1, 2, 2, 3, 3, 3, 4, 4, 4, 4])
>>> sl.count(3)
3
:param value: value to count in sorted list
:return: count
"""
_maxes = self._maxes
if not _maxes:
return 0
pos_left = bisect_left(_maxes, value)
if pos_left == len(_maxes):
return 0
_lists = self._lists
idx_left = bisect_left(_lists[pos_left], value)
pos_right = bisect_right(_maxes, value)
if pos_right == len(_maxes):
return self._len - self._loc(pos_left, idx_left)
idx_right = bisect_right(_lists[pos_right], value)
if pos_left == pos_right:
return idx_right - idx_left
right = self._loc(pos_right, idx_right)
left = self._loc(pos_left, idx_left)
return right - left
def copy(self):
"""Return a shallow copy of the sorted list.
Runtime complexity: `O(n)`
:return: new sorted list
"""
return self.__class__(self)
__copy__ = copy
def append(self, value):
"""Raise not-implemented error.
Implemented to override `MutableSequence.append` which provides an
erroneous default implementation.
:raises NotImplementedError: use ``sl.add(value)`` instead
"""
raise NotImplementedError('use ``sl.add(value)`` instead')
def extend(self, values):
"""Raise not-implemented error.
Implemented to override `MutableSequence.extend` which provides an
erroneous default implementation.
:raises NotImplementedError: use ``sl.update(values)`` instead
"""
raise NotImplementedError('use ``sl.update(values)`` instead')
def insert(self, index, value):
"""Raise not-implemented error.
:raises NotImplementedError: use ``sl.add(value)`` instead
"""
raise NotImplementedError('use ``sl.add(value)`` instead')
def pop(self, index=-1):
"""Remove and return value at `index` in sorted list.
Raise :exc:`IndexError` if the sorted list is empty or index is out of
range.
Negative indices are supported.
Runtime complexity: `O(log(n))` -- approximate.
>>> sl = SortedList('abcde')
>>> sl.pop()
'e'
>>> sl.pop(2)
'c'
>>> sl
SortedList(['a', 'b', 'd'])
:param int index: index of value (default -1)
:return: value
:raises IndexError: if index is out of range
"""
if not self._len:
raise IndexError('pop index out of range')
_lists = self._lists
if index == 0:
val = _lists[0][0]
self._delete(0, 0)
return val
if index == -1:
pos = len(_lists) - 1
loc = len(_lists[pos]) - 1
val = _lists[pos][loc]
self._delete(pos, loc)
return val
if 0 <= index < len(_lists[0]):
val = _lists[0][index]
self._delete(0, index)
return val
len_last = len(_lists[-1])
if -len_last < index < 0:
pos = len(_lists) - 1
loc = len_last + index
val = _lists[pos][loc]
self._delete(pos, loc)
return val
pos, idx = self._pos(index)
val = _lists[pos][idx]
self._delete(pos, idx)
return val
def index(self, value, start=None, stop=None):
"""Return first index of value in sorted list.
Raise ValueError if `value` is not present.
Index must be between `start` and `stop` for the `value` to be
considered present. The default value, None, for `start` and `stop`
indicate the beginning and end of the sorted list.
Negative indices are supported.
Runtime complexity: `O(log(n))` -- approximate.
>>> sl = SortedList('abcde')
>>> sl.index('d')
3
>>> sl.index('z')
Traceback (most recent call last):
...
ValueError: 'z' is not in list
:param value: value in sorted list
:param int start: start index (default None, start of sorted list)
:param int stop: stop index (default None, end of sorted list)
:return: index of value
:raises ValueError: if value is not present
"""
_len = self._len
if not _len:
raise ValueError('{0!r} is not in list'.format(value))
if start is None:
start = 0
if start < 0:
start += _len
if start < 0:
start = 0
if stop is None:
stop = _len
if stop < 0:
stop += _len
if stop > _len:
stop = _len
if stop <= start:
raise ValueError('{0!r} is not in list'.format(value))
_maxes = self._maxes
pos_left = bisect_left(_maxes, value)
if pos_left == len(_maxes):
raise ValueError('{0!r} is not in list'.format(value))
_lists = self._lists
idx_left = bisect_left(_lists[pos_left], value)
if _lists[pos_left][idx_left] != value:
raise ValueError('{0!r} is not in list'.format(value))
stop -= 1
left = self._loc(pos_left, idx_left)
if start <= left:
if left <= stop:
return left
else:
right = self._bisect_right(value) - 1
if start <= right:
return start
raise ValueError('{0!r} is not in list'.format(value))
def __add__(self, other):
"""Return new sorted list containing all values in both sequences.
``sl.__add__(other)`` <==> ``sl + other``
Values in `other` do not need to be in sorted order.
Runtime complexity: `O(n*log(n))`
>>> sl1 = SortedList('bat')
>>> sl2 = SortedList('cat')
>>> sl1 + sl2
SortedList(['a', 'a', 'b', 'c', 't', 't'])
:param other: other iterable
:return: new sorted list
"""
values = reduce(iadd, self._lists, [])
values.extend(other)
return self.__class__(values)
__radd__ = __add__
def __iadd__(self, other):
"""Update sorted list with values from `other`.
``sl.__iadd__(other)`` <==> ``sl += other``
Values in `other` do not need to be in sorted order.
Runtime complexity: `O(k*log(n))` -- approximate.
>>> sl = SortedList('bat')
>>> sl += 'cat'
>>> sl
SortedList(['a', 'a', 'b', 'c', 't', 't'])
:param other: other iterable
:return: existing sorted list
"""
self._update(other)
return self
def __mul__(self, num):
"""Return new sorted list with `num` shallow copies of values.
``sl.__mul__(num)`` <==> ``sl * num``
Runtime complexity: `O(n*log(n))`
>>> sl = SortedList('abc')
>>> sl * 3
SortedList(['a', 'a', 'a', 'b', 'b', 'b', 'c', 'c', 'c'])
:param int num: count of shallow copies
:return: new sorted list
"""
values = reduce(iadd, self._lists, []) * num
return self.__class__(values)
__rmul__ = __mul__
def __imul__(self, num):
"""Update the sorted list with `num` shallow copies of values.
``sl.__imul__(num)`` <==> ``sl *= num``
Runtime complexity: `O(n*log(n))`
>>> sl = SortedList('abc')
>>> sl *= 3
>>> sl
SortedList(['a', 'a', 'a', 'b', 'b', 'b', 'c', 'c', 'c'])
:param int num: count of shallow copies
:return: existing sorted list
"""
values = reduce(iadd, self._lists, []) * num
self._clear()
self._update(values)
return self
def __make_cmp(seq_op, symbol, doc):
"Make comparator method."
def comparer(self, other):
"Compare method for sorted list and sequence."
if not isinstance(other, Sequence):
return NotImplemented
self_len = self._len
len_other = len(other)
if self_len != len_other:
if seq_op is eq:
return False
if seq_op is ne:
return True
for alpha, beta in zip(self, other):
if alpha != beta:
return seq_op(alpha, beta)
return seq_op(self_len, len_other)
seq_op_name = seq_op.__name__
comparer.__name__ = '__{0}__'.format(seq_op_name)
doc_str = """Return true if and only if sorted list is {0} `other`.
``sl.__{1}__(other)`` <==> ``sl {2} other``
Comparisons use lexicographical order as with sequences.
Runtime complexity: `O(n)`
:param other: `other` sequence
:return: true if sorted list is {0} `other`
"""
comparer.__doc__ = dedent(doc_str.format(doc, seq_op_name, symbol))
return comparer
__eq__ = __make_cmp(eq, '==', 'equal to')
__ne__ = __make_cmp(ne, '!=', 'not equal to')
__lt__ = __make_cmp(lt, '<', 'less than')
__gt__ = __make_cmp(gt, '>', 'greater than')
__le__ = __make_cmp(le, '<=', 'less than or equal to')
__ge__ = __make_cmp(ge, '>=', 'greater than or equal to')
__make_cmp = staticmethod(__make_cmp)
def __reduce__(self):
values = reduce(iadd, self._lists, [])
return (type(self), (values,))
@recursive_repr()
def __repr__(self):
"""Return string representation of sorted list.
``sl.__repr__()`` <==> ``repr(sl)``
:return: string representation
"""
return '{0}({1!r})'.format(type(self).__name__, list(self))
def _check(self):
"""Check invariants of sorted list.
Runtime complexity: `O(n)`
"""
try:
assert self._load >= 4
assert len(self._maxes) == len(self._lists)
assert self._len == sum(len(sublist) for sublist in self._lists)
# Check all sublists are sorted.
for sublist in self._lists:
for pos in range(1, len(sublist)):
assert sublist[pos - 1] <= sublist[pos]
# Check beginning/end of sublists are sorted.
for pos in range(1, len(self._lists)):
assert self._lists[pos - 1][-1] <= self._lists[pos][0]
# Check _maxes index is the last value of each sublist.
for pos in range(len(self._maxes)):
assert self._maxes[pos] == self._lists[pos][-1]
# Check sublist lengths are less than double load-factor.
double = self._load << 1
assert all(len(sublist) <= double for sublist in self._lists)
# Check sublist lengths are greater than half load-factor for all
# but the last sublist.
half = self._load >> 1
for pos in range(0, len(self._lists) - 1):
assert len(self._lists[pos]) >= half
if self._index:
assert self._len == self._index[0]
assert len(self._index) == self._offset + len(self._lists)
# Check index leaf nodes equal length of sublists.
for pos in range(len(self._lists)):
leaf = self._index[self._offset + pos]
assert leaf == len(self._lists[pos])
# Check index branch nodes are the sum of their children.
for pos in range(self._offset):
child = (pos << 1) + 1
if child >= len(self._index):
assert self._index[pos] == 0
elif child + 1 == len(self._index):
assert self._index[pos] == self._index[child]
else:
child_sum = self._index[child] + self._index[child + 1]
assert child_sum == self._index[pos]
except:
traceback.print_exc(file=sys.stdout)
print('len', self._len)
print('load', self._load)
print('offset', self._offset)
print('len_index', len(self._index))
print('index', self._index)
print('len_maxes', len(self._maxes))
print('maxes', self._maxes)
print('len_lists', len(self._lists))
print('lists', self._lists)
raise
def identity(value):
"Identity function."
return value
class SortedKeyList(SortedList):
def __init__(self, iterable=None, key=identity):
self._key = key
self._len = 0
self._load = self.DEFAULT_LOAD_FACTOR
self._lists = []
self._keys = []
self._maxes = []
self._index = []
self._offset = 0
if iterable is not None:
self._update(iterable)
def __new__(cls, iterable=None, key=identity):
return object.__new__(cls)
@property
def key(self):
return self._key
def clear(self):
self._len = 0
del self._lists[:]
del self._keys[:]
del self._maxes[:]
del self._index[:]
_clear = clear
def add(self, value):
_lists = self._lists
_keys = self._keys
_maxes = self._maxes
key = self._key(value)
if _maxes:
pos = bisect_right(_maxes, key)
if pos == len(_maxes):
pos -= 1
_lists[pos].append(value)
_keys[pos].append(key)
_maxes[pos] = key
else:
idx = bisect_right(_keys[pos], key)
_lists[pos].insert(idx, value)
_keys[pos].insert(idx, key)
self._expand(pos)
else:
_lists.append([value])
_keys.append([key])
_maxes.append(key)
self._len += 1
def _expand(self, pos):
_lists = self._lists
_keys = self._keys
_index = self._index
if len(_keys[pos]) > (self._load << 1):
_maxes = self._maxes
_load = self._load
_lists_pos = _lists[pos]
_keys_pos = _keys[pos]
half = _lists_pos[_load:]
half_keys = _keys_pos[_load:]
del _lists_pos[_load:]
del _keys_pos[_load:]
_maxes[pos] = _keys_pos[-1]
_lists.insert(pos + 1, half)
_keys.insert(pos + 1, half_keys)
_maxes.insert(pos + 1, half_keys[-1])
del _index[:]
else:
if _index:
child = self._offset + pos
while child:
_index[child] += 1
child = (child - 1) >> 1
_index[0] += 1
def update(self, iterable):
_lists = self._lists
_keys = self._keys
_maxes = self._maxes
values = sorted(iterable, key=self._key)
if _maxes:
if len(values) * 4 >= self._len:
_lists.append(values)
values = reduce(iadd, _lists, [])
values.sort(key=self._key)
self._clear()
else:
_add = self.add
for val in values:
_add(val)
return
_load = self._load
_lists.extend(values[pos:(pos + _load)]
for pos in range(0, len(values), _load))
_keys.extend(list(map(self._key, _list)) for _list in _lists)
_maxes.extend(sublist[-1] for sublist in _keys)
self._len = len(values)
del self._index[:]
_update = update
def __contains__(self, value):
_maxes = self._maxes
if not _maxes:
return False
key = self._key(value)
pos = bisect_left(_maxes, key)
if pos == len(_maxes):
return False
_lists = self._lists
_keys = self._keys
idx = bisect_left(_keys[pos], key)
len_keys = len(_keys)
len_sublist = len(_keys[pos])
while True:
if _keys[pos][idx] != key:
return False
if _lists[pos][idx] == value:
return True
idx += 1
if idx == len_sublist:
pos += 1
if pos == len_keys:
return False
len_sublist = len(_keys[pos])
idx = 0
def discard(self, value):
_maxes = self._maxes
if not _maxes:
return
key = self._key(value)
pos = bisect_left(_maxes, key)
if pos == len(_maxes):
return
_lists = self._lists
_keys = self._keys
idx = bisect_left(_keys[pos], key)
len_keys = len(_keys)
len_sublist = len(_keys[pos])
while True:
if _keys[pos][idx] != key:
return
if _lists[pos][idx] == value:
self._delete(pos, idx)
return
idx += 1
if idx == len_sublist:
pos += 1
if pos == len_keys:
return
len_sublist = len(_keys[pos])
idx = 0
def remove(self, value):
_maxes = self._maxes
if not _maxes:
raise ValueError('{0!r} not in list'.format(value))
key = self._key(value)
pos = bisect_left(_maxes, key)
if pos == len(_maxes):
raise ValueError('{0!r} not in list'.format(value))
_lists = self._lists
_keys = self._keys
idx = bisect_left(_keys[pos], key)
len_keys = len(_keys)
len_sublist = len(_keys[pos])
while True:
if _keys[pos][idx] != key:
raise ValueError('{0!r} not in list'.format(value))
if _lists[pos][idx] == value:
self._delete(pos, idx)
return
idx += 1
if idx == len_sublist:
pos += 1
if pos == len_keys:
raise ValueError('{0!r} not in list'.format(value))
len_sublist = len(_keys[pos])
idx = 0
def _delete(self, pos, idx):
_lists = self._lists
_keys = self._keys
_maxes = self._maxes
_index = self._index
keys_pos = _keys[pos]
lists_pos = _lists[pos]
del keys_pos[idx]
del lists_pos[idx]
self._len -= 1
len_keys_pos = len(keys_pos)
if len_keys_pos > (self._load >> 1):
_maxes[pos] = keys_pos[-1]
if _index:
child = self._offset + pos
while child > 0:
_index[child] -= 1
child = (child - 1) >> 1
_index[0] -= 1
elif len(_keys) > 1:
if not pos:
pos += 1
prev = pos - 1
_keys[prev].extend(_keys[pos])
_lists[prev].extend(_lists[pos])
_maxes[prev] = _keys[prev][-1]
del _lists[pos]
del _keys[pos]
del _maxes[pos]
del _index[:]
self._expand(prev)
elif len_keys_pos:
_maxes[pos] = keys_pos[-1]
else:
del _lists[pos]
del _keys[pos]
del _maxes[pos]
del _index[:]
def irange(self, minimum=None, maximum=None, inclusive=(True, True),
reverse=False):
min_key = self._key(minimum) if minimum is not None else None
max_key = self._key(maximum) if maximum is not None else None
return self._irange_key(
min_key=min_key, max_key=max_key,
inclusive=inclusive, reverse=reverse,
)
def irange_key(self, min_key=None, max_key=None, inclusive=(True, True),
reverse=False):
_maxes = self._maxes
if not _maxes:
return iter(())
_keys = self._keys
# Calculate the minimum (pos, idx) pair. By default this location
# will be inclusive in our calculation.
if min_key is None:
min_pos = 0
min_idx = 0
else:
if inclusive[0]:
min_pos = bisect_left(_maxes, min_key)
if min_pos == len(_maxes):
return iter(())
min_idx = bisect_left(_keys[min_pos], min_key)
else:
min_pos = bisect_right(_maxes, min_key)
if min_pos == len(_maxes):
return iter(())
min_idx = bisect_right(_keys[min_pos], min_key)
if max_key is None:
max_pos = len(_maxes) - 1
max_idx = len(_keys[max_pos])
else:
if inclusive[1]:
max_pos = bisect_right(_maxes, max_key)
if max_pos == len(_maxes):
max_pos -= 1
max_idx = len(_keys[max_pos])
else:
max_idx = bisect_right(_keys[max_pos], max_key)
else:
max_pos = bisect_left(_maxes, max_key)
if max_pos == len(_maxes):
max_pos -= 1
max_idx = len(_keys[max_pos])
else:
max_idx = bisect_left(_keys[max_pos], max_key)
return self._islice(min_pos, min_idx, max_pos, max_idx, reverse)
_irange_key = irange_key
def bisect_left(self, value):
return self._bisect_key_left(self._key(value))
def bisect_right(self, value):
return self._bisect_key_right(self._key(value))
bisect = bisect_right
def bisect_key_left(self, key):
_maxes = self._maxes
if not _maxes:
return 0
pos = bisect_left(_maxes, key)
if pos == len(_maxes):
return self._len
idx = bisect_left(self._keys[pos], key)
return self._loc(pos, idx)
_bisect_key_left = bisect_key_left
def bisect_key_right(self, key):
_maxes = self._maxes
if not _maxes:
return 0
pos = bisect_right(_maxes, key)
if pos == len(_maxes):
return self._len
idx = bisect_right(self._keys[pos], key)
return self._loc(pos, idx)
bisect_key = bisect_key_right
_bisect_key_right = bisect_key_right
def count(self, value):
_maxes = self._maxes
if not _maxes:
return 0
key = self._key(value)
pos = bisect_left(_maxes, key)
if pos == len(_maxes):
return 0
_lists = self._lists
_keys = self._keys
idx = bisect_left(_keys[pos], key)
total = 0
len_keys = len(_keys)
len_sublist = len(_keys[pos])
while True:
if _keys[pos][idx] != key:
return total
if _lists[pos][idx] == value:
total += 1
idx += 1
if idx == len_sublist:
pos += 1
if pos == len_keys:
return total
len_sublist = len(_keys[pos])
idx = 0
def copy(self):
return self.__class__(self, key=self._key)
__copy__ = copy
def index(self, value, start=None, stop=None):
_len = self._len
if not _len:
raise ValueError('{0!r} is not in list'.format(value))
if start is None:
start = 0
if start < 0:
start += _len
if start < 0:
start = 0
if stop is None:
stop = _len
if stop < 0:
stop += _len
if stop > _len:
stop = _len
if stop <= start:
raise ValueError('{0!r} is not in list'.format(value))
_maxes = self._maxes
key = self._key(value)
pos = bisect_left(_maxes, key)
if pos == len(_maxes):
raise ValueError('{0!r} is not in list'.format(value))
stop -= 1
_lists = self._lists
_keys = self._keys
idx = bisect_left(_keys[pos], key)
len_keys = len(_keys)
len_sublist = len(_keys[pos])
while True:
if _keys[pos][idx] != key:
raise ValueError('{0!r} is not in list'.format(value))
if _lists[pos][idx] == value:
loc = self._loc(pos, idx)
if start <= loc <= stop:
return loc
elif loc > stop:
break
idx += 1
if idx == len_sublist:
pos += 1
if pos == len_keys:
raise ValueError('{0!r} is not in list'.format(value))
len_sublist = len(_keys[pos])
idx = 0
raise ValueError('{0!r} is not in list'.format(value))
def __add__(self, other):
values = reduce(iadd, self._lists, [])
values.extend(other)
return self.__class__(values, key=self._key)
__radd__ = __add__
def __mul__(self, num):
values = reduce(iadd, self._lists, []) * num
return self.__class__(values, key=self._key)
def __reduce__(self):
values = reduce(iadd, self._lists, [])
return (type(self), (values, self.key))
@recursive_repr()
def __repr__(self):
type_name = type(self).__name__
return '{0}({1!r}, key={2!r})'.format(type_name, list(self), self._key)
def _check(self):
try:
assert self._load >= 4
assert len(self._maxes) == len(self._lists) == len(self._keys)
assert self._len == sum(len(sublist) for sublist in self._lists)
# Check all sublists are sorted.
for sublist in self._keys:
for pos in range(1, len(sublist)):
assert sublist[pos - 1] <= sublist[pos]
# Check beginning/end of sublists are sorted.
for pos in range(1, len(self._keys)):
assert self._keys[pos - 1][-1] <= self._keys[pos][0]
# Check _keys matches _key mapped to _lists.
for val_sublist, key_sublist in zip(self._lists, self._keys):
assert len(val_sublist) == len(key_sublist)
for val, key in zip(val_sublist, key_sublist):
assert self._key(val) == key
# Check _maxes index is the last value of each sublist.
for pos in range(len(self._maxes)):
assert self._maxes[pos] == self._keys[pos][-1]
# Check sublist lengths are less than double load-factor.
double = self._load << 1
assert all(len(sublist) <= double for sublist in self._lists)
# Check sublist lengths are greater than half load-factor for all
# but the last sublist.
half = self._load >> 1
for pos in range(0, len(self._lists) - 1):
assert len(self._lists[pos]) >= half
if self._index:
assert self._len == self._index[0]
assert len(self._index) == self._offset + len(self._lists)
# Check index leaf nodes equal length of sublists.
for pos in range(len(self._lists)):
leaf = self._index[self._offset + pos]
assert leaf == len(self._lists[pos])
# Check index branch nodes are the sum of their children.
for pos in range(self._offset):
child = (pos << 1) + 1
if child >= len(self._index):
assert self._index[pos] == 0
elif child + 1 == len(self._index):
assert self._index[pos] == self._index[child]
else:
child_sum = self._index[child] + self._index[child + 1]
assert child_sum == self._index[pos]
except:
traceback.print_exc(file=sys.stdout)
for _ in range(input()):
n=input()
a=[int(i) for i in raw_input().split()]
s=SortedList()
s.add(a[0])
for i in range(1,n):
pos=s.bisect_left(a[i])
if pos>0 and s[pos-1]==a[i-1]:
s.add(a[i])
continue
if a[i]==a[i-1]:
continue
pos=s.bisect_left(a[i]+1)
if pos!=len(s) and s[pos]==a[i-1]:
s.add(a[i])
continue
print("NO")
break
else:
print("YES")
``` | output | 1 | 56,636 | 12 | 113,273 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a.
Submitted Solution:
```
from collections import deque
from sys import stdin, gettrace
if gettrace():
def inputi():
return input()
else:
def input():
return next(stdin)[:-1]
def inputi():
return stdin.buffer.readline()
def solve():
n = int(input())
bb = [int(a) for a in input().split()]
pn = {}
pn[bb[0]] = [None, None]
c = bb[0]
for b in bb[1:]:
if b == c:
continue
if b < c:
if pn[c][0] == None or b > pn[c][0]:
if pn[c][0] != None:
pn[pn[c][0]][1] = b
pn[b] = [pn[c][0],c]
pn[c][0] = b
elif b < pn[c][0]:
print('NO')
return
else:
if pn[c][1] == None or b < pn[c][1]:
if pn[c][1] != None:
pn[pn[c][1]][0] = b
pn[b] = [c, pn[c][1]]
pn[c][1] = b
elif b > pn[c][1]:
print('NO')
return
c = b
print('YES')
def main():
t = int(input())
for _ in range(t):
solve()
if __name__ == "__main__":
main()
``` | instruction | 0 | 56,637 | 12 | 113,274 |
Yes | output | 1 | 56,637 | 12 | 113,275 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a.
Submitted Solution:
```
import sys
import math
import heapq
import bisect
from collections import Counter
from collections import defaultdict
from io import BytesIO, IOBase
import string
import fractions
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
import os
self.os = os
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
self.BUFSIZE = 8192
def read(self):
while True:
b = self.os.read(self._fd, max(self.os.fstat(self._fd).st_size, self.BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = self.os.read(self._fd, max(self.os.fstat(self._fd).st_size, self.BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
self.os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
def get_int():
return int(input())
def get_ints():
return list(map(int, input().split(' ')))
def get_int_grid(n):
return [get_ints() for _ in range(n)]
def get_str():
return input().strip()
def get_strs():
return get_str().split(' ')
def flat_list(arr):
return [item for subarr in arr for item in subarr]
def yes_no(b):
if b:
return "YES"
else:
return "NO"
def binary_search(good, left, right, delta=1, right_true=False):
"""
Performs binary search
----------
Parameters
----------
:param good: Function used to perform the binary search
:param left: Starting value of left limit
:param right: Starting value of the right limit
:param delta: Margin of error, defaults value of 1 for integer binary search
:param right_true: Boolean, for whether the right limit is the true invariant
:return: Returns the most extremal value interval [left, right] which is good function evaluates to True,
alternatively returns False if no such value found
"""
limits = [left, right]
while limits[1] - limits[0] > delta:
if delta == 1:
mid = sum(limits) // 2
else:
mid = sum(limits) / 2
if good(mid):
limits[int(right_true)] = mid
else:
limits[int(~right_true)] = mid
if good(limits[int(right_true)]):
return limits[int(right_true)]
else:
return False
def prefix_sums(a):
p = [0]
for x in a:
p.append(p[-1] + x)
return p
def solve_a():
n = get_int()
a = get_ints()
g = 0
for x in a:
g = math.gcd(g, x)
if min(a) < 0:
return False
else:
return list(range(max(a) + 1))
def solve_b():
n = get_int()
s = get_str()
alpha = list(string.ascii_lowercase)
for x in alpha:
if x not in s:
return x
for i in range(26):
for j in range(26):
x, y = alpha[i], alpha[j]
if x + y not in s:
return x + y
for i in range(26):
for j in range(26):
for k in range(26):
x, y, z = alpha[i], alpha[j], alpha[k]
if x + y + z not in s:
return x + y + z
return False
def solve_c():
ratios = defaultdict(int)
n = get_int()
s = list(get_str())
d, k = 0, 0
ans = []
for x in s:
if x == 'D':
d += 1
else:
k += 1
g = math.gcd(d, k)
f = d // g, k // g
ratios[f] += 1
ans.append(ratios[f])
return ans
class SortedList:
def __init__(self, iterable=[], _load=200):
"""Initialize sorted list instance."""
values = sorted(iterable)
self._len = _len = len(values)
self._load = _load
self._lists = _lists = [values[i:i + _load] for i in range(0, _len, _load)]
self._list_lens = [len(_list) for _list in _lists]
self._mins = [_list[0] for _list in _lists]
self._fen_tree = []
self._rebuild = True
def _fen_build(self):
"""Build a fenwick tree instance."""
self._fen_tree[:] = self._list_lens
_fen_tree = self._fen_tree
for i in range(len(_fen_tree)):
if i | i + 1 < len(_fen_tree):
_fen_tree[i | i + 1] += _fen_tree[i]
self._rebuild = False
def _fen_update(self, index, value):
"""Update `fen_tree[index] += value`."""
if not self._rebuild:
_fen_tree = self._fen_tree
while index < len(_fen_tree):
_fen_tree[index] += value
index |= index + 1
def _fen_query(self, end):
"""Return `sum(_fen_tree[:end])`."""
if self._rebuild:
self._fen_build()
_fen_tree = self._fen_tree
x = 0
while end:
x += _fen_tree[end - 1]
end &= end - 1
return x
def _fen_findkth(self, k):
"""Return a pair of (the largest `idx` such that `sum(_fen_tree[:idx]) <= k`, `k - sum(_fen_tree[:idx])`)."""
_list_lens = self._list_lens
if k < _list_lens[0]:
return 0, k
if k >= self._len - _list_lens[-1]:
return len(_list_lens) - 1, k + _list_lens[-1] - self._len
if self._rebuild:
self._fen_build()
_fen_tree = self._fen_tree
idx = -1
for d in reversed(range(len(_fen_tree).bit_length())):
right_idx = idx + (1 << d)
if right_idx < len(_fen_tree) and k >= _fen_tree[right_idx]:
idx = right_idx
k -= _fen_tree[idx]
return idx + 1, k
def _delete(self, pos, idx):
"""Delete value at the given `(pos, idx)`."""
_lists = self._lists
_mins = self._mins
_list_lens = self._list_lens
self._len -= 1
self._fen_update(pos, -1)
del _lists[pos][idx]
_list_lens[pos] -= 1
if _list_lens[pos]:
_mins[pos] = _lists[pos][0]
else:
del _lists[pos]
del _list_lens[pos]
del _mins[pos]
self._rebuild = True
def _loc_left(self, value):
"""Return an index pair that corresponds to the first position of `value` in the sorted list."""
if not self._len:
return 0, 0
_lists = self._lists
_mins = self._mins
lo, pos = -1, len(_lists) - 1
while lo + 1 < pos:
mi = (lo + pos) >> 1
if value <= _mins[mi]:
pos = mi
else:
lo = mi
if pos and value <= _lists[pos - 1][-1]:
pos -= 1
_list = _lists[pos]
lo, idx = -1, len(_list)
while lo + 1 < idx:
mi = (lo + idx) >> 1
if value <= _list[mi]:
idx = mi
else:
lo = mi
return pos, idx
def _loc_right(self, value):
"""Return an index pair that corresponds to the last position of `value` in the sorted list."""
if not self._len:
return 0, 0
_lists = self._lists
_mins = self._mins
pos, hi = 0, len(_lists)
while pos + 1 < hi:
mi = (pos + hi) >> 1
if value < _mins[mi]:
hi = mi
else:
pos = mi
_list = _lists[pos]
lo, idx = -1, len(_list)
while lo + 1 < idx:
mi = (lo + idx) >> 1
if value < _list[mi]:
idx = mi
else:
lo = mi
return pos, idx
def add(self, value):
"""Add `value` to sorted list."""
_load = self._load
_lists = self._lists
_mins = self._mins
_list_lens = self._list_lens
self._len += 1
if _lists:
pos, idx = self._loc_right(value)
self._fen_update(pos, 1)
_list = _lists[pos]
_list.insert(idx, value)
_list_lens[pos] += 1
_mins[pos] = _list[0]
if _load + _load < len(_list):
_lists.insert(pos + 1, _list[_load:])
_list_lens.insert(pos + 1, len(_list) - _load)
_mins.insert(pos + 1, _list[_load])
_list_lens[pos] = _load
del _list[_load:]
self._rebuild = True
else:
_lists.append([value])
_mins.append(value)
_list_lens.append(1)
self._rebuild = True
def discard(self, value):
"""Remove `value` from sorted list if it is a member."""
_lists = self._lists
if _lists:
pos, idx = self._loc_right(value)
if idx and _lists[pos][idx - 1] == value:
self._delete(pos, idx - 1)
def remove(self, value):
"""Remove `value` from sorted list; `value` must be a member."""
_len = self._len
self.discard(value)
if _len == self._len:
raise ValueError('{0!r} not in list'.format(value))
def pop(self, index=-1):
"""Remove and return value at `index` in sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
value = self._lists[pos][idx]
self._delete(pos, idx)
return value
def bisect_left(self, value):
"""Return the first index to insert `value` in the sorted list."""
pos, idx = self._loc_left(value)
return self._fen_query(pos) + idx
def bisect_right(self, value):
"""Return the last index to insert `value` in the sorted list."""
pos, idx = self._loc_right(value)
return self._fen_query(pos) + idx
def count(self, value):
"""Return number of occurrences of `value` in the sorted list."""
return self.bisect_right(value) - self.bisect_left(value)
def __len__(self):
"""Return the size of the sorted list."""
return self._len
def __getitem__(self, index):
"""Lookup value at `index` in sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
return self._lists[pos][idx]
def __delitem__(self, index):
"""Remove value at `index` from sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
self._delete(pos, idx)
def __contains__(self, value):
"""Return true if `value` is an element of the sorted list."""
_lists = self._lists
if _lists:
pos, idx = self._loc_left(value)
return idx < len(_lists[pos]) and _lists[pos][idx] == value
return False
def __iter__(self):
"""Return an iterator over the sorted list."""
return (value for _list in self._lists for value in _list)
def __reversed__(self):
"""Return a reverse iterator over the sorted list."""
return (value for _list in reversed(self._lists) for value in reversed(_list))
def __repr__(self):
"""Return string representation of sorted list."""
return 'SortedList({0})'.format(list(self))
def solve_d():
n = get_int()
b = get_ints()
S = SortedList()
S.add(float('inf'))
S.add(-float('inf'))
m = 0
for x in b:
l = S.bisect_left(m) - 1
r = S.bisect_right(m)
if S[l] <= x <= S[r]:
m = x
S.add(x)
else:
return False
return True
t = get_int()
for _ in range(t):
print(yes_no(solve_d()))
``` | instruction | 0 | 56,638 | 12 | 113,276 |
Yes | output | 1 | 56,638 | 12 | 113,277 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a.
Submitted Solution:
```
import heapq
from collections import Counter
from math import gcd
import io, os
fastin = io.BytesIO(os.read(0,os.fstat(0).st_size)).readline
def solve():
n = int(fastin())
b = list(map(int, fastin().split()))
if n < 3:
print("YES")
return
left, right = [], []
pivot = b[0]
for j in range(1, n):
if pivot == b[j]:
continue
if (right and b[j] > right[0]) or (left and b[j] < -left[0]):
print("NO")
return
if b[j] > pivot:
heapq.heappush(left, -pivot)
pivot = b[j]
if right and b[j] == right[0]:
heapq.heappop(right)
elif b[j] < pivot:
heapq.heappush(right, pivot)
pivot = b[j]
if left and b[j] == -left[0]:
heapq.heappop(left)
print("YES")
return
testcase = int(fastin())
tn = 1
while tn <= testcase:
solve()
tn += 1
``` | instruction | 0 | 56,639 | 12 | 113,278 |
Yes | output | 1 | 56,639 | 12 | 113,279 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a.
Submitted Solution:
```
from __future__ import print_function
import sys
input = sys.stdin.readline
import traceback
from bisect import bisect_left, bisect_right, insort
from itertools import chain, repeat, starmap
from math import log
from operator import add, eq, ne, gt, ge, lt, le, iadd
from textwrap import dedent
try:
from collections.abc import Sequence, MutableSequence
except ImportError:
from collections import Sequence, MutableSequence
from functools import wraps
from sys import hexversion
if hexversion < 0x03000000:
from itertools import imap as map # pylint: disable=redefined-builtin
from itertools import izip as zip # pylint: disable=redefined-builtin
try:
from thread import get_ident
except ImportError:
from dummy_thread import get_ident
else:
from functools import reduce
try:
from _thread import get_ident
except ImportError:
from _dummy_thread import get_ident
def recursive_repr(fillvalue='...'):
def decorating_function(user_function):
repr_running = set()
@wraps(user_function)
def wrapper(self):
key = id(self), get_ident()
if key in repr_running:
return fillvalue
repr_running.add(key)
try:
result = user_function(self)
finally:
repr_running.discard(key)
return result
return wrapper
return decorating_function
class SortedList(MutableSequence):
DEFAULT_LOAD_FACTOR = 1000
def __init__(self, iterable=None, key=None):
assert key is None
self._len = 0
self._load = self.DEFAULT_LOAD_FACTOR
self._lists = []
self._maxes = []
self._index = []
self._offset = 0
if iterable is not None:
self._update(iterable)
def __new__(cls, iterable=None, key=None):
if key is None:
return object.__new__(cls)
else:
if cls is SortedList:
return object.__new__(SortedKeyList)
else:
raise TypeError('inherit SortedKeyList for key argument')
@property
def key(self): # pylint: disable=useless-return
"""Function used to extract comparison key from values.
Sorted list compares values directly so the key function is none.
"""
return None
def _reset(self, load):
values = reduce(iadd, self._lists, [])
self._clear()
self._load = load
self._update(values)
def clear(self):
self._len = 0
del self._lists[:]
del self._maxes[:]
del self._index[:]
self._offset = 0
_clear = clear
def add(self, value):
_lists = self._lists
_maxes = self._maxes
if _maxes:
pos = bisect_right(_maxes, value)
if pos == len(_maxes):
pos -= 1
_lists[pos].append(value)
_maxes[pos] = value
else:
insort(_lists[pos], value)
self._expand(pos)
else:
_lists.append([value])
_maxes.append(value)
self._len += 1
def _expand(self, pos):
_load = self._load
_lists = self._lists
_index = self._index
if len(_lists[pos]) > (_load << 1):
_maxes = self._maxes
_lists_pos = _lists[pos]
half = _lists_pos[_load:]
del _lists_pos[_load:]
_maxes[pos] = _lists_pos[-1]
_lists.insert(pos + 1, half)
_maxes.insert(pos + 1, half[-1])
del _index[:]
else:
if _index:
child = self._offset + pos
while child:
_index[child] += 1
child = (child - 1) >> 1
_index[0] += 1
def update(self, iterable):
_lists = self._lists
_maxes = self._maxes
values = sorted(iterable)
if _maxes:
if len(values) * 4 >= self._len:
values.extend(chain.from_iterable(_lists))
values.sort()
self._clear()
else:
_add = self.add
for val in values:
_add(val)
return
_load = self._load
_lists.extend(values[pos:(pos + _load)]
for pos in range(0, len(values), _load))
_maxes.extend(sublist[-1] for sublist in _lists)
self._len = len(values)
del self._index[:]
_update = update
def __contains__(self, value):
_maxes = self._maxes
if not _maxes:
return False
pos = bisect_left(_maxes, value)
if pos == len(_maxes):
return False
_lists = self._lists
idx = bisect_left(_lists[pos], value)
return _lists[pos][idx] == value
def discard(self, value):
_maxes = self._maxes
if not _maxes:
return
pos = bisect_left(_maxes, value)
if pos == len(_maxes):
return
_lists = self._lists
idx = bisect_left(_lists[pos], value)
if _lists[pos][idx] == value:
self._delete(pos, idx)
def remove(self, value):
_maxes = self._maxes
if not _maxes:
raise ValueError('{0!r} not in list'.format(value))
pos = bisect_left(_maxes, value)
if pos == len(_maxes):
raise ValueError('{0!r} not in list'.format(value))
_lists = self._lists
idx = bisect_left(_lists[pos], value)
if _lists[pos][idx] == value:
self._delete(pos, idx)
else:
raise ValueError('{0!r} not in list'.format(value))
def _delete(self, pos, idx):
_lists = self._lists
_maxes = self._maxes
_index = self._index
_lists_pos = _lists[pos]
del _lists_pos[idx]
self._len -= 1
len_lists_pos = len(_lists_pos)
if len_lists_pos > (self._load >> 1):
_maxes[pos] = _lists_pos[-1]
if _index:
child = self._offset + pos
while child > 0:
_index[child] -= 1
child = (child - 1) >> 1
_index[0] -= 1
elif len(_lists) > 1:
if not pos:
pos += 1
prev = pos - 1
_lists[prev].extend(_lists[pos])
_maxes[prev] = _lists[prev][-1]
del _lists[pos]
del _maxes[pos]
del _index[:]
self._expand(prev)
elif len_lists_pos:
_maxes[pos] = _lists_pos[-1]
else:
del _lists[pos]
del _maxes[pos]
del _index[:]
def _loc(self, pos, idx):
if not pos:
return idx
_index = self._index
if not _index:
self._build_index()
total = 0
# Increment pos to point in the index to len(self._lists[pos]).
pos += self._offset
# Iterate until reaching the root of the index tree at pos = 0.
while pos:
# Right-child nodes are at odd indices. At such indices
# account the total below the left child node.
if not pos & 1:
total += _index[pos - 1]
# Advance pos to the parent node.
pos = (pos - 1) >> 1
return total + idx
def _pos(self, idx):
if idx < 0:
last_len = len(self._lists[-1])
if (-idx) <= last_len:
return len(self._lists) - 1, last_len + idx
idx += self._len
if idx < 0:
raise IndexError('list index out of range')
elif idx >= self._len:
raise IndexError('list index out of range')
if idx < len(self._lists[0]):
return 0, idx
_index = self._index
if not _index:
self._build_index()
pos = 0
child = 1
len_index = len(_index)
while child < len_index:
index_child = _index[child]
if idx < index_child:
pos = child
else:
idx -= index_child
pos = child + 1
child = (pos << 1) + 1
return (pos - self._offset, idx)
def _build_index(self):
row0 = list(map(len, self._lists))
if len(row0) == 1:
self._index[:] = row0
self._offset = 0
return
head = iter(row0)
tail = iter(head)
row1 = list(starmap(add, zip(head, tail)))
if len(row0) & 1:
row1.append(row0[-1])
if len(row1) == 1:
self._index[:] = row1 + row0
self._offset = 1
return
size = 2 ** (int(log(len(row1) - 1, 2)) + 1)
row1.extend(repeat(0, size - len(row1)))
tree = [row0, row1]
while len(tree[-1]) > 1:
head = iter(tree[-1])
tail = iter(head)
row = list(starmap(add, zip(head, tail)))
tree.append(row)
reduce(iadd, reversed(tree), self._index)
self._offset = size * 2 - 1
def __delitem__(self, index):
if isinstance(index, slice):
start, stop, step = index.indices(self._len)
if step == 1 and start < stop:
if start == 0 and stop == self._len:
return self._clear()
elif self._len <= 8 * (stop - start):
values = self._getitem(slice(None, start))
if stop < self._len:
values += self._getitem(slice(stop, None))
self._clear()
return self._update(values)
indices = range(start, stop, step)
# Delete items from greatest index to least so
# that the indices remain valid throughout iteration.
if step > 0:
indices = reversed(indices)
_pos, _delete = self._pos, self._delete
for index in indices:
pos, idx = _pos(index)
_delete(pos, idx)
else:
pos, idx = self._pos(index)
self._delete(pos, idx)
def __getitem__(self, index):
_lists = self._lists
if isinstance(index, slice):
start, stop, step = index.indices(self._len)
if step == 1 and start < stop:
# Whole slice optimization: start to stop slices the whole
# sorted list.
if start == 0 and stop == self._len:
return reduce(iadd, self._lists, [])
start_pos, start_idx = self._pos(start)
start_list = _lists[start_pos]
stop_idx = start_idx + stop - start
# Small slice optimization: start index and stop index are
# within the start list.
if len(start_list) >= stop_idx:
return start_list[start_idx:stop_idx]
if stop == self._len:
stop_pos = len(_lists) - 1
stop_idx = len(_lists[stop_pos])
else:
stop_pos, stop_idx = self._pos(stop)
prefix = _lists[start_pos][start_idx:]
middle = _lists[(start_pos + 1):stop_pos]
result = reduce(iadd, middle, prefix)
result += _lists[stop_pos][:stop_idx]
return result
if step == -1 and start > stop:
result = self._getitem(slice(stop + 1, start + 1))
result.reverse()
return result
indices = range(start, stop, step)
return list(self._getitem(index) for index in indices)
else:
if self._len:
if index == 0:
return _lists[0][0]
elif index == -1:
return _lists[-1][-1]
else:
raise IndexError('list index out of range')
if 0 <= index < len(_lists[0]):
return _lists[0][index]
len_last = len(_lists[-1])
if -len_last < index < 0:
return _lists[-1][len_last + index]
pos, idx = self._pos(index)
return _lists[pos][idx]
_getitem = __getitem__
def __setitem__(self, index, value):
message = 'use ``del sl[index]`` and ``sl.add(value)`` instead'
raise NotImplementedError(message)
def __iter__(self):
return chain.from_iterable(self._lists)
def __reversed__(self):
return chain.from_iterable(map(reversed, reversed(self._lists)))
def reverse(self):
raise NotImplementedError('use ``reversed(sl)`` instead')
def islice(self, start=None, stop=None, reverse=False):
_len = self._len
if not _len:
return iter(())
start, stop, _ = slice(start, stop).indices(self._len)
if start >= stop:
return iter(())
_pos = self._pos
min_pos, min_idx = _pos(start)
if stop == _len:
max_pos = len(self._lists) - 1
max_idx = len(self._lists[-1])
else:
max_pos, max_idx = _pos(stop)
return self._islice(min_pos, min_idx, max_pos, max_idx, reverse)
def _islice(self, min_pos, min_idx, max_pos, max_idx, reverse):
_lists = self._lists
if min_pos > max_pos:
return iter(())
if min_pos == max_pos:
if reverse:
indices = reversed(range(min_idx, max_idx))
return map(_lists[min_pos].__getitem__, indices)
indices = range(min_idx, max_idx)
return map(_lists[min_pos].__getitem__, indices)
next_pos = min_pos + 1
if next_pos == max_pos:
if reverse:
min_indices = range(min_idx, len(_lists[min_pos]))
max_indices = range(max_idx)
return chain(
map(_lists[max_pos].__getitem__, reversed(max_indices)),
map(_lists[min_pos].__getitem__, reversed(min_indices)),
)
min_indices = range(min_idx, len(_lists[min_pos]))
max_indices = range(max_idx)
return chain(
map(_lists[min_pos].__getitem__, min_indices),
map(_lists[max_pos].__getitem__, max_indices),
)
if reverse:
min_indices = range(min_idx, len(_lists[min_pos]))
sublist_indices = range(next_pos, max_pos)
sublists = map(_lists.__getitem__, reversed(sublist_indices))
max_indices = range(max_idx)
return chain(
map(_lists[max_pos].__getitem__, reversed(max_indices)),
chain.from_iterable(map(reversed, sublists)),
map(_lists[min_pos].__getitem__, reversed(min_indices)),
)
min_indices = range(min_idx, len(_lists[min_pos]))
sublist_indices = range(next_pos, max_pos)
sublists = map(_lists.__getitem__, sublist_indices)
max_indices = range(max_idx)
return chain(
map(_lists[min_pos].__getitem__, min_indices),
chain.from_iterable(sublists),
map(_lists[max_pos].__getitem__, max_indices),
)
def irange(self, minimum=None, maximum=None, inclusive=(True, True),
reverse=False):
_maxes = self._maxes
if not _maxes:
return iter(())
_lists = self._lists
if minimum is None:
min_pos = 0
min_idx = 0
else:
if inclusive[0]:
min_pos = bisect_left(_maxes, minimum)
if min_pos == len(_maxes):
return iter(())
min_idx = bisect_left(_lists[min_pos], minimum)
else:
min_pos = bisect_right(_maxes, minimum)
if min_pos == len(_maxes):
return iter(())
min_idx = bisect_right(_lists[min_pos], minimum)
if maximum is None:
max_pos = len(_maxes) - 1
max_idx = len(_lists[max_pos])
else:
if inclusive[1]:
max_pos = bisect_right(_maxes, maximum)
if max_pos == len(_maxes):
max_pos -= 1
max_idx = len(_lists[max_pos])
else:
max_idx = bisect_right(_lists[max_pos], maximum)
else:
max_pos = bisect_left(_maxes, maximum)
if max_pos == len(_maxes):
max_pos -= 1
max_idx = len(_lists[max_pos])
else:
max_idx = bisect_left(_lists[max_pos], maximum)
return self._islice(min_pos, min_idx, max_pos, max_idx, reverse)
def __len__(self):
return self._len
def bisect_left(self, value):
_maxes = self._maxes
if not _maxes:
return 0
pos = bisect_left(_maxes, value)
if pos == len(_maxes):
return self._len
idx = bisect_left(self._lists[pos], value)
return self._loc(pos, idx)
def bisect_right(self, value):
_maxes = self._maxes
if not _maxes:
return 0
pos = bisect_right(_maxes, value)
if pos == len(_maxes):
return self._len
idx = bisect_right(self._lists[pos], value)
return self._loc(pos, idx)
bisect = bisect_right
_bisect_right = bisect_right
def count(self, value):
_maxes = self._maxes
if not _maxes:
return 0
pos_left = bisect_left(_maxes, value)
if pos_left == len(_maxes):
return 0
_lists = self._lists
idx_left = bisect_left(_lists[pos_left], value)
pos_right = bisect_right(_maxes, value)
if pos_right == len(_maxes):
return self._len - self._loc(pos_left, idx_left)
idx_right = bisect_right(_lists[pos_right], value)
if pos_left == pos_right:
return idx_right - idx_left
right = self._loc(pos_right, idx_right)
left = self._loc(pos_left, idx_left)
return right - left
def copy(self):
return self.__class__(self)
__copy__ = copy
def append(self, value):
raise NotImplementedError('use ``sl.add(value)`` instead')
def extend(self, values):
raise NotImplementedError('use ``sl.update(values)`` instead')
def insert(self, index, value):
raise NotImplementedError('use ``sl.add(value)`` instead')
def pop(self, index=-1):
if not self._len:
raise IndexError('pop index out of range')
_lists = self._lists
if index == 0:
val = _lists[0][0]
self._delete(0, 0)
return val
if index == -1:
pos = len(_lists) - 1
loc = len(_lists[pos]) - 1
val = _lists[pos][loc]
self._delete(pos, loc)
return val
if 0 <= index < len(_lists[0]):
val = _lists[0][index]
self._delete(0, index)
return val
len_last = len(_lists[-1])
if -len_last < index < 0:
pos = len(_lists) - 1
loc = len_last + index
val = _lists[pos][loc]
self._delete(pos, loc)
return val
pos, idx = self._pos(index)
val = _lists[pos][idx]
self._delete(pos, idx)
return val
def index(self, value, start=None, stop=None):
_len = self._len
if not _len:
raise ValueError('{0!r} is not in list'.format(value))
if start is None:
start = 0
if start < 0:
start += _len
if start < 0:
start = 0
if stop is None:
stop = _len
if stop < 0:
stop += _len
if stop > _len:
stop = _len
if stop <= start:
raise ValueError('{0!r} is not in list'.format(value))
_maxes = self._maxes
pos_left = bisect_left(_maxes, value)
if pos_left == len(_maxes):
raise ValueError('{0!r} is not in list'.format(value))
_lists = self._lists
idx_left = bisect_left(_lists[pos_left], value)
if _lists[pos_left][idx_left] != value:
raise ValueError('{0!r} is not in list'.format(value))
stop -= 1
left = self._loc(pos_left, idx_left)
if start <= left:
if left <= stop:
return left
else:
right = self._bisect_right(value) - 1
if start <= right:
return start
raise ValueError('{0!r} is not in list'.format(value))
def __add__(self, other):
values = reduce(iadd, self._lists, [])
values.extend(other)
return self.__class__(values)
__radd__ = __add__
def __iadd__(self, other):
self._update(other)
return self
def __mul__(self, num):
values = reduce(iadd, self._lists, []) * num
return self.__class__(values)
__rmul__ = __mul__
def __imul__(self, num):
values = reduce(iadd, self._lists, []) * num
self._clear()
self._update(values)
return self
def __make_cmp(seq_op, symbol, doc):
"Make comparator method."
def comparer(self, other):
"Compare method for sorted list and sequence."
if not isinstance(other, Sequence):
return NotImplemented
self_len = self._len
len_other = len(other)
if self_len != len_other:
if seq_op is eq:
return False
if seq_op is ne:
return True
for alpha, beta in zip(self, other):
if alpha != beta:
return seq_op(alpha, beta)
return seq_op(self_len, len_other)
seq_op_name = seq_op.__name__
comparer.__name__ = '__{0}__'.format(seq_op_name)
doc_str = """Return true if and only if sorted list is {0} `other`.
``sl.__{1}__(other)`` <==> ``sl {2} other``
Comparisons use lexicographical order as with sequences.
Runtime complexity: `O(n)`
:param other: `other` sequence
:return: true if sorted list is {0} `other`
"""
comparer.__doc__ = dedent(doc_str.format(doc, seq_op_name, symbol))
return comparer
__eq__ = __make_cmp(eq, '==', 'equal to')
__ne__ = __make_cmp(ne, '!=', 'not equal to')
__lt__ = __make_cmp(lt, '<', 'less than')
__gt__ = __make_cmp(gt, '>', 'greater than')
__le__ = __make_cmp(le, '<=', 'less than or equal to')
__ge__ = __make_cmp(ge, '>=', 'greater than or equal to')
__make_cmp = staticmethod(__make_cmp)
def __reduce__(self):
values = reduce(iadd, self._lists, [])
return (type(self), (values,))
@recursive_repr()
def __repr__(self):
return '{0}({1!r})'.format(type(self).__name__, list(self))
def _check(self):
"""Check invariants of sorted list.
Runtime complexity: `O(n)`
"""
try:
assert self._load >= 4
assert len(self._maxes) == len(self._lists)
assert self._len == sum(len(sublist) for sublist in self._lists)
# Check all sublists are sorted.
for sublist in self._lists:
for pos in range(1, len(sublist)):
assert sublist[pos - 1] <= sublist[pos]
# Check beginning/end of sublists are sorted.
for pos in range(1, len(self._lists)):
assert self._lists[pos - 1][-1] <= self._lists[pos][0]
# Check _maxes index is the last value of each sublist.
for pos in range(len(self._maxes)):
assert self._maxes[pos] == self._lists[pos][-1]
# Check sublist lengths are less than double load-factor.
double = self._load << 1
assert all(len(sublist) <= double for sublist in self._lists)
half = self._load >> 1
for pos in range(0, len(self._lists) - 1):
assert len(self._lists[pos]) >= half
if self._index:
assert self._len == self._index[0]
assert len(self._index) == self._offset + len(self._lists)
# Check index leaf nodes equal length of sublists.
for pos in range(len(self._lists)):
leaf = self._index[self._offset + pos]
assert leaf == len(self._lists[pos])
# Check index branch nodes are the sum of their children.
for pos in range(self._offset):
child = (pos << 1) + 1
if child >= len(self._index):
assert self._index[pos] == 0
elif child + 1 == len(self._index):
assert self._index[pos] == self._index[child]
else:
child_sum = self._index[child] + self._index[child + 1]
assert child_sum == self._index[pos]
except:
traceback.print_exc(file=sys.stdout)
print('len', self._len)
print('load', self._load)
print('offset', self._offset)
print('len_index', len(self._index))
print('index', self._index)
print('len_maxes', len(self._maxes))
print('maxes', self._maxes)
print('len_lists', len(self._lists))
print('lists', self._lists)
raise
def identity(value):
"Identity function."
return value
for f in range(int(input())):
n=int(input())
b=list(map(int,input().split()))
s=SortedList()
i=0
s.add(b[0])
poss=True
for x in b:
if s[i]==x:
continue
if i>0 and s[i-1]==x:
i-=1
continue
if i<len(s)-1 and s[i+1]==x:
i+=1
continue
if x<s[i]:
if i==0 or x>s[i-1]:
s.add(x)
continue
poss=False
break
if x>s[i]:
if i==len(s)-1 or x<s[i+1]:
s.add(x)
i+=1
continue
poss=False
break
print(["NO","YES"][poss])
``` | instruction | 0 | 56,640 | 12 | 113,280 |
Yes | output | 1 | 56,640 | 12 | 113,281 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a.
Submitted Solution:
```
import time
#start_time = time.time()
#def TIME_(): print(time.time()-start_time)
import os, sys
from io import BytesIO, IOBase
from types import GeneratorType
from bisect import bisect_left, bisect_right
from collections import defaultdict as dd, deque as dq, Counter as dc
import math, string, heapq as h
BUFSIZE = 8192
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
import os
self.os = os
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
def read(self):
while True:
b = self.os.read(self._fd, max(self.os.fstat(self._fd).st_size, BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = self.os.read(self._fd, max(self.os.fstat(self._fd).st_size, BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
self.os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
def getInt(): return int(input())
def getStrs(): return input().split()
def getInts(): return list(map(int,input().split()))
def getStr(): return input()
def listStr(): return list(input())
def getMat(n): return [getInts() for _ in range(n)]
def getBin(): return list(map(int,list(input())))
def isInt(s): return '0' <= s[0] <= '9'
def ceil_(a,b): return (a+b-1)//b
MOD = 10**9 + 7
"""
"""
#https://pastebin.com/sU55scy2
class SortedList:
def __init__(self, iterable=[], _load=200):
"""Initialize sorted list instance."""
values = sorted(iterable)
self._len = _len = len(values)
self._load = _load
self._lists = _lists = [values[i:i + _load] for i in range(0, _len, _load)]
self._list_lens = [len(_list) for _list in _lists]
self._mins = [_list[0] for _list in _lists]
self._fen_tree = []
self._rebuild = True
def _fen_build(self):
"""Build a fenwick tree instance."""
self._fen_tree[:] = self._list_lens
_fen_tree = self._fen_tree
for i in range(len(_fen_tree)):
if i | i + 1 < len(_fen_tree):
_fen_tree[i | i + 1] += _fen_tree[i]
self._rebuild = False
def _fen_update(self, index, value):
"""Update `fen_tree[index] += value`."""
if not self._rebuild:
_fen_tree = self._fen_tree
while index < len(_fen_tree):
_fen_tree[index] += value
index |= index + 1
def _fen_query(self, end):
"""Return `sum(_fen_tree[:end])`."""
if self._rebuild:
self._fen_build()
_fen_tree = self._fen_tree
x = 0
while end:
x += _fen_tree[end - 1]
end &= end - 1
return x
def _fen_findkth(self, k):
"""Return a pair of (the largest `idx` such that `sum(_fen_tree[:idx]) <= k`, `k - sum(_fen_tree[:idx])`)."""
_list_lens = self._list_lens
if k < _list_lens[0]:
return 0, k
if k >= self._len - _list_lens[-1]:
return len(_list_lens) - 1, k + _list_lens[-1] - self._len
if self._rebuild:
self._fen_build()
_fen_tree = self._fen_tree
idx = -1
for d in reversed(range(len(_fen_tree).bit_length())):
right_idx = idx + (1 << d)
if right_idx < len(_fen_tree) and k >= _fen_tree[right_idx]:
idx = right_idx
k -= _fen_tree[idx]
return idx + 1, k
def _delete(self, pos, idx):
"""Delete value at the given `(pos, idx)`."""
_lists = self._lists
_mins = self._mins
_list_lens = self._list_lens
self._len -= 1
self._fen_update(pos, -1)
del _lists[pos][idx]
_list_lens[pos] -= 1
if _list_lens[pos]:
_mins[pos] = _lists[pos][0]
else:
del _lists[pos]
del _list_lens[pos]
del _mins[pos]
self._rebuild = True
def _loc_left(self, value):
"""Return an index pair that corresponds to the first position of `value` in the sorted list."""
if not self._len:
return 0, 0
_lists = self._lists
_mins = self._mins
lo, pos = -1, len(_lists) - 1
while lo + 1 < pos:
mi = (lo + pos) >> 1
if value <= _mins[mi]:
pos = mi
else:
lo = mi
if pos and value <= _lists[pos - 1][-1]:
pos -= 1
_list = _lists[pos]
lo, idx = -1, len(_list)
while lo + 1 < idx:
mi = (lo + idx) >> 1
if value <= _list[mi]:
idx = mi
else:
lo = mi
return pos, idx
def _loc_right(self, value):
"""Return an index pair that corresponds to the last position of `value` in the sorted list."""
if not self._len:
return 0, 0
_lists = self._lists
_mins = self._mins
pos, hi = 0, len(_lists)
while pos + 1 < hi:
mi = (pos + hi) >> 1
if value < _mins[mi]:
hi = mi
else:
pos = mi
_list = _lists[pos]
lo, idx = -1, len(_list)
while lo + 1 < idx:
mi = (lo + idx) >> 1
if value < _list[mi]:
idx = mi
else:
lo = mi
return pos, idx
def add(self, value):
"""Add `value` to sorted list."""
_load = self._load
_lists = self._lists
_mins = self._mins
_list_lens = self._list_lens
self._len += 1
if _lists:
pos, idx = self._loc_right(value)
self._fen_update(pos, 1)
_list = _lists[pos]
_list.insert(idx, value)
_list_lens[pos] += 1
_mins[pos] = _list[0]
if _load + _load < len(_list):
_lists.insert(pos + 1, _list[_load:])
_list_lens.insert(pos + 1, len(_list) - _load)
_mins.insert(pos + 1, _list[_load])
_list_lens[pos] = _load
del _list[_load:]
self._rebuild = True
else:
_lists.append([value])
_mins.append(value)
_list_lens.append(1)
self._rebuild = True
def discard(self, value):
"""Remove `value` from sorted list if it is a member."""
_lists = self._lists
if _lists:
pos, idx = self._loc_right(value)
if idx and _lists[pos][idx - 1] == value:
self._delete(pos, idx - 1)
def remove(self, value):
"""Remove `value` from sorted list; `value` must be a member."""
_len = self._len
self.discard(value)
if _len == self._len:
raise ValueError('{0!r} not in list'.format(value))
def pop(self, index=-1):
"""Remove and return value at `index` in sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
value = self._lists[pos][idx]
self._delete(pos, idx)
return value
def bisect_left(self, value):
"""Return the first index to insert `value` in the sorted list."""
pos, idx = self._loc_left(value)
return self._fen_query(pos) + idx
def bisect_right(self, value):
"""Return the last index to insert `value` in the sorted list."""
pos, idx = self._loc_right(value)
return self._fen_query(pos) + idx
def count(self, value):
"""Return number of occurrences of `value` in the sorted list."""
return self.bisect_right(value) - self.bisect_left(value)
def __len__(self):
"""Return the size of the sorted list."""
return self._len
def __getitem__(self, index):
"""Lookup value at `index` in sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
return self._lists[pos][idx]
def __delitem__(self, index):
"""Remove value at `index` from sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
self._delete(pos, idx)
def __contains__(self, value):
"""Return true if `value` is an element of the sorted list."""
_lists = self._lists
if _lists:
pos, idx = self._loc_left(value)
return idx < len(_lists[pos]) and _lists[pos][idx] == value
return False
def __iter__(self):
"""Return an iterator over the sorted list."""
return (value for _list in self._lists for value in _list)
def __reversed__(self):
"""Return a reverse iterator over the sorted list."""
return (value for _list in reversed(self._lists) for value in reversed(_list))
def __repr__(self):
"""Return string representation of sorted list."""
return 'SortedList({0})'.format(list(self))
"""
X 2 X
X X 1 2 X
X 1 2 X X
X 1 2 3 X X X
"""
def solve():
N = getInt()
B = getInts()
A = SortedList()
A2 = SortedList()
used = set()
for i,b in enumerate(B):
idx = A.bisect_left(b)
idx2 = len(A) - A.bisect_right(b)
#We have between idx and idx2 elements which are < b
if b not in used:
used.add(b)
A.add(b)
else:
continue
if idx*2 > i+1 or idx2*2 > i+1: return "NO"
return "YES"
for _ in range(getInt()):
print(solve())
#solve()
#TIME_()
``` | instruction | 0 | 56,641 | 12 | 113,282 |
No | output | 1 | 56,641 | 12 | 113,283 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a.
Submitted Solution:
```
import random
class Node:
def __init__(self, val):
self.val = val
self.priority = random.random()
self.left = None
self.right = None
self.num_children = 0
def get_num_left_children(self):
if self.left:
return self.left.num_children + 1
return 0
def get_num_right_children(self):
return self.num_children - self.get_num_left_children()
def __repr__(self):
return str((self.val, self.priority))
class SimpleTreap: # with min heap.
def __init__(self):
self.root = None
self.size = 0
def add_node(self, val):
# assumes val doesn't appear in treap yet!
self.size += 1
if not self.root:
self.root = Node(val); return
node = Node(val)
current = self.root
parents = []
while True: # add to BST.
parents.append(current)
current.num_children += 1
if node.val < current.val:
if current.left:
current = current.left
else:
current.left = node; break
elif node.val > current.val:
if current.right:
current = current.right
else:
current.right = node; break
current = node
while parents: # reheap.
parent = parents.pop()
if parent.priority <= current.priority:
return
if parent.val > current.val:
num_A = current.get_num_left_children()
num_B = current.get_num_right_children()
num_C = parent.get_num_right_children()
parent.num_children = num_B + num_C
current.num_children = parent.num_children + num_A + 1
parent.left = current.right
current.right = parent
if parents:
parents[-1].left = current
else:
num_A = parent.get_num_left_children()
num_B = current.get_num_left_children()
num_C = current.get_num_right_children()
parent.num_children = num_A + num_B
current.num_children = parent.num_children + num_C + 1
parent.right = current.left
current.left = parent
if parents:
if parents[-1].val > parent.val:
parents[-1].left = current
else:
parents[-1].right = current
else:
self.root = current
def get_pos(self, val):
# assumes val appears in treap.
current = self.root
to_the_left = 0
while True:
if current.val == val:
return current.get_num_left_children() + to_the_left
elif current.val > val:
current = current.left
else:
to_the_left += current.get_num_left_children() + 1
current = current.right
def solve(n, b):
se = set(); prev = -1
treap = SimpleTreap()
for i in range(n):
if b[i] in se:
x = treap.get_pos(b[i])
if abs(prev - x) >= 2:
return False
prev = x
else:
treap.add_node(b[i])
x = treap.get_pos(b[i])
se.add(b[i])
if not (prev == x or prev == x - 1):
return False
prev = x
return True
def main():
t = int(input())
for _ in range(t):
n = int(input())
b = list(map(int, input().split()))
if solve(n, b):
print("YES")
else:
print("NO")
main()
``` | instruction | 0 | 56,642 | 12 | 113,284 |
No | output | 1 | 56,642 | 12 | 113,285 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a.
Submitted Solution:
```
def gcd(a,b):
if b==0:
return a
return gcd(b,a%b)
for _ in range(int(input())):
n=int(input())
s=input()
ans=[0]*n
d=0
ans[0]=1
m={}
if s[0]=='D':
d=1
m[(1,0)]=1
else:
m[(0,1)]=1
for i in range(1,n):
if s[i]=='D':
d+=1
g=gcd(d,i+1-d)
a,b=d//g,(i+1-d)//g
if (a,b) not in m:
m[(a,b)]=1
else:
m[(a,b)]+=1
ans[i]=m[(a,b)]
print(*ans)
``` | instruction | 0 | 56,643 | 12 | 113,286 |
No | output | 1 | 56,643 | 12 | 113,287 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Uh oh! Ray lost his array yet again! However, Omkar might be able to help because he thinks he has found the OmkArray of Ray's array. The OmkArray of an array a with elements a_1, a_2, …, a_{2k-1}, is the array b with elements b_1, b_2, …, b_{k} such that b_i is equal to the median of a_1, a_2, …, a_{2i-1} for all i. Omkar has found an array b of size n (1 ≤ n ≤ 2 ⋅ 10^5, -10^9 ≤ b_i ≤ 10^9). Given this array b, Ray wants to test Omkar's claim and see if b actually is an OmkArray of some array a. Can you help Ray?
The median of a set of numbers a_1, a_2, …, a_{2i-1} is the number c_{i} where c_{1}, c_{2}, …, c_{2i-1} represents a_1, a_2, …, a_{2i-1} sorted in nondecreasing order.
Input
Each test contains multiple test cases. The first line contains a single integer t (1 ≤ t ≤ 10^4) — the number of test cases. Description of the test cases follows.
The first line of each test case contains an integer n (1 ≤ n ≤ 2 ⋅ 10^5) — the length of the array b.
The second line contains n integers b_1, b_2, …, b_n (-10^9 ≤ b_i ≤ 10^9) — the elements of b.
It is guaranteed the sum of n across all test cases does not exceed 2 ⋅ 10^5.
Output
For each test case, output one line containing YES if there exists an array a such that b_i is the median of a_1, a_2, ..., a_{2i-1} for all i, and NO otherwise. The case of letters in YES and NO do not matter (so yEs and No will also be accepted).
Examples
Input
5
4
6 2 1 3
1
4
5
4 -8 5 6 -7
2
3 3
4
2 1 2 3
Output
NO
YES
NO
YES
YES
Input
5
8
-8 2 -6 -5 -4 3 3 2
7
1 1 3 1 0 -2 -1
7
6 12 8 6 2 6 10
6
5 1 2 3 6 7
5
1 3 4 3 0
Output
NO
YES
NO
NO
NO
Note
In the second case of the first sample, the array [4] will generate an OmkArray of [4], as the median of the first element is 4.
In the fourth case of the first sample, the array [3, 2, 5] will generate an OmkArray of [3, 3], as the median of 3 is 3 and the median of 2, 3, 5 is 3.
In the fifth case of the first sample, the array [2, 1, 0, 3, 4, 4, 3] will generate an OmkArray of [2, 1, 2, 3] as
* the median of 2 is 2
* the median of 0, 1, 2 is 1
* the median of 0, 1, 2, 3, 4 is 2
* and the median of 0, 1, 2, 3, 3, 4, 4 is 3.
In the second case of the second sample, the array [1, 0, 4, 3, 5, -2, -2, -2, -4, -3, -4, -1, 5] will generate an OmkArray of [1, 1, 3, 1, 0, -2, -1], as
* the median of 1 is 1
* the median of 0, 1, 4 is 1
* the median of 0, 1, 3, 4, 5 is 3
* the median of -2, -2, 0, 1, 3, 4, 5 is 1
* the median of -4, -2, -2, -2, 0, 1, 3, 4, 5 is 0
* the median of -4, -4, -3, -2, -2, -2, 0, 1, 3, 4, 5 is -2
* and the median of -4, -4, -3, -2, -2, -2, -1, 0, 1, 3, 4, 5, 5 is -1
For all cases where the answer is NO, it can be proven that it is impossible to find an array a such that b is the OmkArray of a.
Submitted Solution:
```
import time
#start_time = time.time()
#def TIME_(): print(time.time()-start_time)
import os, sys
from io import BytesIO, IOBase
from types import GeneratorType
from bisect import bisect_left, bisect_right
from collections import defaultdict as dd, deque as dq, Counter as dc
import math, string, heapq as h
BUFSIZE = 8192
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
import os
self.os = os
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
def read(self):
while True:
b = self.os.read(self._fd, max(self.os.fstat(self._fd).st_size, BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = self.os.read(self._fd, max(self.os.fstat(self._fd).st_size, BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
self.os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
def getInt(): return int(input())
def getStrs(): return input().split()
def getInts(): return list(map(int,input().split()))
def getStr(): return input()
def listStr(): return list(input())
def getMat(n): return [getInts() for _ in range(n)]
def getBin(): return list(map(int,list(input())))
def isInt(s): return '0' <= s[0] <= '9'
def ceil_(a,b): return (a+b-1)//b
MOD = 10**9 + 7
"""
"""
#https://pastebin.com/sU55scy2
class SortedList:
def __init__(self, iterable=[], _load=200):
"""Initialize sorted list instance."""
values = sorted(iterable)
self._len = _len = len(values)
self._load = _load
self._lists = _lists = [values[i:i + _load] for i in range(0, _len, _load)]
self._list_lens = [len(_list) for _list in _lists]
self._mins = [_list[0] for _list in _lists]
self._fen_tree = []
self._rebuild = True
def _fen_build(self):
"""Build a fenwick tree instance."""
self._fen_tree[:] = self._list_lens
_fen_tree = self._fen_tree
for i in range(len(_fen_tree)):
if i | i + 1 < len(_fen_tree):
_fen_tree[i | i + 1] += _fen_tree[i]
self._rebuild = False
def _fen_update(self, index, value):
"""Update `fen_tree[index] += value`."""
if not self._rebuild:
_fen_tree = self._fen_tree
while index < len(_fen_tree):
_fen_tree[index] += value
index |= index + 1
def _fen_query(self, end):
"""Return `sum(_fen_tree[:end])`."""
if self._rebuild:
self._fen_build()
_fen_tree = self._fen_tree
x = 0
while end:
x += _fen_tree[end - 1]
end &= end - 1
return x
def _fen_findkth(self, k):
"""Return a pair of (the largest `idx` such that `sum(_fen_tree[:idx]) <= k`, `k - sum(_fen_tree[:idx])`)."""
_list_lens = self._list_lens
if k < _list_lens[0]:
return 0, k
if k >= self._len - _list_lens[-1]:
return len(_list_lens) - 1, k + _list_lens[-1] - self._len
if self._rebuild:
self._fen_build()
_fen_tree = self._fen_tree
idx = -1
for d in reversed(range(len(_fen_tree).bit_length())):
right_idx = idx + (1 << d)
if right_idx < len(_fen_tree) and k >= _fen_tree[right_idx]:
idx = right_idx
k -= _fen_tree[idx]
return idx + 1, k
def _delete(self, pos, idx):
"""Delete value at the given `(pos, idx)`."""
_lists = self._lists
_mins = self._mins
_list_lens = self._list_lens
self._len -= 1
self._fen_update(pos, -1)
del _lists[pos][idx]
_list_lens[pos] -= 1
if _list_lens[pos]:
_mins[pos] = _lists[pos][0]
else:
del _lists[pos]
del _list_lens[pos]
del _mins[pos]
self._rebuild = True
def _loc_left(self, value):
"""Return an index pair that corresponds to the first position of `value` in the sorted list."""
if not self._len:
return 0, 0
_lists = self._lists
_mins = self._mins
lo, pos = -1, len(_lists) - 1
while lo + 1 < pos:
mi = (lo + pos) >> 1
if value <= _mins[mi]:
pos = mi
else:
lo = mi
if pos and value <= _lists[pos - 1][-1]:
pos -= 1
_list = _lists[pos]
lo, idx = -1, len(_list)
while lo + 1 < idx:
mi = (lo + idx) >> 1
if value <= _list[mi]:
idx = mi
else:
lo = mi
return pos, idx
def _loc_right(self, value):
"""Return an index pair that corresponds to the last position of `value` in the sorted list."""
if not self._len:
return 0, 0
_lists = self._lists
_mins = self._mins
pos, hi = 0, len(_lists)
while pos + 1 < hi:
mi = (pos + hi) >> 1
if value < _mins[mi]:
hi = mi
else:
pos = mi
_list = _lists[pos]
lo, idx = -1, len(_list)
while lo + 1 < idx:
mi = (lo + idx) >> 1
if value < _list[mi]:
idx = mi
else:
lo = mi
return pos, idx
def add(self, value):
"""Add `value` to sorted list."""
_load = self._load
_lists = self._lists
_mins = self._mins
_list_lens = self._list_lens
self._len += 1
if _lists:
pos, idx = self._loc_right(value)
self._fen_update(pos, 1)
_list = _lists[pos]
_list.insert(idx, value)
_list_lens[pos] += 1
_mins[pos] = _list[0]
if _load + _load < len(_list):
_lists.insert(pos + 1, _list[_load:])
_list_lens.insert(pos + 1, len(_list) - _load)
_mins.insert(pos + 1, _list[_load])
_list_lens[pos] = _load
del _list[_load:]
self._rebuild = True
else:
_lists.append([value])
_mins.append(value)
_list_lens.append(1)
self._rebuild = True
def discard(self, value):
"""Remove `value` from sorted list if it is a member."""
_lists = self._lists
if _lists:
pos, idx = self._loc_right(value)
if idx and _lists[pos][idx - 1] == value:
self._delete(pos, idx - 1)
def remove(self, value):
"""Remove `value` from sorted list; `value` must be a member."""
_len = self._len
self.discard(value)
if _len == self._len:
raise ValueError('{0!r} not in list'.format(value))
def pop(self, index=-1):
"""Remove and return value at `index` in sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
value = self._lists[pos][idx]
self._delete(pos, idx)
return value
def bisect_left(self, value):
"""Return the first index to insert `value` in the sorted list."""
pos, idx = self._loc_left(value)
return self._fen_query(pos) + idx
def bisect_right(self, value):
"""Return the last index to insert `value` in the sorted list."""
pos, idx = self._loc_right(value)
return self._fen_query(pos) + idx
def count(self, value):
"""Return number of occurrences of `value` in the sorted list."""
return self.bisect_right(value) - self.bisect_left(value)
def __len__(self):
"""Return the size of the sorted list."""
return self._len
def __getitem__(self, index):
"""Lookup value at `index` in sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
return self._lists[pos][idx]
def __delitem__(self, index):
"""Remove value at `index` from sorted list."""
pos, idx = self._fen_findkth(self._len + index if index < 0 else index)
self._delete(pos, idx)
def __contains__(self, value):
"""Return true if `value` is an element of the sorted list."""
_lists = self._lists
if _lists:
pos, idx = self._loc_left(value)
return idx < len(_lists[pos]) and _lists[pos][idx] == value
return False
def __iter__(self):
"""Return an iterator over the sorted list."""
return (value for _list in self._lists for value in _list)
def __reversed__(self):
"""Return a reverse iterator over the sorted list."""
return (value for _list in reversed(self._lists) for value in reversed(_list))
def __repr__(self):
"""Return string representation of sorted list."""
return 'SortedList({0})'.format(list(self))
def solve():
N = getInt()
B = getInts()
used = set()
num_used = 0
A = SortedList()
for i,b in enumerate(B):
if b in used:
num_used += 1
continue
idx = A.bisect_left(b)
if 2*(idx-num_used) > i:
return "NO"
used.add(b)
A.add(b)
return "YES"
for _ in range(getInt()):
print(solve())
#solve()
#TIME_()
``` | instruction | 0 | 56,644 | 12 | 113,288 |
No | output | 1 | 56,644 | 12 | 113,289 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Yaroslav has an array, consisting of (2·n - 1) integers. In a single operation Yaroslav can change the sign of exactly n elements in the array. In other words, in one operation Yaroslav can select exactly n array elements, and multiply each of them by -1.
Yaroslav is now wondering: what maximum sum of array elements can be obtained if it is allowed to perform any number of described operations?
Help Yaroslav.
Input
The first line contains an integer n (2 ≤ n ≤ 100). The second line contains (2·n - 1) integers — the array elements. The array elements do not exceed 1000 in their absolute value.
Output
In a single line print the answer to the problem — the maximum sum that Yaroslav can get.
Examples
Input
2
50 50 50
Output
150
Input
2
-1 -100 -1
Output
100
Note
In the first sample you do not need to change anything. The sum of elements equals 150.
In the second sample you need to change the sign of the first two elements. Then we get the sum of the elements equal to 100. | instruction | 0 | 56,657 | 12 | 113,314 |
Tags: constructive algorithms
Correct Solution:
```
#!/bin/python
n = int(input())
a = list(map(int, input().split()))
c = 0
for cc in a:
if cc < 0:
c += 1
if 0 in a:
print(sum(abs(x) for x in a))
elif n % 2 == 0 and c % 2 == 1:
print(sum(abs(x) for x in a) - 2*min(map(abs, a)))
else:
print(sum(abs(x) for x in a))
``` | output | 1 | 56,657 | 12 | 113,315 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Yaroslav has an array, consisting of (2·n - 1) integers. In a single operation Yaroslav can change the sign of exactly n elements in the array. In other words, in one operation Yaroslav can select exactly n array elements, and multiply each of them by -1.
Yaroslav is now wondering: what maximum sum of array elements can be obtained if it is allowed to perform any number of described operations?
Help Yaroslav.
Input
The first line contains an integer n (2 ≤ n ≤ 100). The second line contains (2·n - 1) integers — the array elements. The array elements do not exceed 1000 in their absolute value.
Output
In a single line print the answer to the problem — the maximum sum that Yaroslav can get.
Examples
Input
2
50 50 50
Output
150
Input
2
-1 -100 -1
Output
100
Note
In the first sample you do not need to change anything. The sum of elements equals 150.
In the second sample you need to change the sign of the first two elements. Then we get the sum of the elements equal to 100. | instruction | 0 | 56,658 | 12 | 113,316 |
Tags: constructive algorithms
Correct Solution:
```
n = int(input())
a = list(map(int, input().split()))
t = 0
for i in range(len(a)):
if a[i] < 0:
t += 1
a[i] = -a[i]
if t % 2 == 0 or n % 2 == 1:
print(sum(a))
else:
print(sum(a) - min(a) * 2)
``` | output | 1 | 56,658 | 12 | 113,317 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Yaroslav has an array, consisting of (2·n - 1) integers. In a single operation Yaroslav can change the sign of exactly n elements in the array. In other words, in one operation Yaroslav can select exactly n array elements, and multiply each of them by -1.
Yaroslav is now wondering: what maximum sum of array elements can be obtained if it is allowed to perform any number of described operations?
Help Yaroslav.
Input
The first line contains an integer n (2 ≤ n ≤ 100). The second line contains (2·n - 1) integers — the array elements. The array elements do not exceed 1000 in their absolute value.
Output
In a single line print the answer to the problem — the maximum sum that Yaroslav can get.
Examples
Input
2
50 50 50
Output
150
Input
2
-1 -100 -1
Output
100
Note
In the first sample you do not need to change anything. The sum of elements equals 150.
In the second sample you need to change the sign of the first two elements. Then we get the sum of the elements equal to 100. | instruction | 0 | 56,659 | 12 | 113,318 |
Tags: constructive algorithms
Correct Solution:
```
n = int(input())
a = list(map(int, input().split()))
c = 0
for cc in a:
#print("Element is: " + str(cc))
# Reverse sign
if cc < 0:
c += 1
if 0 in a:
#print("matched first condition")
print(sum(abs(x) for x in a))
elif n % 2 == 0 and c % 2 == 1:
#print("matched second condition")
print(sum(abs(x) for x in a) - 2*min(map(abs, a)))
else:
#print("matched third condition")
print(sum(abs(x) for x in a))
``` | output | 1 | 56,659 | 12 | 113,319 |
Provide tags and a correct Python 3 solution for this coding contest problem.
Yaroslav has an array, consisting of (2·n - 1) integers. In a single operation Yaroslav can change the sign of exactly n elements in the array. In other words, in one operation Yaroslav can select exactly n array elements, and multiply each of them by -1.
Yaroslav is now wondering: what maximum sum of array elements can be obtained if it is allowed to perform any number of described operations?
Help Yaroslav.
Input
The first line contains an integer n (2 ≤ n ≤ 100). The second line contains (2·n - 1) integers — the array elements. The array elements do not exceed 1000 in their absolute value.
Output
In a single line print the answer to the problem — the maximum sum that Yaroslav can get.
Examples
Input
2
50 50 50
Output
150
Input
2
-1 -100 -1
Output
100
Note
In the first sample you do not need to change anything. The sum of elements equals 150.
In the second sample you need to change the sign of the first two elements. Then we get the sum of the elements equal to 100. | instruction | 0 | 56,660 | 12 | 113,320 |
Tags: constructive algorithms
Correct Solution:
```
n = int(input())
a = list(map(int, input().split()))
c = list(map(abs, a))
if len(list(filter(lambda x: x < 0, a))) & 1 and n + 1 & 1:
print(sum(c) - 2 * min(c))
else:
print(sum(c))
# Made By Mostafa_Khaled
``` | output | 1 | 56,660 | 12 | 113,321 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
Yaroslav has an array, consisting of (2·n - 1) integers. In a single operation Yaroslav can change the sign of exactly n elements in the array. In other words, in one operation Yaroslav can select exactly n array elements, and multiply each of them by -1.
Yaroslav is now wondering: what maximum sum of array elements can be obtained if it is allowed to perform any number of described operations?
Help Yaroslav.
Input
The first line contains an integer n (2 ≤ n ≤ 100). The second line contains (2·n - 1) integers — the array elements. The array elements do not exceed 1000 in their absolute value.
Output
In a single line print the answer to the problem — the maximum sum that Yaroslav can get.
Examples
Input
2
50 50 50
Output
150
Input
2
-1 -100 -1
Output
100
Note
In the first sample you do not need to change anything. The sum of elements equals 150.
In the second sample you need to change the sign of the first two elements. Then we get the sum of the elements equal to 100.
Submitted Solution:
```
n = int(input())
data = list(map(int, input().split()))
x = 0
for element in data:
if element < 0:
x =+ abs(x)
print(x)
values = sorted(list(map(abs, data)))
result = sum(values)
if n % 2 == 0 and x % 2 == 1:
result -= 2*values[0]
print(result)
``` | instruction | 0 | 56,661 | 12 | 113,322 |
No | output | 1 | 56,661 | 12 | 113,323 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given a permutation of n numbers p1, p2, ..., pn. We perform k operations of the following type: choose uniformly at random two indices l and r (l ≤ r) and reverse the order of the elements pl, pl + 1, ..., pr. Your task is to find the expected value of the number of inversions in the resulting permutation.
Input
The first line of input contains two integers n and k (1 ≤ n ≤ 100, 1 ≤ k ≤ 109). The next line contains n integers p1, p2, ..., pn — the given permutation. All pi are different and in range from 1 to n.
The problem consists of three subproblems. The subproblems have different constraints on the input. You will get some score for the correct submission of the subproblem. The description of the subproblems follows.
* In subproblem G1 (3 points), the constraints 1 ≤ n ≤ 6, 1 ≤ k ≤ 4 will hold.
* In subproblem G2 (5 points), the constraints 1 ≤ n ≤ 30, 1 ≤ k ≤ 200 will hold.
* In subproblem G3 (16 points), the constraints 1 ≤ n ≤ 100, 1 ≤ k ≤ 109 will hold.
Output
Output the answer with absolute or relative error no more than 1e - 9.
Examples
Input
3 1
1 2 3
Output
0.833333333333333
Input
3 4
1 3 2
Output
1.458333333333334
Note
Consider the first sample test. We will randomly pick an interval of the permutation (1, 2, 3) (which has no inversions) and reverse the order of its elements. With probability <image>, the interval will consist of a single element and the permutation will not be altered. With probability <image> we will inverse the first two elements' order and obtain the permutation (2, 1, 3) which has one inversion. With the same probability we might pick the interval consisting of the last two elements which will lead to the permutation (1, 3, 2) with one inversion. Finally, with probability <image> the randomly picked interval will contain all elements, leading to the permutation (3, 2, 1) with 3 inversions. Hence, the expected number of inversions is equal to <image>. | instruction | 0 | 56,708 | 12 | 113,416 |
Tags: dp
Correct Solution:
```
f = lambda: map(int, input().split())
n, k = f()
p = list(f())
r = range
u = [l * l + l >> 1 for l in r(n + 1)]
v = [(i, j) for i in r(n) for j in r(i + 1, n)]
t = [[p[i] > p[j] for j in r(n)] for i in r(n)]
a = [[0] * n for i in r(n)]
b = [[0] * n for i in r(n)]
c = [[0] * n for i in r(n)]
for l in r(min(k, 1000)):
for j in r(1, n):
s, x = 0, a[j]
for i in r(j):
s += t[i][j]
x[i + 1] = x[i] + s
for i in r(n):
s, y = 0, b[i]
for j in r(n - 1, i, -1):
s += t[i][j]
y[j - 1] = y[j] + s
for d in r(1, n):
s, z = 0, c[d]
for i in r(n - d):
s += t[i][i + d]
z[i + 1] = z[i] + s
for i, j in v:
d = j - i
x, y, z = a[j], b[i], c[d]
s = t[i][j] * (u[i] + u[d - 1] + u[n - j - 1])
s += x[j] - x[i] - x[d - 1]
s += y[i] - y[j] - y[n - d]
s += (i + 1) * (n - j) - z[n - d] + z[n - j - 1] + z[i]
t[i][j] = s / u[n]
print(sum(t[i][j] for i, j in v))
``` | output | 1 | 56,708 | 12 | 113,417 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given a permutation of n numbers p1, p2, ..., pn. We perform k operations of the following type: choose uniformly at random two indices l and r (l ≤ r) and reverse the order of the elements pl, pl + 1, ..., pr. Your task is to find the expected value of the number of inversions in the resulting permutation.
Input
The first line of input contains two integers n and k (1 ≤ n ≤ 100, 1 ≤ k ≤ 109). The next line contains n integers p1, p2, ..., pn — the given permutation. All pi are different and in range from 1 to n.
The problem consists of three subproblems. The subproblems have different constraints on the input. You will get some score for the correct submission of the subproblem. The description of the subproblems follows.
* In subproblem G1 (3 points), the constraints 1 ≤ n ≤ 6, 1 ≤ k ≤ 4 will hold.
* In subproblem G2 (5 points), the constraints 1 ≤ n ≤ 30, 1 ≤ k ≤ 200 will hold.
* In subproblem G3 (16 points), the constraints 1 ≤ n ≤ 100, 1 ≤ k ≤ 109 will hold.
Output
Output the answer with absolute or relative error no more than 1e - 9.
Examples
Input
3 1
1 2 3
Output
0.833333333333333
Input
3 4
1 3 2
Output
1.458333333333334
Note
Consider the first sample test. We will randomly pick an interval of the permutation (1, 2, 3) (which has no inversions) and reverse the order of its elements. With probability <image>, the interval will consist of a single element and the permutation will not be altered. With probability <image> we will inverse the first two elements' order and obtain the permutation (2, 1, 3) which has one inversion. With the same probability we might pick the interval consisting of the last two elements which will lead to the permutation (1, 3, 2) with one inversion. Finally, with probability <image> the randomly picked interval will contain all elements, leading to the permutation (3, 2, 1) with 3 inversions. Hence, the expected number of inversions is equal to <image>.
Submitted Solution:
```
f = lambda: map(int, input().split())
n, k = f()
p = list(f())
a = [[0] * n for i in range(n)]
g = [k * k - k >> 1 for k in range(n)]
for i in range(n):
for j in range(i + 1, n):
if p[i] > p[j]: a[i][j] = 1
else: a[j][i] = 1
for t in range(max(k, 1000)):
b = [[0] * n for i in range(n)]
for i in range(n):
for j in range(i + 1, n):
p = q = 0
for x in range(j):
d = min(i + 1, j - x, x + 1, j - i)
p += d * a[x][j]
q += d
for y in range(i + 1, n):
d = min(n - j, y - i, n - y, j - i)
p += d * a[i][y]
q += d
for s in range(j, i + n):
x, y = s - i, s - j
d = min(i + 1, n - j, y + 1, n - x)
p += d * a[x][y]
q += d
d = g[j - i] + g[i + 1] + g[n - j]
b[i][j] = (p + d * a[i][j]) / (d + q)
a = b
for i in range(n):
for j in range(i + 1, n):
a[j][i] = 1 - a[i][j]
s = 0
for i in range(n):
for j in range(i + 1, n): s += a[i][j]
print(s)
``` | instruction | 0 | 56,709 | 12 | 113,418 |
No | output | 1 | 56,709 | 12 | 113,419 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given a permutation of n numbers p1, p2, ..., pn. We perform k operations of the following type: choose uniformly at random two indices l and r (l ≤ r) and reverse the order of the elements pl, pl + 1, ..., pr. Your task is to find the expected value of the number of inversions in the resulting permutation.
Input
The first line of input contains two integers n and k (1 ≤ n ≤ 100, 1 ≤ k ≤ 109). The next line contains n integers p1, p2, ..., pn — the given permutation. All pi are different and in range from 1 to n.
The problem consists of three subproblems. The subproblems have different constraints on the input. You will get some score for the correct submission of the subproblem. The description of the subproblems follows.
* In subproblem G1 (3 points), the constraints 1 ≤ n ≤ 6, 1 ≤ k ≤ 4 will hold.
* In subproblem G2 (5 points), the constraints 1 ≤ n ≤ 30, 1 ≤ k ≤ 200 will hold.
* In subproblem G3 (16 points), the constraints 1 ≤ n ≤ 100, 1 ≤ k ≤ 109 will hold.
Output
Output the answer with absolute or relative error no more than 1e - 9.
Examples
Input
3 1
1 2 3
Output
0.833333333333333
Input
3 4
1 3 2
Output
1.458333333333334
Note
Consider the first sample test. We will randomly pick an interval of the permutation (1, 2, 3) (which has no inversions) and reverse the order of its elements. With probability <image>, the interval will consist of a single element and the permutation will not be altered. With probability <image> we will inverse the first two elements' order and obtain the permutation (2, 1, 3) which has one inversion. With the same probability we might pick the interval consisting of the last two elements which will lead to the permutation (1, 3, 2) with one inversion. Finally, with probability <image> the randomly picked interval will contain all elements, leading to the permutation (3, 2, 1) with 3 inversions. Hence, the expected number of inversions is equal to <image>.
Submitted Solution:
```
f = lambda: map(int, input().split())
n, k = f()
p = list(f())
a = [[0] * n for i in range(n)]
g = [i * i - i >> 1 for i in range(n)]
for i in range(n):
for j in range(i + 1, n):
if p[i] > p[j]: a[i][j] = 1
else: a[j][i] = 1
for t in range(max(k, 1000)):
b = [[0] * n for i in range(n)]
for i in range(n):
for j in range(i + 1, n):
p = q = 0
for x in range(j):
d = min(i + 1, j - x, x + 1, j - i)
p += d * a[x][j]
q += d
for y in range(i + 1, n):
d = min(n - j, y - i, n - y, j - i)
p += d * a[i][y]
q += d
for s in range(j, i + n):
x, y = s - i, s - j
d = min(i + 1, n - j, y + 1, n - x)
p += d * a[x][y]
q += d
d = g[j - i] + g[i + 1] + g[n - j]
b[i][j] = (p + d * a[i][j]) / (d + q)
a = b
for i in range(n):
for j in range(i + 1, n):
a[j][i] = 1 - a[i][j]
s = 0
for i in range(n):
for j in range(i + 1, n): s += a[i][j]
print(s)
``` | instruction | 0 | 56,710 | 12 | 113,420 |
No | output | 1 | 56,710 | 12 | 113,421 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3. | instruction | 0 | 56,833 | 12 | 113,666 |
Tags: dp, greedy, implementation
Correct Solution:
```
n=int(input())
l=list(map(int,input().split()))
c=[[],[]]
for i in l:
c[i%2].append(i)
c[0].sort()
c[1].sort()
ans=0
for i in c[0]:
if i>0:
ans+=i
ans+=c[1][-1]
for i in range(len(c[1])-2,0,-2):
ans=max(ans,ans+c[1][i]+c[1][i-1])
print(ans)
``` | output | 1 | 56,833 | 12 | 113,667 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3. | instruction | 0 | 56,834 | 12 | 113,668 |
Tags: dp, greedy, implementation
Correct Solution:
```
def main():
input()
aa, l = list(map(int, input().split())), []
for i, a in enumerate(aa):
if a & 1:
l.append(i)
elif a < 0:
aa[i] = 0
l.sort(key=aa.__getitem__)
i = 1 - len(l) & 1
if i:
aa[l[0]] = 0
for j, k in zip(l[i::2], l[i + 1::2]):
if aa[j] + aa[k] >= 0:
break
aa[j] = aa[k] = 0
print(sum(aa))
if __name__ == '__main__':
main()
``` | output | 1 | 56,834 | 12 | 113,669 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3. | instruction | 0 | 56,835 | 12 | 113,670 |
Tags: dp, greedy, implementation
Correct Solution:
```
input()
a=[int(x) for x in input().split()]
oc=0
ps=0
pmo=1e6
nmo=-1e6
for x in a:
if x>0:
ps+=x
if x%2==1 and x>0 and pmo>x:
pmo=x
if x%2==1 and x>0:
oc+=1
if x%2==1 and x<0 and nmo<x:
nmo=x
if oc%2==1:
print(ps)
else:
print(max(ps-pmo,ps+nmo))
``` | output | 1 | 56,835 | 12 | 113,671 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3. | instruction | 0 | 56,836 | 12 | 113,672 |
Tags: dp, greedy, implementation
Correct Solution:
```
n = int(input())
arr = list(map(int,input().split()))
s = sum([a for a in arr if a>0])
if s%2==1:
print(s)
else:
mn_odd = 10001
for ele in arr:
if ele%2==1 and ele>0:
mn_odd = min(mn_odd,ele)
mx_odd = -10001
for ele in arr:
if ele%2==1 and ele<0:
mx_odd = max(mx_odd,ele)
print(max(s-mn_odd,s+mx_odd))
``` | output | 1 | 56,836 | 12 | 113,673 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3. | instruction | 0 | 56,837 | 12 | 113,674 |
Tags: dp, greedy, implementation
Correct Solution:
```
n = int(input())
arr = list(map(int,input().split()))
pos,neg = [],[]
for i in arr:
if i>=0:
pos+=[i]
else:
neg+=[i]
ans = sum(pos)
m_neg = -10**10
s_odd = 10**10
for i in neg:
if i%2!=0:
m_neg = max(m_neg,i)
for i in pos:
if i%2!=0:
s_odd = min(s_odd,i)
if ans%2!=0:
print(ans)
else:
ans1,ans2 = ans,ans
if m_neg!=(-10**10):
ans1 = ans+m_neg
if s_odd!=(10**10):
ans2 = ans-s_odd
# print(ans1,ans2,m_neg,s_odd)
if ans1%2!=0 and ans2%2!=0:
print(max(ans1,ans2))
elif ans1%2==0 and ans2%2!=0:
print(ans2)
elif ans1%2!=0 and ans2%2==0:
print(ans1)
``` | output | 1 | 56,837 | 12 | 113,675 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3. | instruction | 0 | 56,838 | 12 | 113,676 |
Tags: dp, greedy, implementation
Correct Solution:
```
n=int(input())
l=[int(i) for i in input().split()]
s=sum(i for i in l if i>0)
if s%2:
print(s)
else:
m=[abs(i) for i in l if i%2==1]
print(s-min(m))
``` | output | 1 | 56,838 | 12 | 113,677 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3. | instruction | 0 | 56,839 | 12 | 113,678 |
Tags: dp, greedy, implementation
Correct Solution:
```
n =int(input())
ar = list(map(int,input().split()))
odd = [i for i in ar if i%2==1]
even =[i for i in ar if i%2==0]
odd.sort(reverse = True)
even.sort(reverse = True)
sum_even = 0
for i in even:
if i>0:
sum_even+=i
preSum = []
preSum.append(odd[0])
for i in range(1,len(odd)):
preSum.append(preSum[i-1]+odd[i])
max_odd_sum = preSum[0]
for i in range(2,len(odd),2):
max_odd_sum = max(max_odd_sum,preSum[i])
print(max(max_odd_sum+sum_even,max_odd_sum))
# print(odd)
# print(even)
# print(preSum)
``` | output | 1 | 56,839 | 12 | 113,679 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3. | instruction | 0 | 56,840 | 12 | 113,680 |
Tags: dp, greedy, implementation
Correct Solution:
```
n=int(input())
a=list(map(int,input().split()))
d=[]
a.sort()
e=[]
for j in range(n):
if(a[j]>=0):
d.append(a[j])
else:
e.append(a[j])
if(len(d)==0):
for j in range(n-1,-1,-1):
if(a[j]%2==1):
val=a[j]
break
else:
s=sum(d)
if(s%2==1):
val=s
else:
flag=0
s1=s
s2=s
for j in range(len(d)):
if(d[j]%2==1):
flag=1
s1-=d[j]
break
flag1=0
for j in range(len(e)-1,-1,-1):
if(e[j]%2==1):
flag1=1
s2+=e[j]
break
if(flag==0):
val=s2
elif(flag1==0):
val=s1
else:
val=max(s1,s2)
print(val)
``` | output | 1 | 56,840 | 12 | 113,681 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3.
Submitted Solution:
```
input()
items = list(map(int, input().split()))
out=sum(filter(lambda n:n>0, items) )
if out%2==0:
out=out-min(map(abs,filter(lambda n:n%2==1, items)) )
print(out)
# Made By Mostafa_Khaled
``` | instruction | 0 | 56,841 | 12 | 113,682 |
Yes | output | 1 | 56,841 | 12 | 113,683 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3.
Submitted Solution:
```
def makeodd(listt):
c=0
for r in range(len(listt)):
if listt[r]%2==0:
listt[r]="blah"
c+=1
for e in range(c):
listt.remove("blah")
n=int(input())
inp=input()
nums=inp.split(" ")
for w in range(len(nums)):
nums[w]=int(nums[w])
sum=0
tsum=0
neg=[]
pos=[]
an=False
noTroll=True
for i in range(n):
if nums[i]>0:
sum+=nums[i]
pos.append(nums[i])
elif nums[i]<0:
neg.append(nums[i])
if sum%2==1:
print(sum)
else:
neg.sort()
neg.reverse()
pos.sort()
makeodd(neg)
makeodd(pos)
z=0
while sum%2==0 and noTroll:
if len(pos)>z and len(neg)>z:
if pos[z]%2!=0 and pos[z]<neg[z]*-1:
sum-=pos[z]
break
elif neg[z]%2!=0:
for q in range(len(pos)-z):
if pos[q+z]>neg[z]*-1 and pos[q+z]%2!=0:
an=True
break
if an:
sum+=neg[z]
break
elif len(neg)>z:
if neg[z]%2!=0:
sum+=neg[z]
break
else:
if len(pos)>z:
if pos[z]%2!=0:
sum-=pos[z]
break
z+=1
print(sum)
"yay"
``` | instruction | 0 | 56,842 | 12 | 113,684 |
Yes | output | 1 | 56,842 | 12 | 113,685 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3.
Submitted Solution:
```
n = int(input())
A = list(map(int, input().split()))
dp = [[-9999999999, -9999999999]]
for elem in A:
dp += [[0, 0]]
if elem % 2 == 0:
dp[-1][0] = max(dp[-2][0] + elem, dp[-2][0], elem)
dp[-1][1] = max(dp[-2][1] + elem, dp[-2][1])
else:
dp[-1][0] = max(dp[-2][1] + elem, dp[-2][0])
dp[-1][1] = max(dp[-2][0] + elem, dp[-2][1], elem)
print(dp[-1][1])
``` | instruction | 0 | 56,843 | 12 | 113,686 |
Yes | output | 1 | 56,843 | 12 | 113,687 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3.
Submitted Solution:
```
n=int(input())
arr=list(map(int,input().split()))
i=0
arr=sorted(arr,reverse=True)
Sum=0
while i<n and arr[i]>0:
Sum+=arr[i]
i+=1
if Sum%2==1:
print(Sum)
else:
k=i
negSum=Sum
while k<n and negSum%2==0:
if arr[k]%2==1:
negSum+=arr[k]
k+=1
posSum=Sum
j=i-1
while j>=0 and posSum%2==0:
if arr[j]%2==1:
posSum-=arr[j]
j-=1
if posSum%2==0 and negSum%2==1:
print(negSum)
elif posSum%2==1 and negSum%2==0:
print(posSum)
else:
print(max(posSum,negSum))
``` | instruction | 0 | 56,844 | 12 | 113,688 |
Yes | output | 1 | 56,844 | 12 | 113,689 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3.
Submitted Solution:
```
def LOS(L):
x=min(L)
L2=[0]*(len(L))
for i in range(len(L)):
if(L[i]%2!=0):
L2[i]=L[i]
else:
L2[i]=x
for i in range(1,len(L)):
for j in range(i):
if((L[i]+L[j])%2!=0 and L[i]+L[j]>L2[i]):
L2[i]=L[i]+L[j]
return max(L2)
i=int(input(""))
L=[x for x in input("").split()]
for i in range(len(L)):
L[i]=int(L[i])
print(LOS(L))
``` | instruction | 0 | 56,845 | 12 | 113,690 |
No | output | 1 | 56,845 | 12 | 113,691 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3.
Submitted Solution:
```
n = int(input())
l = list(map(int,input().split()))
odd_pos = []
odd_neg = []
even_pos = []
even_neg = []
for i in range(len(l)):
if l[i]%2 == 1 and l[i] > 0:
odd_pos.append(l[i])
elif l[i]%2 == 0 and l[i] > 0:
even_pos.append(l[i])
elif l[i]%2 == 1 and l[i] < 0:
odd_neg.append(l[i])
else:
even_neg.append(l[i])
if len(odd_pos):
odd_pos.sort(reverse=True)
if len(odd_pos)%2 == 0:
odd_pos = odd_pos[:len(odd_pos)-1]
ans1 = sum(odd_pos)
else:
ans1 = sum(odd_pos)
if len(even_pos):
ans2 = sum(even_pos)
else:
ans2 = 0
print(ans1 + ans2)
elif len(odd_neg):
ans1 = max(odd_neg)
if len(even_pos):
ans2 = sum(even_pos)
else:
ans2 = 0
print(ans1+ans2)
``` | instruction | 0 | 56,846 | 12 | 113,692 |
No | output | 1 | 56,846 | 12 | 113,693 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3.
Submitted Solution:
```
n = int(input())
l = list(map(int, input().split()))
odd = []
even = []
for i in range(n):
if l[i] % 2 == 0:
if l[i] > 0:
even.append(l[i])
else:
odd.append(l[i])
odd.sort(reverse=True)
even.sort(reverse=True)
ans = 0
for i in range(len(odd)):
if odd[i] < 0:
if ans == 0:
ans += odd[i]
break
else:
break
else:
ans += odd[i]
print(ans + sum(even))
``` | instruction | 0 | 56,847 | 12 | 113,694 |
No | output | 1 | 56,847 | 12 | 113,695 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given sequence a1, a2, ..., an of integer numbers of length n. Your task is to find such subsequence that its sum is odd and maximum among all such subsequences. It's guaranteed that given sequence contains subsequence with odd sum.
Subsequence is a sequence that can be derived from another sequence by deleting some elements without changing the order of the remaining elements.
You should write a program which finds sum of the best subsequence.
Input
The first line contains integer number n (1 ≤ n ≤ 105).
The second line contains n integer numbers a1, a2, ..., an ( - 104 ≤ ai ≤ 104). The sequence contains at least one subsequence with odd sum.
Output
Print sum of resulting subseqeuence.
Examples
Input
4
-2 2 -3 1
Output
3
Input
3
2 -5 -3
Output
-1
Note
In the first example sum of the second and the fourth elements is 3.
Submitted Solution:
```
n=int(input())
ligne = input().split(" ")
if(n==1):
print(ligne[0])
else:
l=[]
for i in range(n):
l.append(int(ligne[i]))
maxi=max(l)
l.remove(max(l))
j=max(l)
nmaxi=maxi+j
while(maxi%2==0 or len(l)>0):
if(nmaxi>maxi):
if(nmaxi%2!=0):
maxi=nmaxi
l.remove(j)
j=max(l)
nmaxi=maxi+j
else:
l.remove(j)
j=max(l)
nmaxi=maxi+j
elif(nmaxi%2!=0):
#maxi=nmaxi
break
print(maxi)
``` | instruction | 0 | 56,848 | 12 | 113,696 |
No | output | 1 | 56,848 | 12 | 113,697 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9 | instruction | 0 | 56,849 | 12 | 113,698 |
Tags: data structures, divide and conquer, dsu, sortings
Correct Solution:
```
import sys
def pro():
return sys.stdin.readline().strip()
def rop():
return map(int, pro().split())
def a_b_b(a, b):
return a > b
def a_br_b(a, b):
return a >= b
def a_m_b(a, b):
return a < b
def a_mr_b(a, b):
return a <= b
def ry(a, c, t):
q = len(a)
x = [(t, -1)]
r = [0] * q
for i in range(len(a)):
v = a[i]
while c(v, x[-1][0]):
x.pop()
r[i] = x[-1][1]
x.append((v, i))
return r
def tyr(a):
q = len(a)
r = [0] * q
for i in range(q):
r[q - i - 1] = q - a[i] - 1
return r
def main():
s = int(pro())
a = list(rop())
z = a[::-1]
o = 10 ** 9
r = 0
m = ry(a, a_b_b, o)
n = tyr(ry(z, a_br_b, o))
b = ry(a, a_m_b, -o)
v = tyr(ry(z, a_mr_b, -o))
for i in range(s):
r += a[i] * ((i - m[i]) * (n[i] - i) - (i - b[i]) * (v[i] - i))
print(r)
main()
``` | output | 1 | 56,849 | 12 | 113,699 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9 | instruction | 0 | 56,850 | 12 | 113,700 |
Tags: data structures, divide and conquer, dsu, sortings
Correct Solution:
```
#####################################
import atexit, io, sys, collections, math, heapq, fractions,copy, os
from io import BytesIO, IOBase
##################################### python 3 START
BUFSIZE = 8192
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
def read(self):
while True:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
##################################### python 3 END
n = int(input())
ais = list(map(int, input().split()))
def build(nums,iter, flag,mark):
h = {}
stack = collections.deque([])
for i in iter:
if flag:
if mark:
while(stack and nums[stack[-1]] >= nums[i]):
h[stack.pop()] = i
else:
while(stack and nums[stack[-1]] > nums[i]):
h[stack.pop()] = i
else:
if mark:
while(stack and nums[stack[-1]] < nums[i]):
h[stack.pop()] = i
else:
while(stack and nums[stack[-1]] <= nums[i]):
h[stack.pop()] = i
stack.append(i)
return h
def f(ais,n):
ans = 0
for b in [False, True]:
left = build(ais, range(len(ais))[::-1], b, False)
right = build(ais, range(len(ais)), b, True)
'''
print ('MAX' if b is False else 'MIN',ais)
for i in range(len(ais)):
l = (i - (left[i] if i in left else -1) )
r = ((right[i] if i in right else n) - i)
q= l*r
print ('index = ', i, '|',left[i] if i in left else -1, right[i] if i in right else n, 'quantity = ', q)
'''
for i in range(len(ais)):
l = (i - (left[i] if i in left else -1) )
r = ((right[i] if i in right else n) - i)
q = l * r * ais[i]
if b:
q*=-1
ans += q
return ans
print (f(ais,n))
``` | output | 1 | 56,850 | 12 | 113,701 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9 | instruction | 0 | 56,851 | 12 | 113,702 |
Tags: data structures, divide and conquer, dsu, sortings
Correct Solution:
```
#####################################
import atexit, io, sys, collections, math, heapq, fractions,copy, os
from io import BytesIO, IOBase
##################################### python 3 START
BUFSIZE = 8192
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
def read(self):
while True:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
##################################### python 3 END
n = int(input())
ais = list(map(int, input().split()))
def check(a,b,flag, mark):
return (a >= b if mark else a > b) if flag else (a <= b if mark else a < b)
def build(nums,iter, flag,mark,dv):
h,stack = [dv for i in range(len(nums))],collections.deque([])
for i in iter:
while(stack and check(nums[stack[-1]], nums[i], flag, mark)): h[stack.pop()] = i
stack.append(i)
return h
def f(ais,n):
ans = 0
left = [None,None]
right = [None,None]
for b in [False,True]:
left[int(b)],right[int(b)] = build(ais, range(len(ais)-1,-1,-1), b, False, -1),build(ais, range(len(ais)), b, True, n)
for i in range(len(ais)):
ans += ((i - left[0][i]) * (right[0][i] - i) * ais[i]) - (i - left[1][i]) * (right[1][i] - i) * ais[i]
return ans
print (f(ais,n))
``` | output | 1 | 56,851 | 12 | 113,703 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9 | instruction | 0 | 56,852 | 12 | 113,704 |
Tags: data structures, divide and conquer, dsu, sortings
Correct Solution:
```
#####################################
import atexit, io, sys, collections, math, heapq, fractions,copy, os
from io import BytesIO, IOBase
##################################### python 3 START
BUFSIZE = 8192
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
def read(self):
while True:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
##################################### python 3 END
n = int(input())
ais = list(map(int, input().split()))
def check(a,b,flag, mark):
if flag:
return a >= b if mark else a > b
else:
return a <= b if mark else a < b
def build(nums,iter, flag,mark,dv):
h,stack = [dv for i in range(len(nums))],collections.deque([])
for i in iter:
while(stack and check(nums[stack[-1]], nums[i], flag, mark)): h[stack.pop()] = i
stack.append(i)
return h
def f(ais,n):
ans = 0
for b in [False, True]:
left,right = build(ais, range(len(ais))[::-1], b, False, -1),build(ais, range(len(ais)), b, True, n)
for i in range(len(ais)):
l = i - left[i]# if i in left else -1) )
r = right[i] - i #if i in right else n) - i)
q = l * r * ais[i]
ans += q if not(b) else -q
return ans
print (f(ais,n))
``` | output | 1 | 56,852 | 12 | 113,705 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9 | instruction | 0 | 56,853 | 12 | 113,706 |
Tags: data structures, divide and conquer, dsu, sortings
Correct Solution:
```
#####################################
import atexit, io, sys, collections, math, heapq, fractions,copy, os
from io import BytesIO, IOBase
##################################### python 3 START
BUFSIZE = 8192
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
def read(self):
while True:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
##################################### python 3 END
n = int(input())
ais = list(map(int, input().split()))
def check(a,b,flag, mark):
if flag:
return a >= b if mark else a > b
else:
return a <= b if mark else a < b
def build(nums,iter, flag,mark,dv):
h,stack = [dv for i in range(len(nums))],collections.deque([])
for i in iter:
while(stack and check(nums[stack[-1]], nums[i], flag, mark)): h[stack.pop()] = i
stack.append(i)
return h
def f(ais,n):
ans = 0
for b in [False, True]:
left,right = build(ais, range(len(ais)-1,-1,-1), b, False, -1),build(ais, range(len(ais)), b, True, n)
for i in range(len(ais)):
q = (i - left[i]) * (right[i] - i) * ais[i]
ans += q if not(b) else -q
return ans
print (f(ais,n))
``` | output | 1 | 56,853 | 12 | 113,707 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9 | instruction | 0 | 56,854 | 12 | 113,708 |
Tags: data structures, divide and conquer, dsu, sortings
Correct Solution:
```
n = int(input())
a = list(map(int,input().split()))
left = [None for _ in range(n)]
right = [None for _ in range(n)]
left1 = [None for _ in range(n)]
right1 = [None for _ in range(n)]
class Stack(object):
def __init__(self):
self.c = []
def push(self,new):
self.c.append(new)
def pop(self):
return self.c.pop(-1)
def vide(self):
return self.c == []
def peek(self):
return self.c[-1]
def calculleft():
s = Stack()
for i in range(n):
while not s.vide() and a[s.peek()] > a[i]:
s.pop()
if not s.vide():
left[i] = s.peek()
else:
left[i] = -1
s.push(i)
def calculright():
s = Stack()
for i in range(n-1,-1,-1):
while not s.vide() and a[s.peek()] >= a[i]:
s.pop()
if not s.vide():
right[i] = s.peek()
else:
right[i] = n
s.push(i)
def calculleft1():
s = Stack()
for i in range(n):
while not s.vide() and a[s.peek()] < a[i]:
s.pop()
if not s.vide():
left1[i] = s.peek()
else:
left1[i] = -1
s.push(i)
def calculright1():
s = Stack()
for i in range(n-1,-1,-1):
while not s.vide() and a[s.peek()] <= a[i]:
s.pop()
if not s.vide():
right1[i] = s.peek()
else:
right1[i] = n
s.push(i)
res = 0
calculleft()
calculright()
calculleft1()
calculright1()
res2 = 0
for i in range(n):
res += (i-left[i])*(right[i]-i)*a[i]
res2 += (i-left1[i])*(right1[i]-i)*a[i]
print(res2-res)
``` | output | 1 | 56,854 | 12 | 113,709 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9 | instruction | 0 | 56,855 | 12 | 113,710 |
Tags: data structures, divide and conquer, dsu, sortings
Correct Solution:
```
def solve1(a, n):
left=[]
st=[]
i=0
while i<(n):
if not st:
left.append(-1)
st.append(i)
elif st and a[i]<a[st[-1]]:
left.append(st[-1])
st.append(i)
else:
while st and a[i]>a[st[-1]]:
st.pop()
if not st:
st.append(i)
left.append(-1)
else:
left.append(st[-1])
st.append(i)
i+=1
# print(left)
right=[]
st=[]
i=n-1
while i>(-1):
if not st:
right.append(n)
st.append(i)
elif st and a[i]<a[st[-1]]:
right.append(st[-1])
st.append(i)
else:
while st and a[i]>=a[st[-1]]:
st.pop()
if not st:
st.append(i)
right.append(n)
else:
right.append(st[-1])
st.append(i)
i-=1
right=right[::-1]
c=0
for i in range(len(left)):
x=(right[i]-i)*(i-left[i])*a[i]
# print(x)
if x==0:
c+=a[i]
else:
c+=x
# print(c)
# print(a)
# print(left)
# print(right)
return c
def solve2(a, n):
left=[]
st=[]
i=0
while i<(n):
if not st:
left.append(-1)
st.append(i)
elif st and a[i]>a[st[-1]]:
left.append(st[-1])
st.append(i)
else:
while st and a[i]<a[st[-1]]:
st.pop()
if not st:
st.append(i)
left.append(-1)
else:
left.append(st[-1])
st.append(i)
i+=1
# print(left)
right=[]
st=[]
i=n-1
while i>(-1):
if not st:
right.append(n)
st.append(i)
elif st and a[i]>a[st[-1]]:
right.append(st[-1])
st.append(i)
else:
while st and a[i]<=a[st[-1]]:
st.pop()
if not st:
st.append(i)
right.append(n)
else:
right.append(st[-1])
st.append(i)
i-=1
right=right[::-1]
# print(right)
c=0
for i in range(len(left)):
x=(right[i]-i)*(i-left[i])*a[i]
# print(x)
if x==0:
c+=a[i]
else:
c+=x
# print(c)
# print(a)
# print(left)
# print(right)
return c
n=int(input())
arr=[int(x) for x in input().split()]
# solve2(arr,n)
# print((solve1(arr,n)))
# print((solve2(arr,n)))
# print()
# (solve2(arr,n))
print(solve1(arr,n)-solve2(arr,n))
``` | output | 1 | 56,855 | 12 | 113,711 |
Provide tags and a correct Python 3 solution for this coding contest problem.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9 | instruction | 0 | 56,856 | 12 | 113,712 |
Tags: data structures, divide and conquer, dsu, sortings
Correct Solution:
```
#Bhargey Mehta (Sophomore)
#DA-IICT, Gandhinagar
import sys, math, queue, bisect
#sys.stdin = open("input.txt", "r")
MOD = 10**9+7
sys.setrecursionlimit(1000000)
def mxr(x):
n = len(x)
ans = [n for i in range(n)]
s = []
for i in range(n):
while len(s) > 0 and x[s[-1]] < x[i]:
ans[s.pop()] = i
s.append(i)
return ans
def mnr(x):
n = len(x)
ans = [n for i in range(n)]
s = []
for i in range(n):
while len(s) > 0 and x[s[-1]] > x[i]:
ans[s.pop()] = i
s.append(i)
return ans
def mxl(x):
ans = [-1 for i in range(n)]
s = []
for i in range(n-1, -1, -1):
while len(s) > 0 and x[s[-1]] <= x[i]:
ans[s.pop()] = i
s.append(i)
return ans
def mnl(x):
ans = [-1 for i in range(n)]
s = []
for i in range(n-1, -1, -1):
while len(s) > 0 and x[s[-1]] >= x[i]:
ans[s.pop()] = i
s.append(i)
return ans
n = int(input())
a = list(map(int, input().split()))
maxl = mxl(a)
minl = mnl(a)
maxr = mxr(a)
minr = mnr(a)
ans = 0
for i in range(n):
mxrng = (maxr[i]-i)*(i-maxl[i])
mnrng = (minr[i]-i)*(i-minl[i])
ans += a[i]*(mxrng-mnrng)
print(ans)
``` | output | 1 | 56,856 | 12 | 113,713 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9
Submitted Solution:
```
import os
import sys
from io import BytesIO, IOBase
BUFSIZE = 8192
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
def read(self):
while True:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
#'%.9f'%ans
#sys.stdout.flush()
##########################################################
def add(i,val):
while i<=n:
bit[i].append(val)
i+=(i&-i)
def query(i,x):
s=0
while i>0:
pos=bisect.bisect_left(bit[i],x)
s+=len(bit[i])-pos
i-=(i&-i)
return s
from collections import Counter
import math
bit=[]
import bisect
from string import ascii_lowercase as al
#for _ in range(int(input())):
#n=int(input())
#n, k = map(int, input().split())
#arr=list(map(int,input().split()))
#for i in range(n):
def find_max():
left=[]
st=[]
for i in range(n):
if not st:
left.append(-1)
st.append(i)
elif st and arr[i]<arr[st[-1]]:
left.append(st[-1])
st.append(i)
else:
while st and arr[st[-1]]<=arr[i]:
st.pop()
if not st:
st.append(i)
left.append(-1)
else:
left.append(st[-1])
st.append(i)
right=[]
st=[]
for i in range(n-1,-1,-1):
if not st:
st.append(i)
right.append(n)
elif arr[i]<arr[st[-1]]:
right.append(st[-1])
st.append(i)
else:
while st and arr[st[-1]]<arr[i]:
st.pop()
if st:
right.append(st[-1])
st.append(i)
else:
right.append(n)
st.append(i)
ans=0
right=right[::-1]
for i in range(n):
x=(right[i]-i)*(i-left[i])*arr[i]
if x==0:
ans+=arr[i]
else:
ans+=x
return ans
def find_min():
left = []
st = []
for i in range(n):
if not st:
left.append(-1)
st.append(i)
elif st and arr[i] > arr[st[-1]]:
left.append(st[-1])
st.append(i)
else:
while st and arr[st[-1]] >=arr[i]:
st.pop()
if not st:
st.append(i)
left.append(-1)
else:
left.append(st[-1])
st.append(i)
right = []
st = []
for i in range(n - 1, -1, -1):
if not st:
st.append(i)
right.append(n)
elif arr[i] > arr[st[-1]]:
right.append(st[-1])
st.append(i)
else:
while st and arr[st[-1]] >arr[i]:
st.pop()
if st:
right.append(st[-1])
st.append(i)
else:
right.append(n)
st.append(i)
ans = 0
right = right[::-1]
for i in range(n):
x = (right[i] - i) * (i - left[i])*arr[i]
if x == 0:
ans += arr[i]
else:
ans += x
#print("%%%%%%%%%%%%")
return ans
n=int(input())
arr=list(map(int,input().split()))
print(find_max()-find_min())
``` | instruction | 0 | 56,857 | 12 | 113,714 |
Yes | output | 1 | 56,857 | 12 | 113,715 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9
Submitted Solution:
```
#####################################
import atexit, io, sys, collections, math, heapq, fractions,copy, os
from io import BytesIO, IOBase
##################################### python 3 START
BUFSIZE = 8192
class FastIO(IOBase):
newlines = 0
def __init__(self, file):
self._fd = file.fileno()
self.buffer = BytesIO()
self.writable = "x" in file.mode or "r" not in file.mode
self.write = self.buffer.write if self.writable else None
def read(self):
while True:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
if not b:
break
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines = 0
return self.buffer.read()
def readline(self):
while self.newlines == 0:
b = os.read(self._fd, max(os.fstat(self._fd).st_size, BUFSIZE))
self.newlines = b.count(b"\n") + (not b)
ptr = self.buffer.tell()
self.buffer.seek(0, 2), self.buffer.write(b), self.buffer.seek(ptr)
self.newlines -= 1
return self.buffer.readline()
def flush(self):
if self.writable:
os.write(self._fd, self.buffer.getvalue())
self.buffer.truncate(0), self.buffer.seek(0)
class IOWrapper(IOBase):
def __init__(self, file):
self.buffer = FastIO(file)
self.flush = self.buffer.flush
self.writable = self.buffer.writable
self.write = lambda s: self.buffer.write(s.encode("ascii"))
self.read = lambda: self.buffer.read().decode("ascii")
self.readline = lambda: self.buffer.readline().decode("ascii")
sys.stdin, sys.stdout = IOWrapper(sys.stdin), IOWrapper(sys.stdout)
input = lambda: sys.stdin.readline().rstrip("\r\n")
##################################### python 3 END
n = int(input())
ais = list(map(int, input().split()))
def check(a,b,flag, mark):
return (a >= b if mark else a > b) if flag else (a <= b if mark else a < b)
def build(nums,iter, flag,mark,dv):
h,stack = [dv for i in range(len(nums))],collections.deque([])
for i in iter:
while(stack and check(nums[stack[-1]], nums[i], flag, mark)): h[stack.pop()] = i
stack.append(i)
return h
def f(ais,n):
ans = 0
left = [None,None]
right = [None,None]
for b in [False,True]:
left[int(b)],right[int(b)] = build(ais, range(len(ais)-1,-1,-1), b, False, -1),build(ais, range(len(ais)), b, True, n)
for i in range(len(ais)):
q = (i - left[0][i]) * (right[0][i] - i) * ais[i]
ans += q
q = (i - left[1][i]) * (right[1][i] - i) * ais[i]
ans -= q
return ans
print (f(ais,n))
``` | instruction | 0 | 56,858 | 12 | 113,716 |
Yes | output | 1 | 56,858 | 12 | 113,717 |
Evaluate the correctness of the submitted Python 3 solution to the coding contest problem. Provide a "Yes" or "No" response.
You are given an array a consisting of n elements. The imbalance value of some subsegment of this array is the difference between the maximum and minimum element from this segment. The imbalance value of the array is the sum of imbalance values of all subsegments of this array.
For example, the imbalance value of array [1, 4, 1] is 9, because there are 6 different subsegments of this array:
* [1] (from index 1 to index 1), imbalance value is 0;
* [1, 4] (from index 1 to index 2), imbalance value is 3;
* [1, 4, 1] (from index 1 to index 3), imbalance value is 3;
* [4] (from index 2 to index 2), imbalance value is 0;
* [4, 1] (from index 2 to index 3), imbalance value is 3;
* [1] (from index 3 to index 3), imbalance value is 0;
You have to determine the imbalance value of the array a.
Input
The first line contains one integer n (1 ≤ n ≤ 106) — size of the array a.
The second line contains n integers a1, a2... an (1 ≤ ai ≤ 106) — elements of the array.
Output
Print one integer — the imbalance value of a.
Example
Input
3
1 4 1
Output
9
Submitted Solution:
```
#taken from https://stackoverflow.com/questions/30698441/optimal-way-to-find-sums-of-all-contiguous-sub-arrays-max-difference
def max_sums(d):
stack = [(-1, float('inf'))]
sum_ = 0
for i, x in enumerate(d):
while x > stack[-1][1]:
prev_i, prev_x = stack.pop()
prev_prev_i, prev_prev_x = stack[-1]
sum_ += prev_x * (i - prev_i) * (prev_i - prev_prev_i)
stack.append((i, x))
while len(stack) > 1:
prev_i, prev_x = stack.pop()
prev_prev_i, prev_prev_x = stack[-1]
sum_ += prev_x * (len(d) - prev_i) * (prev_i - prev_prev_i)
return sum_
def max_differences_sum(d):
return max_sums(d) + max_sums([-x for x in d])
n=int(input())
l=list(map(int,input().split()))
print(max_differences_sum(l))
``` | instruction | 0 | 56,859 | 12 | 113,718 |
Yes | output | 1 | 56,859 | 12 | 113,719 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.