We invite you to participate in CodeChef’s Starters142, this Wednesday, 10th July, rated for till 5-Stars(ie. for users with rating < 2200).
Time: 8:00 PM — 10:00 PM IST
Joining us on the problem setting panel are:
Setters: Rachit knightFighterr Choudhary, Sezal sezal Mittal, Souradeep Paul, Archit Pols_Agyi_Pols, Bernard BERNARD Ibrahimcha, Amr Amrharb Harb.
Tester:Manan mexomerf Grover.
Text Editorialists:Nishank IceKnight1093 Suresh.
Statement Verifier:Kanhaiya notsoloud1 Mohan.
Contest Admin : Yash yash_daga Daga.
Note: Some problems have subtasks
Written editorials will be available for all on discuss.codechef.com. Pro users can find the editorials directly on the problem pages after the contest. The video editorials of the problems will be available only to Pro users.
Also, if you have some original and engaging problem ideas, and you’re interested in them being used in CodeChef's contests, you can share them here. Hope to see you participating.
Good Luck!
Hoping no problems will be there containing Game Theory. IMHO, CC should give Game Theory some rest.
stfu no
BERNARD orz
Another week another contest not rated for 7* coders :sob;
Suffering from success
Suffering from success
Amr Harb is the best.
Whenever you have subtasks you must have separate samples where sample 1 fits the constraints of subtask 1, sample 2 fits the constraints of subtask 2, etc.
And explanations of samples must be after all tests.
SpeedChef
Was anyone able to solve Sum of Xors(SMXR)?
I thought it would be just (N*M*3)/2 but that's wrong, can someone please explain?
Try to think when K=6 and N=M=4.
That was such a random guess, it is not a coding problem.
What will happen when K is of 4*x+2 form? Do you think this is random guessing?
I looked at $$$N=M=4$$$, $$$K=6$$$ and immediately guessed that $$$F(K)=0$$$ for all $$$K$$$ except $$$2$$$ and $$$NM-2$$$, without proving that such a 'closed tour' actually exists.
As a matter of fact, one does not need a closed tour of alternating row and column moves for the construction, which further shows that my guess was not supported by much thought. For example, $$$N=M=6$$$, $$$K=10$$$ has a construction that is not a connected tour:
That may be your experience.
You know most people are guessing and not even trying to prove when accuracy is 8.2% (Div. 1), 1.23% (Div. 2), 0.6% (Div. 3), and 0.84% (Div. 4).
The comment below shows the correct arrangement not the one I cooked up.
Why is the sample test cases promoting the answer (n*m*3)/2? It would have been better if you had included similar corner cases in the samples.
Sample test cases are made to explain the problem statement, not to include corner cases.
You can get $$$F(X) = 0$$$ for $$$K \equiv 2 (\mod 4)$$$ for some constructions.
For example, consider $$$N = 4, M = 4, K = 6$$$
Oh, thank you very much, now I understand
[WRONG]
Sorting the array completely breaks the number of $$$[L, R]$$$ that satisfy the equation.
For example consider the input:
$$$n = 6, x = 1, A = [8, 8, 4, 4, 8, 8]$$$
In this array, only $$$i = 3, j = 4$$$ satisfy the equation, so $$$F(L, R) = 1$$$ for all $$$1 \leq L \leq 3, 4 \leq R \leq 6$$$ giving an answer of $$$9$$$.
However, if you sort the array, it becomes $$$[4, 4, 8, 8, 8, 8]$$$ which has $$$F(L, R) = 1$$$ for $$$L = 1, 2 \leq R \leq 6$$$ giving an answer of only $$$5$$$.
Thanks!
Not a fan of "knowledge-based" problems like "Number Hunt". Its pretty much just "Do you know that the prime gap is small even at $$$10^9$$$?".
Maybe it makes sense to put it in Starters since they are meant to be "educational" contests, but it still feels out of place.
Yeah, I added these constraints because I found this fact very interesting, when I got to know about it as a beginner. So thought it might be educational for starters.
Did you know that a twin prime gap is also small? https://leetcode.com/problems/closest-prime-numbers-in-range/solutions/2979147/python-twin-primes-no-sieve/
I think anyone who solved this problem before can solve today's sum of Xor in 2-3 minutes
Today's E (LENGTHX) , its editorial was based on segment trees , any suggestions where to learn seg trees from ?
For xor sum , to prove all even $$$i \gt 4$$$ and $$$i \lt NM - 2$$$ , $$$F[i] = 0 $$$ , you can first construct $$$K = 6.$$$ However , it may get confusing when many $$$1s$$$ have been filled , to solve that case , observe that You can invert all $$$1s$$$ to $$$0$$$ and $$$0s$$$ to $$$1$$$ , and $$$F[i]$$$ remains same. Hence the case $$$K = 6$$$ is same as $$$K = NM - 6$$$ .
include <bits/stdc++.h>
using namespace std;
int main() { int t; cin >> t; while (t--) { int n; cin >> n; vector v(n); int y = 0; for (int i = 0; i < n; ++i) { cin >> v[i]; y = y | v[i]; }
}this code is for 4th que of div4 in codechef(array removal) and it got accepted in codechef but this code gives wrong output for the input(n=4) {2,8,9,10}
are u gonna do anything about wrong submissions that got accepted for array removal problem,it had massive impact in rankings of div2
Cheaters Cheaters everywhere :
https://www.youtube.com/watch?v=lroJG3QAcRA
Today I was hoping to turn three star on Codechef but thanks to them
Sorry for mentioning you but the tester's code given for the problem
Pairs Sum with Length x
in today's contest does not work. Please fix it.Submission link
mexomerf IceKnight1093.
There must be a mistake, this solution is for a prev version of the problem
Number Hunt Detailed Video Editorial
https://youtu.be/fhOLz-Or9NA