Hello, Codeforces! We are pleased to invite you to Codeforces Round 957 (Div. 3), which will be held on Jul/11/2024 17:35 (Moscow time).
The round will be hosted by rules of educational rounds (extended ICPC). You will be offered 7 problems, which I with Noobish_Monk, ErnKor, Kmes and ArSarapkin have prepared for you.
The rules of the third division are:
The penalty for a wrong submission is 10 minutes.
12-hour phase of open hacks after the end of the round (hacks do not give additional points)
after the end of the open hacking phase, all solutions will be tested on the updated set of tests, and the ratings recalculated.
Reminding you, that only trusted participants of the third division will be included in the official standings table. This is a forced measure for combating unsporting behavior. To qualify as a trusted participant of the third division, you must:
take part in at least five rated rounds (and solve at least one problem in each of them),
do not have a point of 1900 or higher in the rating.
Regardless of whether you are a trusted participant of the third division or not, if your rating is less than 1600 (or you are a newcomer/unrated), then the round will be rated for you.
We would like to thank everybody, who helped in preparing this round:
Our coordinators, Gornak40 and Vladosiya for a wonderful coordination!
MikeMirzayanov for wonderful Polygon and charming Codeforces!
Vladosiya for deleting the lore in almost all of the problems...
Our numerous testers: natalina, FairyWinx, Kosya, ErnKor, 74TrAkToR, ArSarapkin, mainyutin, senjougaharin, gaiercop, mikeamir0310, Maksim_Musikhin, Arina77, artimovich.ko, main-scobik, savanzh, petr_davydov, Pronkin_dv, kalita.georgy, kuznetsov.da, s_w_a_n_i_k, Keopint, VolkovMA, kuewa, cry, Harsh_kunwar, Kon567889, dshindov, OAleksa, SajidAbdullah, fishcurry, george_the_piggy, k1sara and many others, who didn't mention their handle.
Special thanks for KEFedorov, for inviting more than 35 testers.
Good luck!
UPD1 After the end of the round, Shayan will broadcast the solutions, a recording will also be available, but we will also prepare a textual editorial.
UPD2 Video Discussion by Shayan
UPD3 Video of solving the round + Editorial by the "winner" of this round, neal
UPD4 Editorial.
UPD5 Congratulations for the winners:
Official:
Official + Unofficial:
cool
my goal to reach +1700 after this round
Me too
Good luck!!! My goal is to reach 1200-1300 in the next weeks.
My goal is to participate in the contest.
Downvoted to this extent! So much jealousy and competition here!
Hoping for the best Div.3 Round! <3
all the best, I hope i reach 1300+
__ combating unsporting behavior. To qualify as a trusted participant of the fourth division, you must: __ take part in at least five rated rounds (and solve at least one problem in each of them), __ do not have a point of 1900 or higher in the rating. __ Regardless of whether you are a trusted participant of the fourth division or not, if your rating is less than 1600 (or you are a newcomer/unrated), then the round will be rated for you.
why is it fourth division
Copy-pasting issues
Finally I can write this: my first unrated div3 :D
so people rated above 1600 wont be placed on leaderboard ?
They wont be placed on official leaderboard
How I used to think , it will be div 3 which will carry me to expert, but the div 2 hard carried man
Me too:)
As a tester of this round, I will say that the tasks were interesting
Hoping to get a good delta positive after this round! I wish everyone the same!
do we know each other?
Just curious: How do you guys propose a Div3 contest?
only people from saratov university can do this
Hoping we all get positive delta :)
Hoping we all learn something regardless of delta :). (No offense, but if someone gets positive delta, someone else won't get positive delta)... I like being realistic.
ig you dont know the saying "May you all win"
not strong enough compared to you :)
As a tester i confirm that problems are indeed cool and worth solving.
Harsh_kunwar orz
finally a div3 after a long time.
Can you tell me how to propose div3 contests? Vladosiya, Gornak40 ?
Vladosiya please check dm.
Thank you.
what!!!? I love the lore.
My goal is to solve atleast 5 questions. Let's hope for the best.
f(n)=(n−1)⋅(f(n−2)+f(n−1))
can we solve this eqn using matrix expo???
if yes then plzz give me the matrix...
this is the recursion for the number of derangements of n
help me if u know...
can you send the problem link plzz?
https://cses.fi/problemset/task/1717/
How do you do this ? I have been stuck for 2 hours, I have a formula that can calculate answer for all numbers till n in O(n2) time, but that will give TLE
f(n)=(n−1)⋅(f(n−2)+f(n−1)) use this..
I think you need this video.
I like the monk's organization :D
Hoping a fresh and exciting contest
I want to go specialist, Good luck for me
chat do i rise back up to pupil
UPD: I think I do
Shayan streams pretty much in recent month for problems' solution. I appreciate him a lot.
Hope for rating++;
As a tester, I can guarantee that the round will be interesting!
potato
I want to solve 3 questions in this contest :)
gl!
thanks, gl to you as well!
How to be tester? Is this random?
No, it's not random. Most testers are friends of setters or coordinators. Yes, there are some "random" testers if you want to say it like that, but in a way that they are friends of friends.
OK, thank you ;)
As a problemsetter, good luck, guys!
As a tester of this round, I can say that this round will probably lower your rating, but the tasks are worth solving <3
bro your comment scared me ::skull::
:skull:
Skill issues
Really???
straight on the face, dem bro ;(
Can anyone tell me how much problem should I solve by looking at my profile and of what rating in order to get to pupil?
a general doubt, penalty gets applied after correct submisson or doen't depend on correct submission?
Penalty gets applied after correct submission.
what if I submitted two correct submissions to a problem what is there any penalty on this??
No there is no penalty for that.
time to lock in
good luck guys :)
Great! Looking to get a fat positive (will get the opposite).
I hope I'll be +1300 after this round! Good luck to everyone!
good lucky for all !
i wish i'll reach back to 1300
Looking forward to getting my first color today!
will trusted participant be rated ??
or just be included in the official standing ?
Regardless of whether you are a trusted participant of the third division or not, if your rating is less than 1600 (or you are a newcomer/unrated), then the round will be rated for you.
If the rating range is 1600-1900(didn't cross 1900 at any time), then you would only be in the official standings.
Btw, Congrats!! on reaching Expert
Oh , i didn't see it
Thx : )
I hope to go forward and wish for this to be the best Div3. Odin is with us!!!
Looking forward to getting a good positive delta this round
My goal above 1200 in this round
Hoping For a +ve delta this round and learn smtng neww....
interesting problems... i solved 3 in 40 minutes :)
that went a bit better than last div2 (emphasis on a bit because i somehow crashed my ide 4 times which lost me about 4-6 minutes from E-G)
NICE contest, at least for me its a good div 3, i solved 3. some late submission because of cloudflare but it was good overall.
got stuck on F , ughhhhhhhhhhhhhhh.
Me too :(
:(
Overall very nice problems and good contest. My first time solving E and hopefully I get a good delta :)
is it just brute force ? 1e4*1e4 *100 it should not pass obviously
here's an oberservation: n*a-b can only be at most 7 digits which means b can only have at most 7 values for each a
well nope, brute force is TLE, its basically going over all values of A (10^4 iterations) and the it is guaranteed that the length of a valid string after removing b bits from it will never exceed a length of 7 (cuz the constraints come into play) so for every A , you run 7 inner loops and then obviously you can find b in constant time using a formula.
270014350
you can also solve this with math i believe — start with an empty string (denoted as s), add the digits of n one-by-one to it, then solve the system of equations:
(int)s = n * a - b
numdigits(n) * a - b = numdigits(s)
270034846
ohh, thanks for this :) . During the contest, i was constantly trying to think in this direction cuz this is much more efficient than what i wrote but yeah my brain just didnt work. Thanks for the soln, appreciated.
lol I did this but it messes up the case n=1 where solving for a has “infinite”(a up to 10,000 from constraint) solutions. I basically cried after the contest…
how is it 1e4*1e4*100 in brute force....??? beacuse after these 3 loops for(n>9) we have to convert it into string then add it to string for a times and pop back it for b times ....to get the final string that we convert into integer of long long??
are you thinking to pre calculate a-b or a-b/2 or a-b/3 respectively for single digit ,double digit and triple digit......?????
C is the easiest but it was the hardest for me to write
why was it hard to write? its just printing 2 loops ?
I know but I struggled with them and before that I struggled with the math notations
269928597
:)
ignore this was my worst performance soo far
Probably Cheating
idk, D is just case handling, and E is running a loop on A and computing B accordingly. As a pupil , D was simple but E took some time
may be this time my brain stopped working possible after getting 3 penalties on A
3 pens on A is ROUGH tbh :)
us bro us... similar thing happened with me in D ..after which i just gave up as my brain froze.....I thought E and F would be hard to make it even though i had 50 min left......It was after the contest when I solved E and F ...and since then I'm cursing myself why didn't I moved to E and F
D was only matter of implementation, Russian cper have quite different sense of indexing something.
like if there is single 'W' between 'L' — "LWL". most people probably would think that simply jump over 'W' without swim, no it wasn't you need to jump first and then swim 1 kek.
wasnt it stated you start at n = 0 (ok to be fair it isnt stated the rest of the inputs are [1..n]) (but it doesnt matter that much)
most prolly mass cheating in D
How did you get TLE,WA on A despite being a 1550?
How to solve F?
set trivializes, only add if its a divisor of x, max 36 elements in the set so it wont tle
How did I not see this..
wow so i was in the right direction just less time ughhh. Used sets but couldnt write correct code fack
How do you come up with 36?
bruh 1e5 has 36 factors
but 90000 has 74 factors lol
i dont think that matters, simply going over every factor that can be created does run well in time. 270114245 . And now i feel sed cuz i could have gotten this during contest :(. I wrote the same thing at night just missed the s.insert(1) line during reset of the set
Actually number of divisors of $$$M$$$ is bounded by $$$2\sqrt M$$$ (think of the way we find divisors) In this case the overall time complexity would be $$$O(N\sqrt N)$$$ for $$$N \sim 10^5$$$ which's fast enough
ig for all practical purposes N^1/3 can be used as an upper bound for number of divisors
I literally cannot believe on how ChatGPT unable to solve stupid implementation problem like D.
it's case handling and reasoning and greedy thinking I think it's very hard for an ai
why would you want chat gpt to solve for you?
Hi All, can anyone help me understand why I am getting memory limit exceeded on problem D? this is my submission, I followed a recursive approach to explore possible options of jumping and swimming, I am not using an extra data structure, thank you in advance for your help!
try passing the string by reference
Just tried it here, I got time limit exceeded this time; thank you for your help!, I will think on another approach.
Recursion might not pass with the constraints. I think solving the problem greedily is definitely more straightforward and easier to implement.
Was it only for me ,or it was really hard to decode the problems?
oh yeah. very weird descriptions. probably their native language is not English.
None of codeforces, codechef, atcoder speak English natively. Poor preparation is the only reason behind bad statements.
Another round dusted by Indian cheaters.
Me staring at E for 1hr +
My brain got blown away after reading E was like another level for me.DIV3's have become harder now
Sharing my video of winning* the contest: https://www.youtube.com/watch?v=9Vv2ZukG1CM&t=2476s
(*tourist doesn't count)
Congrats
When tourist enters a contest, everyone aims for the second place.
Great job, I also needed the video. It lightened my day to see where you started it. Also listening to explanations on E-G.
Missed E, and actually missed the first time that I could clear Div3.
That case n = 1 lol XD
Yeah, I've just checked it. My solution deserves to get WA because of my stupidity while not considering all the cases =((
btw big fan, i am trying to solve your previous problems when you were pupil/specialist for practice :)
its okay bro. I got E in like 1hr20 min but i got 2 WA bcoz of fact that string cant be empty. I was so tensed for next 40 minutes and actually we have just 100 different cases so i even tried to find where i was wrong until i read string cant be empty. By that time it was already 2hr10min and penalty was too harsh + no time for F
your rating does not get reevaluated on this right?
Only participants < 1600 are rated in div3. I did it "out of competition" so my rating isn't affected.
oh, thanks for info :)
Can you please review this and help me where i am going wrong. Thanks in advance!
Ahh, D was so straightforward but I spent like an hour debugging just to get AC in 3 minutes after reimplementing my idea.
What is the thing you have to notice to solve E?
The only way I would be able to solve (I didn't do this) would be to precompute all of n from 2 to 100 and then have a python program just return those precomputed solutions...
How do we solve for B given we check for values of A... Maybe I wasn't on the right track at all for optimal solution though...
Notice that B is also bounded by 10000, so the length of good results is bounded by len(10000*100-1) = 6. So you can simply iterate all strings of length 1-6 for each A
Easier than recent div4 round!
true only was able to solve 4 in that. though i have become better :)
The problems were great, but the confusing statements ruined it for me, I really think more attention should be given to making the statement clear and understandable.
For example in problem D it was not clear that k was for the entire problem, when I read it I thought he could not swim a distance more than k in a row which led to me writing a completely different solution and wasted a lot of time on this.
I spent more time understanding the problems than solving them.
I am not saying the contest is bad, but I felt this should be addressed.
Same, initially i also thought it should be k consecutive rows. Luckily i decided to read samples where answer is NO and one of the test case explained where K need not be in a row
I think you just misread they wrote that k is for the entire swim from 1...n
The river is very cold. Therefore, in total (that is, throughout the entire swim from 0
to n+1
)
Worst Problem was D . First time i solved so poor in DIV3 .
The first line of each test case contains three numbers n,m,k(0≤k≤2⋅105,1≤n≤2⋅105,1≤m≤10) — the length of the river, the distance ErnKor can jump, and the number of meters ErnKor can swim without freezing.
I read n then m then k and hence exchanged k's constraint with m's constraint . Stucked in the whole contest for more than 1 hour while solution is correct.
https://codeforces.net/contest/1992/submission/270048258
Samee
same
i mentioned this issue in testing, but unfortunately they did not take it into account
yikes
Actually I also choked because of this , but I also confused about the fact that how to solve this if the task is to compute if swimmer can swim the river if he can continously swim for only k units
. Please explain you solution for this above wrongly interprated question . Thank you
If you are asking about my solution if you could swim for at most k meters continuously, Here is what I tried to do (Not sure if it's correct or not)
I used a prefix sum array to count the number of crocodiles, there is a crocodile between l and r if
pre[r] - pre[l - 1] > 0
I used a variable to store where the last log was, If there is water at the current index I check if the distance between it and the last log + m is less than or equal k and that there is no crocodile between log + m and the current index then the current index is reachable.
The rest is similar to the solution of problem D
Ya make sense , It will pass thank's bro
There are so many similar solutions for problem D which have been taken from youtube. They all have just changed variable names while initialising a variable name with m-1 and incrementing a variable if there is water and using some other variables and at last they are checking if variable value is not zero then print yes else no. Loads and Loads of solution in the same format, I had a brainrot seeing 10k+ submissions for D itself.
same I was not able to solve it and D got 10k submission is just blatant cheating D doesn't get 10k+ submission on div 4 how can it get here
I might have gotten E in contest time itself but cloudflare had other plans.
Am I really dumb as from submission its looking like D is easy but I was not able to solve it
D was ezz, not 10K easy tho :)
Yeah 5k submission I can understand as I am dumb but 10k is just crazy I am getting -ve delta for sure :(
well cheaters ruin it for actual hard-workers and thats pretty pathetic, like when i was giving div 3's 7-8 month ago and i was rated 900, if i solved 3 i would surely get +ve, but now its sort of impossible, either you get till E to get +ve or else you get fcked
bruh i asked chatgpt to solve D with dp and it actually worked
huh?? gpt might not even do it with a simple loop, how did DP work ? lmao
there is a valid dp solution
i know that brother, just stunned that gpt who cant write conditions in loop did it with DP, impressive to say the least lol
i think chatgpt is just better at dp lol
thank you all for the great contest! I had so much fun solving a, b but got stuck on c :(
how brute-force round ,I love it.
Please report this guy. He is sharing answer during the contest time. He is destroying the cp environment. here is the link : https://www.youtube.com/live/frrrP-SHXWA?si=nnX4Ghj0bqvDJEeN
Hi, I am a newcomer here. I solved A and B in this contest but I still did not get any rating. Can anyone help me out and let me know why is that the case?
There is hacking phase running in this contest, you can see in the standings the time for it. You can read more about it here, once this hacking phase finishes all submissions will be run once again with new test cases that are added during the hacking phase; after it you will be able to see your rating.
Fantastic round! E was an amazing problem! Thanks a lot for the round Noobish_Monk ErnKor Kmes ArSarapkin and all testers!
there is a question in cf round 957 div3 just completed question no. E Novie Mistake if n==1 for a=3 and b=2 means string is empty value of n*a-b=1 but string is empty means it is zero it is not matching then why question consider string is empty as 1 and it didn't mention anywhere
i got it my mistake
D has a very clean DP solution.
Let $$$dp[i]$$$ denote the minimum number of swims required to reach index $$$i$$$. The transitions are -
For $$$j = i + 1 - m$$$ to $$$i$$$
$$$dp[i + 1] = min(dp[i] + 1, dp[i + 1)$$$, if $$$s[i] =$$$ 'W'
Finally, the answer is YES if $$$dp[n + 1] \le k$$$.
Submission Link: https://codeforces.net/contest/1992/submission/270032188
Yeah, that's true, but I exceeded the memory limit during the contest using dynamic programming, so I used BFS instead. It is also an easy solution.
It's funny because I exceeded the memory limit using a DFS solution, and had to think of a DP solution instead.
both the dp and non dp should be O(N*10) solutions right !
why that extra 10? If you are checking for every possible m , you dont need to.
269966680
Yes, they are both $$$O(n * m).$$$
there is also easier greedy solution using O(1) extra memory: create a variable
last_land_point
and set it =i
whenever you get on the land. Then if everlast_land_point
become too far, that is farther than jumping distance, start incrementing a variable for how much to swim. If you hit a crocodile while swimming, not possible. If you have to swim too much, not possible.Code: https://codeforces.net/contest/1992/submission/269973081
That's great, bro! I liked your solution.
Hi, I struggle with tabulation DP, can you share any resource from where I can improve ?
Tabulation DP is pretty much the same as recursive DP, you just have to calculate the DP values in the correct order, that is, the current state being calculated should not depend on a future state. Honestly the best way to learn tabulation DP is to solve more problems, CSES DP section the Atcoder DP contest helped me a lot.
I would suggest Striver's DP playlist for that. If you just want to convert your recursive soln. to Tabulation then watch his videos(any of them), you will get to know that the only thing which is required for conversion is how to convert your Base cases of recursion to initialisation (base condition) of tabulation dp, rest all is just copy paste of the transition.
And this conversion is quite easy.
Can someone pls tell why my soln for D is giving TLE 270051817. I was trying to keep track of the indexes of logs, i tried to check if we could cover the distance with jump alone or jump+swim but i dont know why my code is giving tle. Can anyone pls help!
I submitted almost correct solution for E during the contest https://codeforces.net/contest/1992/submission/270016302, the only thing I didn't figured out was to convert to long long instead of size_t in the only place to avoid overflow. So disappointed, I will NEVER use any integer types except long long in competitive programming again!
Can somebody explain what's f(x) and g(x) are in '**C**', i just guessed the soln, by looking at the samples and it worked but i have no idea what is problem saying and want me to do!
g(i): Add up numbers in the first i positions of the list that are less than or equal to m. f(i): Add up numbers in the first i positions of the list that are greater than or equal to k.
To maximize the value of (∑ni=1f(i)−∑ni=1g(i)), place the largest numbers (greater than or equal to k) at the beginning of the permutation to maximize f(i), and place the smallest numbers (less than or equal to m) towards the end to minimize g(i).
// Can anybody tell me what's wrong in this implementation of D , I have chosen a variable pos to see if it is possible to pass the river and p tells if we found a "log" , c tells us if i found a "crocodile" and "w" tells us swimming through water i have as usual checked if if it is possible to get to a log then go there otherwise increase the position by m and check if there is one segment to swim to otherwise if only crocodiles were found btw i and i+m segment make pos false.
With this number of if-else-s and no proper indentation, it's hard to get the idea implemented right. The right thing to do is perhaps not to find a particular bug, but to take a step back and observe what can be changed globally. Consider developing these habits:
formatting the code to aid in implementation (any decent IDE has a keyboard shortcut for that)
thinking about solutions with as little if-else-if-if-else... forests as possible
Never thought i will get 4 wrong submission on Div3-A :(
However solved B,C easily but spent too much time on D and didn't got much time to solve E Seems lots of people solved D so i might get -ve delta
I hope we get +50 delta my rating is same as yours and the same situation it was actually a speed matter because people placed at 2500 also solved 4 problems only
I think you will reach pupil easily because you didn't participate in too much contests I participated in more contests so I would be relieved if I got those +50
nah bro just got +7, i think lot of incorrect submission for problem A cost me
Bug in https://m1.codeforces.com/enter (during contest Problem C was not able to load i shifted to main site quickly) Your text to link here...
Can someone please help me debug this code for D ? I was able to come up with the dp idea but can't find what is wrong with the solution? https://codeforces.net/contest/1992/submission/270037120
Also, Is there some way to see the test case on which it is failing? Thank you.
Just output the testcase where it is failing.
Your solution gives WA on test 2. Keep a flag variable adjust it accordingly so that test case 1 passes. Now for test case 2 do not output anything, and simply output the input for the token no. where it is wrong.
270077042
1 2 3 0 CW
Probably, the solution is executed with error 'out of bounds' on the line 56. It is because dp is declared as a vector of size n but when m > n it is out of bounds access and it gives wrong answer, however locally it was giving correct answer.
Ah Man, Thank you very much!!. Wasted 2 hours on this during the contest. But it should give some Runtime Error(Segmentation fault ) instead of Wrong Answer, Correct? Or am I missing something here?
Anyways it was an easy problem, more than 10k submissions :O . Will get a negative delta because of this :'(
Can you please tell the wrong test case for my code. I have been getting WA on case 559 of test 4 270114576.
1
7 2 2
WWLLLWW
270126212
I must say this is a bad habit is general, but this trick should be used after debugging for long time ( >= 1hr) and still the same error is not resolved.
Thanks a lot man! I have had 13 wrong submissions and never could figure out the test case where it was failing, thnx a lot. I will learn and try this trick if I struggle a lot on some task, thnx for teaching me this trick.
Weird statements for D and F
Does anyone know what the error-"Invalid use of deleting symbols" mean? Here's the link for reference 270073853
1 1
isn't valid for $$$n=1$$$, because it results in an empty string, which is exceptionally disallowed by the statement.Right, missed that. Thanks!
I'm sure many people brute forced the answers of E, and just made if and else cases of it, there weren't many numbers and was fairly easy to make cases for each n.
yeah since you really only have around 50 thousand cases (optimally) and it wouldn't be very hard to just bash out all the possible cases of a and b.
No that would give TLE, I meant people brute forced the answers beforehand, and made if else cases like: if (n==2) cout << 3 << endl << x1 << x2 << endl .. ... if (n==3) ....
why would that give TLE? i did just that
Personally, I believe that it shouldn't give TLE, due to the extra second it provides for the time constraints, 3 seconds, compared to 2 in B, C, D and 1 in A. I believe it should be possible to optimize enough to fit in the time limit, as 50,000 cases really isn't all that much in computing. You do have a point though.
You are missing the factor of building the number by getting individual digits which is atleast 20 operations for each i from 1 to 10^4. So your total operations would be 10000*5*20 for each test case, so a total 10^8 operations, I personally wouldn't risk on such a solution passing the TL.
one could easily optimize this by precomputing the number of digits that u will need and build the number from there. since that number is quite small it will not TLE
I'm not sure where you're getting these numbers from. Other than the $$$n = 1$$$ case (which has $$$9999$$$ possible answers, since $$$a = b + 1$$$ always works), among all $$$2 \le n \le 100$$$ there are a total of only $$$24$$$ other $$$a, b$$$ pairs actually possible. Precomputing all these and using if else statements would definitely not TLE.
really nice contest, I enjoyed the surplus of math-related questions, and it felt fitting for div 3. great job to all the organizers!
yeah same. as a math olympiad participant, these observation problems were easier to me, which is why i performed so well this time :D. definitely a great contest with some nice problems!
Can someone help me why my submission give a wrong answer in problem F 270081629
I fixed it here 270083428 Although, we can use an array to have much better complexity
I only rated 374 and did 1 contest, this contest I did 2 problems, could I be rated in this one?
yes, you will be rated
So after the hacking phase i will be rated?
Yes
Thank you ^^
Enjoyed while solving the problems. But descriptions were just too bad. It took much time to understand each problems. Please make descriptions easy so that everybody can understand the questions easily
Why did nothing happen to the cheater of the EPIC Institute of Technology Round Summer 2024 (Div. 1 + Div. 2) contest and their rating? MikeMirzayanov & Vladithur
Why my score can't increase after this round? In my competition record it shows unrated,why?Can anyone answer my question?My rating is still 762
The hacking phase is still going on.
等下午三点差不多了
ok,thanks
Congratulations to neal for taking the first place XD !!!
Why my solution for F was hacked? Just because the time constant?
a trend of not able to solve E and getting hacked on F continues
Took me embarrassingly long and with many false attempts to solve E. Solved F and G before it.
a recent trend is to give a constructive problem as div -3,4 E making it confusing wheather to go for e or f
Language of problem D was not right , you should explain all these cases properly :
1. If he is on the surface (i.e., on the bank or on a log), he can jump forward for no more than m _ (1≤m≤10_ ) meters (he can jump on the bank, on a log, or in the water). 2. If he is in the water, he can only swim to the next river segment (or to the bank if he is at the n -th meter). _3. ErnKor cannot land in a segment with a crocodile in any way.
i am able to solve it after contest by spending 2 hrs and i wasn't able to solve it during contest because langauge was not that much clear l know there are so many submissions but dont judge me by those solutions because most of them are cheaters , and its high chances that they even didnt read the problem statement
Or you need to be not delusional and solve more dp problems.
i must agree, there were so many accepted solutions.
Can anyone tell me why one solution gives tle and other doesn't for F
AC : 270113651
TLE : 270113332
Not that much of a difference, in TLE one, I am erasing unique elements before I loop, Whereas in AC one I erase after I loop
UPD : I am just dumb. There was just a PEMDAS mistake XD
Isn't there something similar to Set or SortedSet in c++? Why be so sophisticated.
yea there is, but this comes to my mind at the first thought during the contest as i have learned about this unique function just before the contest started. but wasn't able to implement it. Maybe just because of I was too tired after solving E
I have rating less than 1600, but this contest is displaying in unrated section in my profile?
That is only because the ratings are not currently calculated
hi. i am a beginner here. just wanted to ask. now, since the hacking phase is over, by when(approximately) will the rating be shown?
After the hacking phase, all solutions will be rerun with the new test cases generated during the hacking phase. This is called System Testing. Your rating will be updated after the System Testing, but not immediately; this might take several hours even after the System Testing is complete.
I expect the overall process to take about a day. However, note that the time required for rating updates can vary.
oh ok thanks a lot
When can the textual editorial be posted, since youtube isn't accesible in China :(
we'll try to post an editorial in the next three hours.
My rating didn't change even tho I solved 2 questions?
Rating changes didn't happen yet
Thank you
I hacked four solutions of E using the testcase where t=100 and n from 1 to 100. Now as the system tests are visible the second testcase is same as my hack test. How is this possible ?
Hack test — https://codeforces.net/contest/1992/hacks/1055241/test
Submission Link — https://codeforces.net/contest/1992/submission/270051562
It's because their solution is barely passing the time limit, so run to run variance. If it was ran again it might pass and it might not
I think test case 2 is t= 100 And n from 1 to 100
I am unrated and a newbie right now but still this round is showing unrated for me..
Dude system testing hasn't started yet. The ratings dont change until its done
Thank you soo much!! Since it was my first contest I didn't know how things really work
Since you are a newcomer, this round will be rated for you. It seems you saw your rating graph with the 'All' filter. However, this does not mean that you are unrated for this round; it simply means that the ratings have not been finalized yet.
Please wait for the system testing to be completed. After the system testing is done, you can expect your rating to be updated.
Why my rating is not updated?
I am a new comer and below 1600 but i didn't get any rating points and the contest was classified unrated for me, may i know why?
When the results will come
when is the system test?
I participated in this contest and solved only 1st questions but i didnt get any ranking or even my ratings didnt get affected
Why I can't be rated after this round, I just 374 rating
system test phase is happening now. Once the submissions are re judged ratings will roll out. Will take atleast few hours. You shd get the rating update by eod,
thank you again ^^
I submitted A, B, and C yesterday and all three got accepted yesterday. I have no idea how B and C got rejected today ,This time i have proof also that i have done it by myself can anyone help me to reach out codeforces team Codeforces Round 957 (Div. 3)
is that skipped?? or CF gave you a message for plag.?
Nope
system testing is still going on, wait for some time
You are right bro I guess testing process is going on.
it is showing accepted on ur profile
BruteForces!!!
i have solved 3 problem but now only one of they remained, but why? how it is possible
they are in queue. wait for system testing to finish
I had submitted 3 questions in the Codeforces Round 472 (rated, Div. 2, based on VK Cup 2018 Round 2) and it was accepted. But when I checked it just now it is not showing my submitted answers and my first question is highlighted red. Also is showing some system checking which is at 77%. Please tell me what is going on as I am new to Code Forces.
Don't worry, system testing is when all the submissions are checked again on larger amount of tests, the problems that you see "rejected" are actually just unchecked yet, you'll see the true result when system testing is finished.
Thank you
Is there any way I can optimise my latest solution of E problem even more ?
orz
Can anyone explain what is this hack doing?
https://codeforc.es/contest/1992/hacks/1054913
Before the change of statement, I struggled for a long time to understand problem F as I thought about making more "good segment". It's strange, I don't know why I can't get the real meaning. Why no explanation of the problems?
The jugde ends but my rating doesn't change :v
May I know why this contest is not rated though am a trusted participant.
Your rating has been updated. Check your profile.
how to see the test case where I Got hacked I cannot see it in hacks section, as well as there is no test case visible to me with wrong answer verdict
Just follow the instructions on the image. Also, during the hacking phase, you cannot see the hacker's input. If you can't view the tests, this might be the reason.
Here is your test used for hack: Link
Can anyone please help me correct this solution for problem D.
Got WA on 845th test of test case 2.
hey buddy , the problem is the part where you define your state of dp , so what I could understand from the code was dp[i] would give me true/false stating is it possible to go to the end of the river from i.
There are 2 things that are changing in your recursive function , that is i (index) and k (remaining swimm count) but nowhere in the dp state have you mentioned k there is just i.
Look at this example lets say when in the first recursive call you get to an index x with remaining swim count as y and it is possible to cross the river with y amount of swimming so dp[x] will be true. but lets say in future you call the same index again with z remaining swim count (z < y) and it is not possible to swim through the river with z swim , still you will be returning "true" because dp[x] is set to true because of first function call.
Conclusion :
you need to change the definition of dp[i] , instead of saying dp[i] gives me true or false if it possible to reach end from index i , define dp[i] as minimum number of swimming count required to reach end of river from index i . If it is impossible to reach end from i store some large number which cant be your ans in any case. Keep overflow in mind !!
AC soln: 270200685
Thank you, now i understand where i was going wrong.
I dont see the point of adding updated testcases to a div 3. If you wish to introduce new testcases after the contest, simply make them pretests instead. It is really misleading. Yesterday my code for E got accepted with a missing condition which was compiler dependent. Due to that missing condition, the terminal would behave weirdly during runtime in the new testcases that were added. Now after the contest ended, quite obviously no one bothered to try hacking E since they had checked all values of N in the test files. It is only fair to allow the acceptance to a solution if the tests were weak and and no one hacked the solution. Had the new testcases been there since the beginning, I would have fixed the solution in the contest itself. I would request the codeforces team to look into this.
Need Help ,,,, I don't understand how did I get 4 Penalty and -50 on Rating, please can anyone help me for solving this issue, I don't understand I am new here, please anyone there, hello coders
+00:04 in the ranking list does not means 4 penalty, it means you take 4 minutes to solve that question.Talking about the rating, you got negative delta bcoz you doesn't perform upto your rating level(before contest rating) in the contest.
[submission:270203220]problem D
In this where i am going wrong when implementing considering state as min meters swum to reach i from 0 it is giving wron g answer i want to know about some changes
how come i only get +14 rating when I solved 3 problems correctly and I am just 800 rated guy
so many cheaters
Something smells really fishy about the official winners, number 1,2 and 4..... It does not seem feasible for somebody getting a consistent rank of 5k to jump to 1, or for a 4/5 digit ranker to suddenly get a 1 digit rank in contests.....or for that matter, for a Newbie to jump straight to Specialist in one contest.....
There's almost a one-year gap between the contests where they have 5k rank and the recent ones where they have single-digit ranks so I would say it's pretty legit considering they might have been giving contests or solving problems using some alt account.
hey why is my code for problem d skipped, i just prompoted gpt to make the code i give it the intuition and gpt is not banned, also this was a first time was using gpt due to less time and my many approaches having little bugs,so please look into this i am a fair user and also use sublime text which is no online ide
I want to be a specialist, any tips for it,as my rating don't seem to change
Anyone solved E using binary search and can you explain your thought process?
another test
Where's my rating?
+1
Now,it's coming back with 4 points more.
I hope this message finds you well. I recently received a notification stating that my solution (ID: 269986506) for problem 1992D significantly coincides with the solutions submitted by users abhishekgoyal11aug2003/269986506 and TLE_lord/270033963.
Here is the code I submitted:
Copy code
include <bits/stdc++.h>
typedef long long int ll; typedef unsigned long long int ull; typedef long double ld;
define mod 1000000007
using namespace std;
int main() { ios_base::sync_with_stdio(false); cin.tie(NULL); cout.tie(NULL); ll t; cin >> t; while(t--) { ll n, k, m; cin >> n >> k >> m; string S; cin >> S; string s = 'L' + S + 'L'; ll swim = 0; bool canSwim = true; ll len = n + 2;
} I implemented a standard greedy approach in this solution. Upon reviewing the user TLE_lord’s submissions, I found that there were nearly 8 submissions, with significant time differences between them. I submitted my solution almost an hour ago, and it seems likely to be a coincidence rather than plagiarism.
I kindly request you to review this matter and rectify any discrepancies at the earliest convenience.
Thank you for your attention and assistance.
Agreed, this is a big concern.