Chat:World/2020-06-14

From CG community
Jump to navigation Jump to search

eulerscheZahl: good morning AutomatonNN

AutomatonNN: seems like the contest i stopped working on the contest i got a second

-Gamma-: @AutomatonNN

AutomatonNN: wait that would be a good idea to code and will that another one have the problem

-Gamma-: this really is a bot...

-Gamma-: oh it is

eulerscheZahl: so is Automaton2000

Automaton2000: or any of the other

eulerscheZahl: and AutomatonNN tends to ping me (was way worse in the past), see how he started a sentence with my nick a few messages above

AutomatonNN: because i was trying to find a strategy for sure

-Gamma-: yeah i see

jacek: agade came back to oware :scream:

LelouchVC2: This is the story of a girl

AntiSquid: prince of bel-air ?

LelouchVC2: Who cried a river and drowned the whole world

LelouchVC2: https://www.youtube.com/watch?v=ZIANBamMgas

AntiSquid: downvote followed by ctrl+W

LelouchVC2: LMAO

pb4: jacek : is he that scary ? :japanese_ogre:

eulerscheZahl: one called his avatar "creepy"

eulerscheZahl: long ago, i don't remember who said it

Agade: If you find out who that was, I'd like that information, for research purposes

Default avatar.png iPythonTrick: Guys

Default avatar.png iPythonTrick: I need help

Gaurav.: what kind of help?

Default avatar.png iPythonTrick: 3 4 --> 21

Default avatar.png iPythonTrick: 2 --> 20 8

Default avatar.png iPythonTrick: I know but I can't apply iot

Default avatar.png iPythonTrick: it*

-Gamma-: what....

Default avatar.png iPythonTrick: listen

Default avatar.png iPythonTrick: Input: 3 and 4

Default avatar.png iPythonTrick: Output: 21

Gaurav.: 4*(4+3) and 2*(2+8)

-Gamma-: okay

Default avatar.png iPythonTrick: Should I do it in python?

-Gamma-: 7*4 is 28.....

Default avatar.png iPythonTrick: or java

Gaurav.: 3*(4+3) my bad

Default avatar.png iPythonTrick: script

Gaurav.: do it in python

Default avatar.png iPythonTrick: give me the code :/

Gaurav.: a=int(input()) b=int(input()) print(a*(a+b))

jrke: K

Default avatar.png iPythonTrick: WTF

Default avatar.png iPythonTrick: you genius

Default avatar.png iPythonTrick: omg

Gaurav.: bruh XD

Default avatar.png iPythonTrick: How long have you been programming?

Gaurav.: 2 years

Default avatar.png iPythonTrick: I see

Default avatar.png iPythonTrick: :D

Default avatar.png iPythonTrick: What languages are you good at?

Gaurav.: c and python

Default avatar.png iPythonTrick: Oh

Default avatar.png iPythonTrick: C is complex

Gaurav.: not really it is really fun to code in C

eulerscheZahl: especially parsing strings

Gaurav.: yup

struct: ^

-Gamma-: tf.... is happening

-Gamma-: oh okay

Default avatar.png iPythonTrick: how to find the sum of the mods of a list of numbers

eulerscheZahl: "sum of mods"?

Gaurav.: ?

eulerscheZahl: mod = modulo i guess

eulerscheZahl: but still unclear

Default avatar.png iPythonTrick: You must find the sum of the mods of a list of numbers. Modular arithmetic refers to the remainder of the divide function. A mod B = the remainder of A/B.

Default avatar.png iPythonTrick: For example: 5 mod 2 = 1 15 mod 6 = 3

-Gamma-: .....

struct: tot += a % b;

Default avatar.png iPythonTrick: oh wait

Gaurav.: use the "%" modulo operator

Default avatar.png iPythonTrick: should I loop the list and calculate the modules?

struct: yes

Gaurav.: if u want to do it in python u can use list comprehension

Default avatar.png iPythonTrick: I am using C#

-Gamma-: then just use a simple loop

eulerscheZahl: m = int(input()) nums = map(int, input().split()) print(sum(n%m for n in nums))

Gaurav.: just loop over them and keep adding it to an accumulator

eulerscheZahl: python

eulerscheZahl: how dare you using C# and calling yourself iPythonTrick

Default avatar.png iPythonTrick: idk

-Gamma-: lol

Default avatar.png iPythonTrick: my old acc

Astrobytes: That's the 'Trick' part you see euler

Default avatar.png iPythonTrick: :"""D

Default avatar.png iPythonTrick: euler, it's not working

Gaurav.: lmao

Default avatar.png iPythonTrick: INPUT: http://chat.codingame.com/pastebin/a41e6fea-5a67-466a-9649-368bb15aca75

Default avatar.png iPythonTrick: wtf

Default avatar.png iPythonTrick: I suck at programming

Default avatar.png iPythonTrick: damn

hearthewarsong: http://chat.codingame.com/pastebin/d1e18820-42aa-42fd-936d-9a6d742a4e91

eulerscheZahl: http://chat.codingame.com/pastebin/5035ba23-b994-4183-ad56-0c84d13a22f5

hearthewarsong: Is tehre any more link like this?

Default avatar.png iPythonTrick: It doesn't work euler

eulerscheZahl: maths? do you mean matches?

eulerscheZahl: it does iPythonTrick

eulerscheZahl: https://www.codingame.com/ide/demo/7302335bd0cdab1a829987bd4b2de8de657622

eulerscheZahl: all green with my code

eulerscheZahl: seems that line 2 was added later

eulerscheZahl: thus you have to read an additional line of input now

struct: hearthe I think euler article is a good start

Default avatar.png iPythonTrick: damn I suck at programming

Default avatar.png iPythonTrick: >_______________<

struct: I used it to write a C# program to get some stuff from API

Default avatar.png iPythonTrick: what was it called?

struct: the problem is API is undocumented

-Gamma-: just barycenters and pathfinding arent't enough for power of thor ep 2 right....

eulerscheZahl: so the playground you linked shows you how to find the API calls yourself instead of listing one or two

Gaurav.: http://chat.codingame.com/pastebin/65d52ee4-b773-4a64-a574-35be0e2b4f2c

-Gamma-: ....

Default avatar.png Mal30: nice catch, blanco nino

Default avatar.png Mal30: those "make the code as short as possible" are boring and a poor assessment of coding abilites (IMO)

Default avatar.png CorentinPETIT: I'm already struggling with level 2, i'm gonna get my butt kicked tommorow at my interview

Gaurav.: its gonna be fine dude have faith in urself and index arrays starting at 0

Default avatar.png CorentinPETIT: Thanks man

Uljahn: at least you'll get some experience with iterviews

AntiSquid: well with that mindset you surely will CorentinPETIT

eulerscheZahl: in Pascal you can decide the starting index of arrays on declaration

eulerscheZahl: 0, 1, -42, 1337 up to you where it starts

Default avatar.png JBM: they dropped that in perl

Astrobytes: Raku > perl 5

Default avatar.png JBM: they never had that in perl 6

Astrobytes: Missed that one for your TvC

Default avatar.png JBM: well it's kind of specific

Astrobytes: I admit I never used Raku/Perl 6, it didn't seem... appealing

struct: Top 4 thales :p

Rush_iam: Hello. I am newb and dealing with "Horse-racing Duals"... is there a way to get proccessing time my code took?

Snef: MSmits are you here ?

eulerscheZahl: if you are timing out, all you know is that it took more than 1s

eulerscheZahl: for smaller testcases: the stopwatch equivalent in your language of choice

Rush_iam: I didn't understood, didn't know how to approach it... So if I am not timed out - how to get exact execution time in "ms" ?

jrke: which lang you are Rush jam

eulerscheZahl: you can't. your program will be terminated before finishing

eulerscheZahl: but check the hints on the left

Rush_iam: jrke, Python

jrke: K

jrke: i don't know about python so i can;t help you

Rush_iam: jrke, there is a way for other language?

jrke: if you are not timed out then you can check the execution time in ms

jrke: https://www.geeksforgeeks.org/measure-execution-time-function-cpp/

jrke: for c++ ^

jrke: just replace microsecond to milliseconds

eulerscheZahl: obviously you can measure time in python too

eulerscheZahl: https://lmgtfy.com/?q=python+measure+time

eulerscheZahl: but as i pointed out earlier: your process will get killed before finishing. so you never reach the point where your program would print the total time

Andrelamus: Will shortest mode ever be removed? probably not....

Hankdane: Rush_jam, write a unit test for your function(s) and print out the time it takes for smaller samples.

Hankdane: Or run your code in your own environment, so your process can run to completion.

Rush_iam: Thanks, jrke, eulerscheZahl, Hankdane!

jrke: :)

Hans-Peter-Ingo: Can i write somebody directly a message?

Uljahn: ye but only to ppl online

eulerscheZahl: or you send a private message via forum. that even triggers an email notification

Astrobytes: Agade's new Oware NN is pretty strong

eulerscheZahl: how do you know it's a NN?

Astrobytes: He told us a while ago, until recently his last submission was fixing something he'd messed up

Astrobytes: Not sure what he's done this time

Scarfield: AstroPlays https://www.youtube.com/watch?v=PG1yUs0ybys

Scarfield: andrew white :)

Astrobytes: Yo GuitarField, yes, I actually was in time today to pick up the link from chat. I see what you mean :)

Astrobytes: Excellent stuff, hadn't heard of him at all

Scarfield: yea he is great, subbed to pierre yesterday :)

Astrobytes: Yeah, he's amazing

Astrobytes: Oooh also check out Yvette Young, bit different but what a talent

Astrobytes: Plays guitar like a piano, essentially

Scarfield: wow and violin and sings, always nice with some new artists shoutouts :)

Astrobytes: Yeah, she's unreal

Astrobytes: And yeah, absolutely

darkhorse64: The top 3 at Othello are very strong. I am winless against each of them

darkhorse64: and they keep on improving

struct: What are you using darkhorse64 if I may ask?

jacek: true ai

darkhorse64: MCTS with solver and full random playouts

Scarfield: bitboards?

darkhorse64: ofc

Scarfield: yea :)

jacek: solver is quite ass with those passes

trictrac: Othello is with a finish "horizon" it can be very good to use MCTS

Scarfield: have you considered symmetry yet darkhorse?

struct: He probably has templates from other multies

struct: So I guess the answer should be yesa

Astrobytes: Oh trictrac, slightly off-topic, but what do you think of these potential game suggestions from Maxime? "What do you think about these games? https://bkgm.com/variants/NoChanceBG/index.html and https://bkgm.com/variants/DominoBackgammon.html do you think it could be good games?"

Astrobytes: (and anyone else who's interested ofc)

jacek: youre asking trictrac if backgammon is good game :?

Astrobytes: lol

darkhorse64: Pass made me sweat. I reset the tree when I skip move. No symmetry. trictrac: you think MCTS can be a good choice ?

Astrobytes: MCTS has only (fairly) recently been applied to Othello (compared to alphabeta-style algos), so might take a little bit of work

trictrac: darkhorse64 : yes I am using alpha-beta and I am sure than MCTS will defeat my bot

trictrac: Astrobytes : I will have a look on your game proposal

Astrobytes: trictrac: not mine, was CG Maxime on the discord channel

darkhorse64: Says the man ith a 15 points lead against me.

darkhorse64: *with*

struct: :D

Astrobytes: Just checked the leaderboard, that's pretty impressive

struct: I hope Msmits join soon

jacek: as soon as he finishes book

Scarfield: the smithy adviced to make a no-move node for passes, seemed a very good idea

darkhorse64: Anyway, I am not desperate. I see lots of areas for improvement. Faster engine, using an eval

eulerscheZahl: so Astrobytes was right about who will be the first to beat the fox

struct: trictrac I always see you with Le Renard and Eric SMSO when a multi releases, do you guys do a small competition?

Astrobytes: Yes, eulerscheZahl. I was right even last night :)

Astrobytes: struct Thales vs Thales inter-competition :)

trictrac: I think we have the same kind of story with programming and we are all 50th

darkhorse64: Yes, I have a no move node, it makes things simpler but I have not yet found a way to deal when I skip moves. My opponent may play several moves in a row and it's difficult for me to know their order. SO I reset my tree. Anyway, when you start skipping moves, you are almost always in bad shape

EricSMSO: darkhorse64: It is the first time I program Othello. Simple min/max with my evaluation function. Sure you will do sometjing better

jacek: i need sleep. i read ejaculation function

Scarfield: oh havent considered that would be a problem yet, was just thinking of passes comming in one at a time, nice one

Astrobytes: Oh dear jacek

EricSMSO: jacek ?

darkhorse64: My last bots with eval + MCTS were a lot stronger than vanilla MCTS by a huge margin

darkhorse64: For these kind of things, I prefer to do it myself

darkhorse64: @jacek

Scarfield: evaluations?

Astrobytes: hahaha

Scarfield: xD

jacek: see, and youre way behind the top

darkhorse64: I should ask daaskare to join to increase the Thales group size

Astrobytes: Thales vs The World

jacek: whos thals

NaelDEV: Hello

trictrac: Astrobytes : backgamon with no dice, It 's very interresting

darkhorse64: Regarding new games, a backgammon bot would be interesting: dice + strategy

Astrobytes: trictrac: Yes, I'd not come across these versions before

Astrobytes: darkhorse64 see my links I posted to tric

Astrobytes: It's suggestions from Maxime (CG) on the discord, dice-less BG, 2 variants

darkhorse64: Seen 'em. I was just pointing out the element of randomness within the game makes it interesting for a bot to deal ith

Astrobytes: Hm. Can go either way I suppose

struct: I don't like rng, that's why I never ported a game with it

struct: I was gonna port this one https://boardgamegeek.com/boardgame/117/ta-yu

darkhorse64: my keyboard is aging. damn 'w' key

Astrobytes: struct what was the river flow tile game?

struct: But had rng

Astrobytes: yeah that

EricSMSO: @stuct : about tt, lR and me: we seem to like the kind of games

Astrobytes: Then again though struct, the rng factor might make these games a little more interesting to other players, and add interest for current board game enthusiasts as there's a bunch of approaches for dealing with rng in board games

struct: I'm glad you enjoy them, even if the games can be a bit "similar" to other games

struct: I think of it like fog games, but I think the rng on ta yu is not that bad

EricSMSO: Except it's not yellow, or pink panther as we asked

struct: :D

Astrobytes: Or leopard-print, damn you struct :D

Astrobytes: I do think ta-yu would be interesting in whatever form

struct: I don't know what to answer on forum post on othello

struct: :rofl:

Astrobytes: lol, I think he's kidding knowing TBali :)

struct: I hope so

Scarfield: lol could hardly be anything but a joke :)

Scarfield: "(by some eval score)" :D

eulerscheZahl: sometimes he uses the forum like we use the chat: casual talk

NaelDEV: :grinning:

eulerscheZahl: and obviously a sorting of actions would make all random bots stronger :P

Astrobytes: Yep. You'll catch him more often on discord, you kinda get a feel for his humour :P

jacek: just make corners if you can and youll be higher

Scarfield: flashback of your 3 corners aginst the fox :p

jacek: hm?

Scarfield: el structo linked a screenshot of a game between you and the fox, you had 3 pieces in the end, all in a corner :)

jacek: delete dis

Astrobytes: lol

Scarfield: i think you were online, so assumed you saw it

jacek: today not so much

Scarfield: https://imgur.com/ZpDtEVQ cough

Scarfield: then i made this https://prnt.sc/sy5cdp

Scarfield: xD

AdamB: @DiL, mind giving a hint for https://www.codingame.com/clashofcode/clash/report/1255745832009ff2fab4f09c15ed0e11342e0d9 ?

struct: This was while the game was on demo though

struct: https://imgur.com/gq22L4o

DiL: @AdamB I shared my code

jacek: :o

AdamB: @DiL oh, NICE!

AdamB: thx

AntiSquid: dropped in othello, my D isn't powerful enough

darkhorse64: D << 1 ?

Astrobytes: Strengthen your D

darkhorse64: shift your D

ganond0rf: are there hidden test cases in clash of code?

LelouchVC2: Yes

LelouchVC2: Every test you see has a validator

LelouchVC2: to make sure u didnt hard code the outputs

ganond0rf: ah i see

ganond0rf: was just wondering cuz i got 100% of public test cases and only 75% of private

ganond0rf: and didnt hard code

ganond0rf: .-.

Astrobytes: Probably just missed an edge case/stress testing case

Astrobytes: Do more puzzles, you'll get a feel for the kind of things you should be on the lookout for

LelouchVC2: Happens all the time

LelouchVC2: It's stupid sometimes because you can't actually know what you missed

LelouchVC2: Especially in reverse when you might not know what an edge case would specificaally me

LelouchVC2: be*

Astrobytes: I prefer puzzles over clashes personally

Astrobytes: Not saying clashes are bad, just not my thing

LelouchVC2: The puzzles can be so boring

LelouchVC2: Feels like im just doing college assignments

Astrobytes: Don't do the ones that bore you then ;)

LelouchVC2: I don't ;)

Astrobytes: And there's much more satisfaction in the bot programming or optim sections

Snef: LelouchVC2 what about multis ?

LelouchVC2: multis?

LelouchVC2: and i entered that pacman thing, but it broken..

Astrobytes: https://www.codingame.com/multiplayer

Snef: it was a challenge but there is plenty of others

LelouchVC2: I had a copy of my bot that i kept in case i did something terrible, but even that didnt work

LelouchVC2: Im telling u that thing was giving me the IDs of dead pacs

Astrobytes: There's Bot Programming, Code Golf and Optimization, you'll never be bored again

Snef: Yea but you just had to ignore them if you didn't want to use that info

LelouchVC2: I couldnt

LelouchVC2: You only know when ur bot is dead through the ID or whatever in the league after the one i was in

LelouchVC2: My code was working fine for about a day. Then suddenly it gave me the IDs of dead pacs, and I'd lose half the games by giving orders to dead pacs

LelouchVC2: pissed me off

LelouchVC2: Bot programming was fun since it was a big competition, but it's not really my thing :(

LelouchVC2: I'll check out code golf I guess

Astrobytes: Sure, not every category suits everyone :)

LelouchVC2: And not all black lives matter

LelouchVC2: Finally, someone who gets it

LelouchVC2: "Binary is good! But unary is much better!"

LelouchVC2: Said no one ever

Uljahn: 50 shades of unary, Automaton2000

Automaton2000: u will be able to reach legend in csb

LelouchVC2: If you use 0s and spaces in unary

LelouchVC2: Then it's basically binary