Chat:World/2022-04-05

ASM_MOV_INT: website took forever to load

ASM_MOV_INT: take taco

ASM_MOV_INT: run.oprah() - everyone gets a taco!

LelouchVC2: id rather get ur number

ASM_MOV_INT: 0118 999 881 999 119 7253

ASM_MOV_INT: only use it for emergencies though...k?

LelouchVC2: oh.. i already texted

ASM_MOV_INT: :joy:

ASM_MOV_INT: i hope you paused before typing that last number

LelouchVC2: e google

LelouchVC2: ez*

SilverEagle: take taco

CodingFasterThanWifi: i dont even know how to code bruh

ASM_MOV_INT: bruh

ASM_MOV_INT: you'll be smarter than a russian hacker in no time!

CodingFasterThanWifi: its so damm hard i hate this

derjack: :upside_down:

21stCenturyPeon: Nicer ambulances! Faster response times! Better-looking drivers!

TheAwardWinningGiantOfWisdom_d0a9: hi folks

TheAwardWinningGiantOfWisdom_d0a9: are there head hunting opportunities for participating in a contest?

iK4oS: Is clash of code broken?

isimkullanmiyorum: is clash of code closed ?

Pythonista_TN: and most important question : when it will be fixed ?!

[CG]Thibaud: :eyes:

ETET: Hello

ETET: Does Coc work for yall

ETET: ?

ETET: https://www.codingame.com/clashofcode/clash/2326334ae4a4cca2f9b76efd9e9f941710c3fa7

ETET: join if you want to

ETET: take taco

ETET: TAKE TACO

ETET: why does it not work???

Passifi: antiwonto is sustained by sunflower oil and or wontimos love and they are both all gone

ETET: ?

Passifi: the bot is down :)

ETET: lol

Uljahn: guess the bot is self-hosted on Wonto's PC, so it's not available 24/7

Passifi: no why somebody would leave their pc long enough to warrant shutting it down is beyond me of course

Passifi: but its a free country

ETET: lol

ETET: how long do u spend on this web?

Passifi: actively maybe an hour or two or so

Pacut: like 2 hours

SaN099: can anyone teach me javascript?

TheStrategist97: Just realized I have stayed up until now past 6 in the morning making something. I really should go to bed.

SaN099: i am new here

SaN099: good night

PainEXE: gn

virus_coder99: hi

Passifi: hola

Pythonista_TN: hey

RECH..: lol bruseve

Myles22: howd'y

burnpanck: take taco

qemcx: *

Batai22: hablo beans

Hugesn0wduck: hello!

Thorcode: omg

Thorcode: this is the first time why the game infor can be time out?

Hugesn0wduck: wdym?

Thorcode: a nothing

Hugesn0wduck: oh ok

Thorcode: just trying to solve escaping the cat XD

Hugesn0wduck: oh xd

Hugesn0wduck: i dont really participate on this website anymore

Hugesn0wduck: just use it for help from time to time

Thorcode: yep me too

Thorcode: but sometime i want to try some puzzle for fun

Hugesn0wduck: thats great. the more practice the better

AndreasD: My questmap is bugged, does anyone know how to get on from the first mission?

TanninCole: I am having the same issue with my quest map.

Thorcode: like?

TanninCole: I have solved the Onboarding puzzle multiple times, but the quest map says I need to solve it to continue.

Thorcode: try to restart google ?

Hugesn0wduck: tannincole in what way? is it saying you have correctly solved it or that it doesnt say that you were correct and that you cant continue?

TanninCole: I have several indicators that I have correctly solved it (like "My progress: 100%" and I have the ACHIEVEMENT "Onboarding Master"

Thorcode: I think it might be memory exceed on google

Hugesn0wduck: maybe

Uljahn: Codingame's database is having issues today, so quest map updates are paused atm, they'll be resumed later

Astrobytes: Good afternoon.

TanninCole: Thank you

Hugesn0wduck: oh i see

Uljahn: https://www.codingame.com/forum/t/onboarding-puzzle-discussion/119/424

Hugesn0wduck: astrobytes good afternoon to you aswell!

eulerscheZahl: DB issues? did Thibaud inform the devs already?

eulerscheZahl: also: hi

Astrobytes: it's known yeah, Tibo posted in the forum link Uljahn gave

Thorcode: hi Astro

eulerscheZahl: sorry, didn't click

Astrobytes: hello Thorcodo

eulerscheZahl: didn't expect onboarding to contain info about the quest map

Astrobytes: Knowledge. Knowledge everywhere :P

eulerscheZahl: and a contest is close

eulerscheZahl: hype

Astrobytes: The hype is not intense enough yet I feel, though I am looking forward to it

Astrobytes: Just hope most of the usual suspects will participate

eulerscheZahl: 7.5k registered. not bad, not impressive

Astrobytes: there will be more I'm sure

eulerscheZahl: last bot contest had 16.8k registered

eulerscheZahl: and 6.8k playing

Astrobytes: It's only the 5th currently ;)

eulerscheZahl: true, still some time to register. i predict 4k participants who submit

Astrobytes: I'll take 5k, feeling optimistic

eulerscheZahl: both is high compared to similar events (topcoder, RAIC, ...) i'm glad we get another chance

Thorcode: I hope this contest gonna be hard

Astrobytes: Very true, and yes, I'm grateful for it

Astrobytes: Also it's a bonus given the highly unfortunate lack of RAIC

eulerscheZahl: TCO finals should be around November. Yes, I watched parts of their weekly stream recording just to obtain this information

eulerscheZahl: and they hope to do it in person

Astrobytes: lol, oh right. Where is it held?

eulerscheZahl: USA most probably. they switch the city/state

Astrobytes: As long as it's not FL it should be fine then :D

eulerscheZahl: i don't get the joke about florida

eulerscheZahl: is that where Trump lives?

Astrobytes: it's a hotbed of covid

Astrobytes: and covid denial/deniers

eulerscheZahl: last contest was in Texas I think

eulerscheZahl: *looks it up* Houston

Astrobytes: Nice. I have a friend who lives there, incidentally.

eulerscheZahl: you can still qualify via TCO stage 4 ;)

Astrobytes: lol, I doubt it

eulerscheZahl: the hotel they took in 2019 now currently costs about 210€ per night :o

Astrobytes: They pay all expenses upfront right?

eulerscheZahl: no idea

eulerscheZahl: just thinking about how much such an event costs

Astrobytes: A lot. But they have a lot of financial backing

eulerscheZahl: 210€/night * a few nights + flight and multiply by 70 finalists

Astrobytes: + function suite + refreshments * (finalists + staff)

eulerscheZahl: + money prizes

Astrobytes: indeed

Hugesn0wduck: is anybody good with history? i need someone to test my game and see if cloud variables work

Hugesn0wduck: nvm got it to work

esd: i do it for 100$

therealbeef: how do you make an opening book? do you let your algorithm run for a long time, or do you let your bot play against itself and check which openings lead to wins?

antiwonto: [auto] Hey therealbeef, here is a :taco: for loggin in today while it is quiet. You now have 36 tacos

therealbeef: drop taco

therealbeef: drop :taco:

therealbeef: :table flip:

PluckyPenguin: *eat tacco*

Uljahn: throw taco

Uljahn: therealbeef: yes, you play offline with larger timelimits and record most promising moves

Mindarius: take taco

antiwonto: [auto] aww, there are no tacos to take

Mindarius: shakedown antiwonto

antiwonto: [auto] you need at least 3 :taco:s to do a shakedown

Mindarius: oof

facutito: cock

MaliciouslyCrypticUsername: Astrobytes im back :D

Drisnor: hello

jacek: astroback?

Astrobytes: now and then jacek

Astrobytes: and hi MCU o/

jacek: say hello, Automaton2000

Automaton2000: if you're going to get a better score

YoanD: TACO TUESDAYYYYY

Uljahn: Automaton2000: wah

Automaton2000: that would make sense to me

Ose: I

Ose: Definitely going to leave CodinGame bye everyone Bye

KiwiTae: cool by Ose

Wontonimo: weird

Uljahn: :tada:

jacek: :taco: derjack

antiwonto: [auto] jacek has awarded derjack 10 tacos. derjack now has 57 taco. jacek now has 33 taco

eulerscheZahl: disappointment about the next contest not being an escape?

eulerscheZahl: bye Ose

eulerscheZahl: :taco: derjack

antiwonto: [auto] eulerscheZahl has awarded derjack 10 tacos. derjack now has 67 taco. eulerscheZahl now has 35 taco

eulerscheZahl: eat taco

antiwonto: [auto] eulerscheZahl has eaten 10 :taco: and your Swag is now 2. You have 25 left

Wontonimo: shakedown derjack

antiwonto: [auto] Using your superior Intelligence of 2 you still fail, dropping 3 tacos and crushing 1

eulerscheZahl: tako taco

Wontonimo: shakedown derjack

antiwonto: [auto] Using your superior Strength of 3 you shake 10 :taco:s off of derjack. 9 fall on the floor, 1 got crushed, and you lost 1

eulerscheZahl: take taco

antiwonto: [auto] eulerscheZahl has taken 6 :taco:s off the floor and now has 31. There are 5 left

Wontonimo: take taco

antiwonto: [auto] Wontonimo has taken 3 :taco:s off the floor and now has 29. There are 2 left

eulerscheZahl: take taco

antiwonto: [auto] eulerscheZahl has taken 2 :taco:s off the floor and now has 33. There are 0 left

eulerscheZahl: shakedown derjack

antiwonto: [auto] Although your Style of 0 is inferior you luck out and shake 10 :taco:s off of derjack. 9 fall on the floor, 1 got crushed, and you lost 1

Wontonimo: take taco

antiwonto: [auto] Wontonimo has taken 5 :taco:s off the floor and now has 34. There are 4 left

eulerscheZahl: take taco

antiwonto: [auto] eulerscheZahl has taken 2 :taco:s off the floor and now has 35. There are 2 left

Wontonimo: take taco

antiwonto: [auto] Wontonimo has taken 2 :taco:s off the floor and now has 36. There are 0 left

jacek: eat taco

antiwonto: [auto] jacek has eaten 10 :taco: and your Swag is now 3. You have 23 left

eulerscheZahl: i still have no style? :(

Wontonimo: have you been keeping track ?

eulerscheZahl: no

jacek: no? https://i.imgur.com/ernDlo3.jpg

YoanD: take taco

antiwonto: [auto] aww, there are no tacos to take

YoanD: :(

YoanD: eat tacco

YoanD: eat taco

VizGhar: sudo make me taco

Wontonimo: taco powers

antiwonto: [auto] That taco command has move to the #taco channel

Wontonimo: you can now get your attributes using "get attributes" or "taco powers" in the #taco channel

Wontonimo: also, shakedown has moved there as well

Wontonimo: hope everyone enjoys

Wontonimo: also, "drop" now works in addition to "throw"

Wontonimo: drop tacos

antiwonto: [auto] Wontonimo has thrown 10 :taco:s on the floor for the taking but 1 was eaten by a software bugs!

Wontonimo: take :taco:

antiwonto: [auto] Wontonimo has taken 5 :taco:s off the floor and now has 31. There are 4 left

Wontonimo: later :wave:

Mindarius: take taco

antiwonto: [auto] Mindarius has taken 2 :taco:s off the floor and now has 3. There are 2 left

TimberStalker: take taco

facutito: take taco

antiwonto: [auto] aww, there are no tacos to take

TimberStalker: why did it just completely ignore me

jacek: :upside_down:

antiwonto: [auto] ':upside_down:' was defined as ' ɥǝllo '

jacek: i dont know. i think ill go outside and play some balls game

antiwonto: [auto] jacek be respectful and watch your language https://www.codingame.com/playgrounds/40701/help-center/code-of-conduct

Astrobytes: lol

KiwiTae: >< ball ball

eulerscheZahl: so, balls is bad

antiwonto: [auto] eulerscheZahl be respectful and watch your language https://www.codingame.com/playgrounds/40701/help-center/code-of-conduct

eulerscheZahl: what about boobs?

eulerscheZahl: that's sexist

Astrobytes: wonder what other fun words are on the list

Astrobytes: or not, as the case may be

macobo: one way to find out

macobo: a

macobo: aa

macobo: aaa

Astrobytes: yeah, no thanks

Astrobytes: a touch longwinded :D

macobo: hey, brute force is my best algorithm

macobo: don't diss

Astrobytes: haha

MrBear1337: what's the best way to _memorize_ algorithms? i have a job interview with amazon and it seems like i need to know 20 or so common algorithms by heart

Astrobytes: use them

Astrobytes: repeatedly. To solve actual problems

Astrobytes: Also, make sure you fully understand them

eulerscheZahl: https://github.com/panchishin/cgchat/blob/main/responseHandlers.js#L100

MrBear1337: thanks @astrobytes

Astrobytes: eulerscheZahl: I was too lazy to look :D

eulerscheZahl: we discussed it the other day

eulerscheZahl: found it in the chat logs

Astrobytes: I haven't had the willpower to check the chat logs, far too much to catch up on recently

eulerscheZahl: you didn't miss much, mostly taco spam

Astrobytes: Ah, not so bad then

jacek: amazon? why dont you try amazons game :v

eulerscheZahl: i did Amazonial already

Astrobytes: the other one

jacek: reddit ddosed?

eulerscheZahl: works for me

jacek: but is slow

Astrobytes: blame redditors

jacek: its not even friday

eulerscheZahl: got a reddit hug of death?

jacek: ~

Drisnor: hello

LX_The_Entity_Bounty_Sniper_303: HELLO

Zorg1: hello

antiwonto: [auto] Hey Zorg1, here is a :taco: for loggin in today while it is quiet. You now have 4 tacos

Westicles: wow 3 new puzzles, I'm having a balls

antiwonto: [auto] Westicles be respectful and watch your language https://www.codingame.com/playgrounds/40701/help-center/code-of-conduct

Westicles: *ball

DestyNova: :joy:

antiwonto: [auto] Hey DestyNova, here is a :taco: for loggin in today while it is quiet. You now have 1 tacos

Passifi: me want to

Synergyforge: I like my tacos like I like my coffee: without coffee or milk.

CoolandonRS: Is there a way to put the certification tests into dark mode?

LelouchVC2: i have "dark reader"

LelouchVC2: free in the chrome store, works really well on almost every website

nvhoangpascal: hello

LelouchVC2: goodbye

Tropical-Erix: GAY

Tropical-Erix: :grin:

ASM_MOV_INT: anybody else gay other than tropical?

Lisa-Has-Ideas: "not that there's anything wrong with that"

Tropical-Erix: Fuck up

antiwonto: [auto] Tropical-Erix be respectful and watch your language https://www.codingame.com/playgrounds/40701/help-center/code-of-conduct

Tropical-Erix: kill y

Tropical-Erix: kill yourself

Tropical-Erix: :rage:

ASM_MOV_INT: take taco

antiwonto: [auto] aww, there are no tacos to take

Rebollo: Guys anything related to Natural Language Processing in CG?

Wontonimo: yeah

Wontonimo: https://www.codingame.com/training/easy/markov-text-generation

Wontonimo: but nothing advanced

Rebollo: thanks

Rebollo: what would be the best approuch to create a Natural Language to SQL app

Wontonimo: ?

Wontonimo: what do you mean?

Wontonimo: do you want to translate natural language into working SQL ?

Rebollo: yes that´s the idea..

Wontonimo: ha

Rebollo: I mean best ML algorithm, or laguage, or library

Rebollo: I would like to use js

Wontonimo: hmm ... let me think ... cutting edge domain, do a task that 99% of the population doesn't know how to do, no training data ...

Wontonimo: got it, use tensorflow.js

Wontonimo: how though? You'll definitely want to write a paper about it because it'll be a breakthrough

Wontonimo: seriously, that would be crazy difficult

Wontonimo: perhaps start with something a lot easier

Wontonimo: like spelling and grammar correctionj

CoolandonRS: Is there a way to reorder test cases on a contribution?

Rebollo: :upside_down:

antiwonto: [auto] ':upside_down:' was defined as ' ɥǝllo '

Rebollo: lol lol lol lol

Wontonimo: sorry if i came off as unsupportive Rebollo ;(

Wontonimo: tensorflow.js is a good js based neural net framework. you can definitely do NLP with it

Rebollo: that´s ok always good a dose of reallity

Rebollo: but maybe I can learn something during the process even I can't achieve a reasonable result

Wontonimo: have you made a text generator NN yet using either a CNN or RNN ?

Rebollo: no never..

Rebollo: maybe i will start doing so

Wontonimo: oh, start with that! it's super fun to see your NN create an infinite amount of bad monte python

Wontonimo: or lord of the rings

Wontonimo: a really simple RNN that just is character based is easy enough to make and fast enough to train that you can get lots of experience trouble shooting ...

Wontonimo: and there is TONS of trouble shooting ... god is there ever

Rebollo: cool thanks for that

Rebollo: definitely will start with that

Wontonimo: http://karpathy.github.io/2015/05/21/rnn-effectiveness/

Rebollo: I was checking a JS library simple as that

Rebollo: http://chat.codingame.com/pastebin/e2b1cc66-5f35-43e9-829c-3d1f4add5e03

Wontonimo: https://www.tensorflow.org/text/tutorials/text_generation

Rebollo: probably won´´t work, right?

Rebollo: even if a provide tons os samples

Wontonimo: do you want that as a training example of question/answer so that given the question the NLP will respond with the answer?

Rebollo: I would provide a question like

Rebollo: How many employees does the company 'X' have?

Rebollo: And the result woyuld be soemthing like

Rebollo: Select count(*) from EMPLEYEES where COMPANY = 'x'

SilverEagle: take taco

antiwonto: [auto] aww, there are no tacos to take

Rebollo: Its a challenge but nothing´s impossible..

Rebollo: Imagine thinking modern smart phones 100 years ago..

ASM_MOV_INT: you are parsing natural language into SQL code

Rebollo: yes thats the idea

Rebollo: but would like to be able to perform more complex queries

ASM_MOV_INT: how much flexibility does the input have? like what if they reword the input: What if I wanted to know the number of employees at company 'X'.

Wontonimo: "for company 'x' tell me the number of employees"

ASM_MOV_INT: interesting idea

Rebollo: Maybe determining some keywords, classifying them, should do the work

ASM_MOV_INT: exactly what I was thinking...! :)

ASM_MOV_INT: if the given statement makes sense as only one command in SQL, then you have successful parse

Rebollo: number, how many, quatity, amount, value, price, cost

Rebollo: soething like that

ASM_MOV_INT: if more than one sql statement is possible from user input, ask a clarification question?

ASM_MOV_INT: or if not enough to make a complete statement

Wontonimo: i've worked on a few "4th level languages" since 1998. Languages that were billed as "replace developers with business analysts because programming with this is so easy". The problem is that people don't write very good english and don't rreally know exactly what things mean

Wontonimo: i'm not saying it isn't possible to do. I'm saying that non-programmers wouldn't be able to tell if the sql was want was intended

Rebollo: I can also provide some "HELP" functionality

Wontonimo: how do you plan on training it?

Rebollo: By giving questions, or maybe only words, and as result SQL snippets or aybe coplete queries

Wontonimo: to give some context, I have a NLP that just tells the sentiment of a sentence (if it is positive or negative) and it needs 100,000 training examples

Wontonimo: something more complicated like langauge-->sql will probably need a lot more examples

ASM_MOV_INT: I found a paper on the subject with a quick google search: natural language to sql

Wontonimo: +100

Rebollo: I got it but maybe parsing queries should be more straight forward once senntiment analysis is pretty subjective

Rebollo: sql seems to be more objective

Wontonimo: okay, back to your count(*) from employees example

Rebollo: yes

Wontonimo: shouldn't it filter out employees that have been terminated?

Rebollo: sure

Rebollo: maybe onlt if asked

Rebollo: or the result should provide by "categories"

Rebollo: or by all toher coluns that hasnt been used

Rebollo: or before executing

Rebollo: would use something like

Rebollo: "WOuld u like to be ore specific? The source table has tje following possible filters": http://chat.codingame.com/pastebin/4e8f7ad0-63c1-4a00-a582-1df7c5f1c5d4

PatrickMcGinnisII: gigo is perfectly exemplified in sql. I always have to do back-end maintenance as front-end users invariably find ways to multiply records or screw up field data. Natural language example of "Set reminder for 6pm" Still needs a delete at some point.

Rebollo: Like directing the user to best way to use the tool

Rebollo: I was thing soething like read only tool at first

Rebollo: for analyytics purposes

Rebollo: I would be happy if someone like to join me in the jorney :grimacing:

Rebollo: deafening silence

ASM_MOV_INT: i'm working on escaping the cat

ASM_MOV_INT: so far I just end up heading the mouse directly to the cat!

a-Rye: same, and then I had the mouse go in circles until it ran out of turns...

antiwonto: [auto] Hey a-Rye, here is a :taco: for loggin in today while it is quiet. You now have 2 tacos

PatrickMcGinnisII: escaping the cat is pretty easy

Wontonimo: outch! I found it hard

PatrickMcGinnisII: there's only two points to choose from, and they both involve the opposite angle of the cat. *shrug* yea, i guess at first ... but they make it easy cause the center of pool is at 0,0

PatrickMcGinnisII: tmi

Westicles: there's a nice 1-line solution (minus io) in python