Day 13 Discussion

[stone belfry]

umpkin

I have no idea how to do this one

Need to put my thinking cap on

The fact A and B cost a different amount messes me up

[sharp stratus]

I still need to do part 2 from yesterday. Found another edge case

Sounds like the bus stop problem from a couple of years ago

That was Chinese Remainder Theorem

Probably not though. The subreddit blew up xD

Math day today

[vivid fractal]

part 2

?

part 1 is easy

||just solve two simultaneous with Cramer||

||me when I use polar coords because I'm edgy||

[stone belfry]

||Isn't that the guy from seinfeld||

[vivid fractal]

wait that actually worked??

actually used something from linear algebra

hint: ||stop thinking of it as a discrete math problem for a moment||

||idk why the problem is worded as it is when there's just one combination that will lead you to the answer||

||or infinite?||

oh god I was so wrong

wait was i though

thonk

no idea why this doesn't work

nvm i just mistyped the answer

https://tenor.com/view/goku-laughing-hysterically-gif-19073610

LMAO that worked for part 2 as well

i might be a genius

[soft bramble]

Oh wow this looks hard

My initial idea is ||a regression|| but I'm just gonna get a local maxima and be fucked

[soft bramble]

Ok ||cramer's rule|| works for part one

I input the example result into the box instead of my input so I have to wait an entire minute

[vivid fractal]

lol my code gives a floating point result so i retyped it into the box and mistype

[lilac nacelle]

what is this

solving linear equations?

or am i missing something

(part 1)

[lilac nacelle]

@vivid fractal am i wrong 😭
why is there a minimum when theres only one possible solution

[vivid fractal]

ur supposed to use dp

i think

but do you need to?

different question

[lilac nacelle]

madoomed

solving equatin worked for test example

[lilac nacelle]

any reason it wouldnt work for full example?

getting too low

[vivid fractal]

the world is your cloyster

https://tenor.com/view/anime-pokemon-kingler-crab-hammer-gif-12668960

[stone belfry]

What about something like this
Button A: X+50, Y+5
Button B: X+10, Y+1
Prize: X=60, Y=6

[lilac nacelle]

bruh

[bronze hazel]

my matrix math is off somewhere

[lilac nacelle]

You estimate that each button would need to be pressed **no more than 100 times **to win a prize

is this a condition

[stone belfry]

I reaaaally don't like todays question

[stone belfry]

||This also doesn't come into play, at least in my input||

[lilac nacelle]

||maybe its a way to limit our search space||

[lilac nacelle]

this kinda of case doesnt seem to be there in my input 🤔

||nvm it was because i Floored instead of Round||

[stone belfry]

||it isn't in my input either||

[lilac nacelle]

it worked for part 1.. should technically work for part 2 as well

wow it did

[bronze hazel]

afaik 2x2 systems of equations only have more than one answer when one equation can simplify to the other and then they have infinite solutions

[lilac nacelle]

yea

and that type of input doesnt seem to appear

so whats the with "minimum cost"

[tardy hatch]

trying to throw you off perhaps

[lilac nacelle]

thought i had a chance to use binary search finally

[vivid fractal]

precisely

[vivid fractal]

you're not meant to linear eq it afaik

but it could just be bait who know

[normal sonnet]

I did and it worked but it was nasty

underline, emphasis, italics on nasty

[bronze hazel]

what was the intended way?

[normal sonnet]

do the math right so u don't have rounding errors. Other than that I can't say

[bronze hazel]

i just used integer division and checked the answer it spit out against the prize coordinates

not the most efficient way but it worked

[normal sonnet]

I'm probably just dumb and kept doing the algebra wrong lol

[bronze hazel]

i just built matrix and vec structs with methods for getting the determinant, inverting, and multiplication, it worked well enough but i had to apply the scalar division by the determinant last when using integers otherwise everything is 0

[normal sonnet]

shut up I'm stupid this is probably why I had rounding issues

[unborn harness]

basic ||arithmetics||, you just solve a ||system of 2 equations||
today i did my best time in 3 years

[vivid fractal]

i dont think its the intended way though

im pretty sure you can solve it entirely discretely with dynamic programming

[bronze hazel]

even part 2?

[soft bramble]

What how

[vivid fractal]

not htinking about it

its like those other dynamic programming problems

idk abt part 2

pretty sure htose constraints exist for a reason

[normal sonnet]

i.e. the knapsack problem I think

or maybe the counting change one

[bronze hazel]

could you do something similar to the stones problem

or maybe similar to day 7

where you have 3 different operations to choose from at every step and they all might get you closer or farther from the goal, here you have two operations with a cost associated

idk how it would work on part 2 because its all unbounded

[normal sonnet]

idk man what do I look like a scientist or something

[uneven whale]

I'm doing it in typescript this year and just spend the afternoon trying to figure out why my calculation is so off and looking into how to handle || large numbers ||... well, then I noticed this.

makes me miss Go