Jelly Hypertraining

« first day (4 days earlier) ← previous day next day → last day (2365 days later) »

3:16 AM

in The Nineteenth Byte, 8 hours ago, by Leaky Nun

CMC: print a circle of radius n. You must pick two consistent symbols. The grid is of radius 2n+1, with center being (n,n). A grid is "colored" if the floored Euclidean distance from the grid to the center is n.

in The Nineteenth Byte, 8 hours ago, by Leaky Nun

e.g. for n=5:

in The Nineteenth Byte, 8 hours ago, by Leaky Nun

00111111100
01100000110
11000000011
10000000001
10000000001
10000000001
10000000001
10000000001
11000000011
01100000110
00111111100

in The Nineteenth Byte, 8 hours ago, by Leaky Nun

Sample implementation in Python:

in The Nineteenth Byte, 8 hours ago, by Leaky Nun

n = eval(input())
size = 2*n+1
for i in range(size):
	r = ""
	for j in range(size):
		if int(((i-n)**2 + (j-n)**2)**0.5) == n:
			r += "1"
		else:
			r += "0"
	print(r)

in The Nineteenth Byte, 8 hours ago, by Leaky Nun

Try it online!

in The Nineteenth Byte, 8 hours ago, by Leaky Nun

code-golf

@HyperNeutrino ^

user165474

Ooh.

user165474

Alright. Interesting!

Leaky Nun

@HyperNeutrino are you doing it?

user165474

I'm working on it. I'm trying to start with going as short as possible instead of making a crazy solution and then golfing it. (as in, golf the algorithm first, not the implementation later)

Leaky Nun

alright

@HyperNeutrino actually I haven't tried it lol

user165474

3:32 AM

lol I see :P

user165474

Erik hasn't been here for a long time lol :P

Leaky Nun

3:44 AM

@HyperNeutrino are you doing it?

user165474

Yes. I have Ḥ‘Ḷ_ so far to generate the distances, but I realized that that's unnecessary and can be shortened (because distances are absolute lol)

user165474

Seems to be the shortest way though. lol

user165474

Is there any way to take something like [a, b, c] and return [[[a, a], [a, b], [a, c]], [[b, a], [b, b], [b, c]], [[c, a], [c, b], [c, c]]]?

Leaky Nun

yes

user165474

I feel like it has to do with , for pair.

Leaky Nun

3:52 AM

16 bytes for me

user165474

For the whole thing?

Leaky Nun

yes

user165474

Nice.

Leaky Nun

15 bytes

user165474

Ooh. Nice.

user165474

3:56 AM

Can you give me a hint as to how to get the pairwise combinations of the array with itself?

Leaky Nun

þ

user165474

Ah. I'm using that right now, but it's giving me the pairwise combinations of the array with the range [1, argument].

user165474

Ḥ‘Ḷ_,þ

Leaky Nun

because it's still a dyad

tio.run/nexus/jelly#@69zeN////8NdYx0jP@b6JgCAA

click that

user165474

Okay.

user165474

4:00 AM

Okay, I see.

user165474

Hm.

user165474

So wait, this currently is [monad][monad][monad][dyad][product_table_dyad].

user165474

I think that [dyad] in there might be what's causing the problem?

user165474

Aha Ḥ‘Ḷ_µ,þ

Leaky Nun

I mean, the combined link is one dyad

user165474

4:05 AM

Oh.

Leaky Nun

,þ is one dyad

user165474

Oh, okay.

user165474

The above works for me for what I want, but I will keep that in mind.

user165474

I may need to change it as well.

user165474

Ḥ‘Ḷ_µ,þµ²

user165474

4:11 AM

Gives me the relative x and y absolute distances squared.

user165474

Something like this:

[[[4, 4], [1, 4], [0, 4], [1, 4], [4, 4]], [[4, 1], [1, 1], [0, 1], [1, 1], [4, 1]], [[4, 0], [1, 0], [0, 0], [1, 0], [4, 0]], [[4, 1], [1, 1], [0, 1], [1, 1], [4, 1]], [[4, 4], [1, 4], [0, 4], [1, 4], [4, 4]]]

user165474

I need to figure out how to get the sums of all of the pairs.

Leaky Nun

Did you confuse the radius with the diameter?

user165474

I hope not..

user165474

If I sum them, square root them, and floor them, it should give me the distances to the center.

Leaky Nun

4:15 AM

look at my sample output

user165474

Hm. If my implementation works correctly, it theoretically should give the correct output.

Leaky Nun

@HyperNeutrino but I only see a 5x5 grid here

user165474

This is for n = 2.

Leaky Nun

oh ok

figure out what þ does

user165474

I think it takes two arrays, takes all pairs of one of the left and one of the right array, and applies the specified dyad to them.

Leaky Nun

4:22 AM

alright

Leaky Nun

4:42 AM

@HyperNeutrino you still aren't seeing it...

user165474

Oh, I had two failed ideas, I just need to take half of each and put them together ;_;

user165474

But Dennis says there's a shorter way (unbase 1), which I should go look at to see what it doesn.

Leaky Nun

ftlog you can use + as the dyad to the outer product quick

user165474

You mean replace the , with a +?

Leaky Nun

yes

user165474

4:49 AM

That won't work.

user165474

That works for manhattan distance

Leaky Nun

square before adding

user165474

...oh

user165474

Is this what you meant? Ḥ‘Ḷ_µ²+þ

Leaky Nun

yes

user165474

4:56 AM

It gives 2 -> [[2, -1, -2, -1, 2], [3, 0, -1, 0, 3], [4, 1, 0, 1, 4], [5, 2, 1, 2, 5], [6, 3, 2, 3, 6]]

Leaky Nun

heh, the right argument is not good

use a backtick

user165474

Where?

Leaky Nun

at the end

user165474

Oh okay.

user165474

Yay.

user165474

4:59 AM

This: Ḥ‘Ḷ_µ²µ+þ: also works, but it ends with a dyad.

user165474

5:10 AM

I have a 14 byte solution. @LeakyNun

Leaky Nun

btw my solution is 11 bytes now

user165474

Nice.

user165474

Mine is:

user165474

Ḥ‘Ḷ_µ²+þ`Æ½=³Y

Leaky Nun

our codes are identical except for the Ḥ‘Ḷ_µ part

which I replaced with 2 bytes

user165474

5:13 AM

Ooh. I see.

user165474

I think I found your solution:

user165474

rN²+þ`Æ½=³Y

Leaky Nun

nice

user165474

yay :D

user165474

It hit output cache when I ran it lol

user165474

5:17 AM

cache is strange ;_;

Leaky Nun

lol

user165474

yay

user165474

Do you have another code task that I could probably complete in half an hour or so?

Leaky Nun

given two coprime integers x and y, output two integers a and b such that ax+by=1

bonus points for pointing out the relavant theorem

user165474

Hm. Interesting.

user165474

5:31 AM

Anyway, I'm too tired to continue thinking (because this requires actual math not just clever implementations), so I'll continue this challenge "tomorrow" morning. Bye!

Leaky Nun

alright

7 hours later…

user165474

12:17 PM

I think I can prove that if ax + by = 1 for coprime x and y, we can always have ax < xy.

user165474

@LeakyNun ^

user165474

Or equivalently by < xy.

user165474

Because either a or b must be negative.

Leaky Nun

sure

user165474

That just helps me restrict the values a bit, as in |a| < y and |b| < x.

Leaky Nun

12:38 PM

@HyperNeutrino are you doing it?

user165474

Yes. I'm trying to do it mathematically first and I've gotten this:

user165474

ax + by = 1
a + by/x = 1/x
a = (1 - by) / x

Leaky Nun

not sure how that helps...

user165474

So essentially I need to find b such that by - 1 is divisible by x.

user165474

I'm not very sure either, but ^

Leaky Nun

12:39 PM

alright

Leaky Nun

12:51 PM

After you finished that, you might want to attempt this

user165474

1:37 PM

Alright. I'll do that after I finish the coprime task.

Leaky Nun

nice

Leaky Nun

2:01 PM

@HyperNeutrino progress?

user165474

@LeakyNun I'm in class right now so I can't really work on it right now. I'll think about it over lunch break.

Leaky Nun

2:50 PM

@HyperNeutrino progress?

user165474

@LeakyNun I have to go now, but I will be thinking about it over lunch. Bye!

Leaky Nun

alright

« first day (4 days earlier) ← previous day next day → last day (2365 days later) »

all rooms

Transcript for

May1

May '172

May3

Practice your Jelly :) Rules and stuff are here: golfingsucces...

class code-golf jelly practice

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC