₦airaland Forum

Deicide:
if you sorting then the time complexity is not O(n)

yeah, I missed that, in the worst case it's O n log n i.e there are no duplicates but in the average the time should be lower.

Ogori8:
As an engineering student, which between .net programming and python would you advice me to learn?

If you're interested in anything data/ai python else .Net. It doesn't matter which you choose to learn first though, since .net is a full ecosystem its gonna more be difficult to wrap your head around.

TastyFriedPussy:
One of the most ugly pieces of web dev sh*t I have ever seen in my life

cope

[quote author=tensazangetsu20 post=115782462][/quote]Wow, at least they are still getting paid, they just need to improve on programming, which shouldn't be hard at these point, they're still in pretty good shape, even if they get fired lol.

SavageBoy:
In order words, a solid knowledge in DSA will give one the Davidic courage to face the Goliaths of the tech industry. That's interesting.

unsurprisingly yes, there's a guy I know, he'll basically get any job if he's able to land the interview, and it's not because of his stack rather it's DSA and system design, basically get so good at it that you stun the interviewer(s).

If you really want a comp. science degree well you could buy one, depends on the university though, cost is about 500k.

DrLevi:
The FAANG dream is a highly ambitious one but the Leetcode DSA thing is quite real especially with established companies, I hear American devs talk about it a lot.

Still your employment will always require expertise with certain languages and frameworks because you have to be productive immediately after the onboarding.

Advising people ignore productive skills for optimisation hacks and efficiency(DSA) is quite misleading because at the end, you won't be employed as a DSA engineer but a ROR or React or Python developer.
Makes no sense passing the interview only to get fired a month into the job for being totally useless to the team

Onboarding in Google and generally FAANG takes 3 to 6 months, that's enough time to learn any tech stack if you're competent enough, though this is not the case for mid and smaller tech companies, you'll need a stack or language at the very least.

MadarasBlade:
I just downloaded the book lol, i am never leaving nairaland.

I was learning how to read and write maths proofs as a prerequisite for learning dsa before now, it seems this book's writing is in plain english. Thanks bro

uwc. There are many other books and courses in plain English too, the book will give you a high level overview of all DSA but you'll probably need more, your knowledge of proofs would come handy some other time.

TastyFriedPussy:
exactly. Respects when you fork the logic in reverse cow bro style, the decency will dramatically release

werey

tensazangetsu20:
I have already read as many DSA books as I could. Grokking algorithms, computer science distilled and cracking the coding interview. Focus now is on leetcoding myself to death.

Nice, for me, I decided to go a bit deeper since much of my exposure to DSA is from blog posts, random vids and the book "a common sense guide to data structures and algorithms". There are good books like clrs, algorithm design manual e.t.c but I'm not a fan large readings, evident that the book above is the only "programming" book I've completed even though I have tons of them even the ones you listed in my pdf library lol.

tensazangetsu20:
I tried learning this too. Just a huge clusterfuck. Probably cause I no send any new tools again till I get insanely good at data structures.

Yeah, though the borrow checker model is actually good, if it compiles, then it's guaranteed to run.
Speaking of data structures, are you doing Leetcode only? or is it in combination with a course or book.

For me it's rust, at the start it was doable but when I tried anything substantial the complexity snowballs.

dodgelord:
Programming is nothing without If & else Prove me wrong

Smiles in pattern matching.

Deicide:
Using your python logic in rust, code translate to

fn shuffle(nums: &mut Vec<i32> ) -> Vec<i32> {
    let mut rng = rand::thread_rng();
    for num in 0..nums.len() {
        let x: u32 = rng.gen_range(0..nums.len() as u32);
        nums.swap(num, x as usize);
    }
    return nums.to_vec();
}

This is fucking fast, cause it's using multiple thread and random results are cached

Now I can finally rest and watch the sun set on a great universe

I see you rustecean, flexing Rust features anyhow
You can't rest o, it's night here already

chukwuebuka65:
Yea, but I think u didnt get the concept. If you get the concept u can fix it.

What concept? unless you want me to debug your code lol.

chukwuebuka65:
function shuffle( arr ) {
Const seenArray = [ ]
Const shuffledArray = [ ]
for ( i=0,i<arr.length-1,i++) {
let randomIndex = math.floor( math.random( )
*arr.length)
If (!seenArray.include(randomIndex)) {
shuffledArray [ randomIndex] = arr[i]
seenArray[i] = randomIndex
} else {
for ( k=0,k<arr.length,k++) {
If ( shuffledArray[k] == null) {
shuffledArray[k] = arr[i]
break;
}
}

}


}
Return shuffledArray;
}

God of bugs
In all seriousness lots of syntax errors and even after fixing the errors does not give the desired output. I bet you typed it on nairaland.

chukwuebuka65:
I think there is a chance this might go into infinite loop

I know, this is not my solution, I was trying to remove the recursion from op's solution, mine's on the first page of the thread.

Deicide:
That's the thing about programming if your solution is not optimised you would know. So there is always room for growth till you move to the next thing.

The randrange you used in python should also depend on random but it's abstracted though am not sure of this.

No it's not like that. In my case, any value returned by the random() function would be used. In your case any value wouldn't do, it would have to return a new value not inside of the HashSet.

Deicide:
Yeah, Things I learnt on leetcode

Yes the complexity is dependent on the random number. So it can be infinite ): so don't try on large dataset.

Another way to express your solution without the recursion would be;

function shuffle(Array arr) 
{
   HashSet seenIndices = new HashSet()
   Array result = new Array()
   for (int i in 0 to arr.length())
   {
      rand_idx = random num in range(0, arr.length())
      while (seenIndices.contains(rand_idx)
      {
        rand_idx = random num in range(0, arr.length())
      }
      result.add(arr[rand_idx])
      seenIndices.add(rand_idx)
    }
  return result;
}

This I think follows the same idea, but would still depend on the random function, so not recommended.

Deicide:
That's why this code wouldn't work if they we duplicate in the input array. Who knows maybe if I stroll out later the solution might come to my head

You should definitely go for walk
You used the hashset to reduce lookup from O(n) to O(1) didn't you? nice one.
Also an instance where random() keeps returning an index to a value that's already in the hashset continuously till count gets to zero would yield undesired results.

Deicide:
like this shuffle([1,2,3,4,5], 5);
5 here is the length of the vector.

Ok sure.

Deicide:
Even on reddit they would advise you not to implement a double linked list or even a Linked list in rust exexpt you wanna run mad But you can use Box<T> to make the running mad slower

My code actually runs fast despite the recursion, it mostly just depends on if it gets the random number right. Best case would be O(n) but that's impossible but it can happen just that the chance are small especially if the input is large.

I'll rewrite this code in C++ so I can test the complexity properly. The lazy_load for HasSet I did up there I don't have the crate for it

I remember the Box<T> construct lol, and to think I was able to implement a linkedlist (single) in it I don't think I have the code anymore but I'll check my GitHub later.

Ofcourse it would run fast, rust if fast af. I do think it's not that great to let your algorithm depend on "picking the right number" but I get you.

Deicide:
That's the base case of the recursion; if that's not there the code would run forever

Ok, say for instance I called the function on a vector, [1,2,3,4,5];
what would pass as argument for the count parameter?

Deicide:
It'll be easy with Js

Sorry, what's the role of the count parameter?

Deicide:
Yeah I agree there are something that are Impossible in rust without using the unsafe key work like improbable did in my code to reduce time & space complexity. But I love the language it makes you as a programmer to be very careful with what you write.

Can you guess the complexity of my code since lookup in a hashmap is O(1) so am thing O(n^2 + random(n)) I don't even know what that means

My guy even with the unsafe keyword it gets complex real quick, I know what I faced when I tried implementing a doubly linkedlist in Rust, it was crazy. Memory management in C/C++ is much simpler than in Rust.
I agree though learning rust will make you better developer overall.

Guy you're brave, using recursion inside a while loop, I hail thee loolz, well let me see if I can figure out the complexity...

Deicide:
Nice one

I modified my post, check it out.

Ikechukwu183:
Brilliant one

Python for the win

Deicide:
Was able to come up with this using rust.
...
Can someone improve on this? use whatever programming language you like. And are there other rust programmers on this section? Complexity of this code? O(2^n) yeah it's that terrible.

There is a problem that causes infinity loop in this code can you spot it?

from random import randrange

def shuffle(arr):
  for i in range(len(arr)):
    rand_idx = randrange(len(arr))
    arr[rand_idx], arr[i] = arr[i], arr[rand_idx]
  return arr
  
  
arr = [1,2,3,4,5]
print(shuffle(arr))

Here's something pretty simple, time complexity should be O (n log n), since the time complexity of the randrange function is O (log n).
It's being quite a while since I did anything in rust, I don't want to fight the borrow checker right now Rust used to be my favorite language but it's just too painful to do anything substantial with it, try implementing a doubly linkeldlist in Rust and you'll agree that C++ is easy compared to Rust.

benalvino3:
I apologize if i wrongly tagged you...
it is not a matter of preference because both of them do not do the same thing or try to solve the same problem.
CSS is not an alternative to c++ that is the point I am trying to make. it will make sense to say i rather touch C++ over C, go, java, c#, etc than CSS.

I believe I'm the one you wanted to quote.

What I'm really saying is, I'd rather implement a functionality in C++ (difficult lang) than design something in CSS, obviously it's not directly comparison.

truthCoder:
learn tailwind css.
thank me later

Sure, thanks.

LienwaltAbel:
I still dey try find my level. It's been long since I last programmed, I'm thinking of getting back to learning python and C++ Sha. Then I'll try my luck with gameplay programming and backend (maybe). The issue now is that I need money for laptop (I'm on that for now).

Ma guy gameplay dry o, hope say you know wetin dey sup currently. Backend and mobile dev seem the most reasonable right now, front end dev is getting saturated.

silento:
If u know basic fundamental concepts of programming

You can learn any language and build complex stuff with it under two weeks

Css is also not hard , c is the hardest language I have come across

Am not talking about c improved with libraries
All the higher languages allows u to mess up and get away with it

You see pointer in c language na there most people dey run and that's the most powerful stuff in c language

If u understand pointer in c language very well
Just know u are good to go

Am talking from experience

If u are not in a hurry to make money start programming with c language then try to understand the fundamental concepts of programming

Syntax
Variable
Datatypes and structures
Operators
Decision making
Loops
Files IO
Function and oop then if language supports oop

This are things u need to understand very well and u will see programming is very easy and fun

Why would you consider C the most difficult? pointers? memory management?
C is not that hard to me, however it requires you to implement the basic things that come built-in in other languages, which is where I agree the difficulty really creeps in.