A few years ago while going through old computer science papers I stumbled upon Smoothsort Revisited by Coenraad Bron and Wim H. Hesselink which introduces a new sorting algorithm inspired by Dijkstra's smoothsort that they call poplar sort. The algorithm is based on the construction and sorting of a forest of "poplars", which is the name the authors give to some specific kinds of heap.

Back then I decided to implement this poplar sort algorithm in C++ for fun. Over the years I kept coming back to this simple algorithm, trying to find optimizations which led to a variety of experiments. Eventually I managed to turn the backing poplar heap into an implicit data structure and to implement functions for it matching the heap functions from the standard library's <algorithm> header. I turned those experiments into an "article" some time ago, so if you're interested into algorithms, data structures and C++ optimization, you can read it here: https://github.com/Morwenn/poplar-heap

Most of the space and time complexities mentioned in the repository are guesstimates waiting for proofs that I'm unlikely to provide myself. That said the same heap data structure was also described by Nicholas J. A. Harvey and Kevin C. Zatloukal under the name post-order heap and their paper has complexity proofs for similar algorithms (theirs differs slightly from ours since it uses two additional variables to keep track of the state of the heap).

The article only describes my successful experiments. If you want to look further into the research area, I've also a Gist with another experiment where I used a secondary heap (a normal binary heap this time) to try to reduce the amount of comparison performed by the sort_heap algorithm: https://gist.github.com/Morwenn/4441d4d1d01dbaf613cf539ad9c13a2a

If you are currently stuck at home like I am, I hope that this might help you to kill time a bit.

Like if you're reading someone's code and you see a function that takes an argument, how do you know what that argument should be? Should it be a tuple or a list or a string or an object defined elsewhere? In Java it's clear as day. But what about in languages where there aren't explicit data types?

Or if you wrote code and then haven't looked at it in months, and you come back and you're thinking "okay now wtf was foobar supposed to be again??"

I'm sure good variable naming convention is important here, but you can't always append "string" or "list" or "dict" to the end of every variable name, you know?

I don’t know how I got to that point of my studies , I feel like I don’t know coding to be in 300 lvl classes and it terrifies me to the point when I can’t eat and I can’t stop shaking . One of my classes will require to build a project with other people and I am scared they will laugh at me or kick me out for being bad at coding . Plus having only boys in my team also intimidates me … I am feeling so overwhelmed and just shitty and idk what to do..

many people online are emphasize so much on the relevance of the knowledge data structures and algorithm as a programmer. So it has left me wondering, if Its applies to one going into machine learning. having spent the last 3 months learning python, I dunno If I should take a detour and learn it before going on to study machine learning maths?

Thanks in advance for your time

I was thinking of starting to learn DSA and was looking for resources online.

So far, I've looked into these:

1. Algorithms and Data Structures EdX
2. Geek for geeks
3. VisuAlgo.net

I mostly like VisuAlgo coz i'm a visual learner (for the most part). I was wondering what resources for DSA do people on this subreddit love/trust the most.

There are a lot of resources on the internet on these topics. I just don't know what to select(as a beginner interested in competitive programming) in order to build a solid foundation.

I am absolutely overwhelmed. It doesn't matter if the resource is paid or not. I'm just looking for a structured plan as I'm tired of jumping from articles to articles and not seeing any knowledge gains.

Thanks in advance!

So, imma post again because I barely got any attention in my first post and I'm a desperate cry-baby! So, I need some advices. I'm kinda stuck in Data structures. I've understood the basics but when it comes to combining all the techniques to code something I get a lost and I'm up to a very frustrating point where anything I find is really basic (or same things over and over again) or too complex to the point I get unmotivated.. My question is, do I stick learning this good, or do I move on and understand as I progress? Because I know that this part is important to know well. Thanks!

Any good resources except CodeWars type of stuff? Because these don't seem to help me a lot for the moment

Loved CS50 and was fascinated by these concepts. Was wondering if anyone has recommendations for courses I can do, or should I go straight to trying leetcode?

Hello. Hope you're all having a great winter break.

It goes without saying Data Structures is one of the most well-known courses at RPI. CS students can recall the course's hardships, heartbreaks, and hair-pulling homework. And as a student taking Data Structures this spring semester for the first time, I thought I would ask these students for advice, namely what students wished they knew when first taking Data Structures.

What's the best way to study for Data Structures? Is there an online course or YouTube video playlist that you found really helped? Perhaps you can recall the title of a book that helped as a reference guide? Maybe a resource at the library, or a method of studying that made homework/exams easier?

Of course, I am going to invest as many hours into the homework and studying for exams as possible. I am going to go to office hours as often as I can. I have already completed most of the C++ crash course on Submitty. Still, the more help I can get, the better.

Thank you in advance.

Edit: Thank you so, so much for all the advice! I honestly didn't think this post would yield this much attention. I have read all your responses and really do appreciate the time you all took in making them.

I began taking a beginner Udacity course on Python (the name of which I'll add to this post once my headache goes away) a couple days ago after deciding I'm very, very tired of my current line of work. The beginning was easy enough: learn the syntax, learn the rules, learn their uses. Today I watched probably a dozen videos on sets, dictionaries, lists and whatever the hell else, and for the first time I retained abosolutely nothing. I decided enough was enough and I'm going to take a Python break for the rest of the day.

I do actually have a project in mind to start on once I get the basics, and I know that's a change of pace for this sub. I'm a former music teacher, and my project idea is a simple calculator, if you will, for generating the scale degrees given two input data. For example, if the user inputs

Starting Note: G

and

Scale Type: Major

the program outputs G A B C D E F# G precisely in that order. The plan is to add scale degree numbers, both types of solfege syllables, and whatever else I can cram in, but for now I want to focus strictly on note names.

After watching the data sctructure videos today, I realized that these structures are what I need to hang on to this information. (Note names would be strings in this case, right?) My issue is that I'm havibg trouble understanding each structure, the syntax involved, and how I even go about using and recalling these structures for output. Like, should I use a list here? A set? I plan to start from the beginning again either tonight or tomorrow, but in the meantime if there is one best answer for this project's purposes, I'd greatly appreciate some feedback. Do not write the code for me. I am not looking for answers. I am looking for advice, best practices and perhaps links to ELI5 resources.

And it's everyone's (mine) favorite time of the quarter! SEC Market Structure Data Update!

Let's see how GME has been behaving when it comes to hidden orders, odd lots, and exchange volume.

Hidden Rate

https://preview.redd.it/dn1mikup7zv71.png?width=1199&format=png&auto=webp&s=d229c3d06f506c991c14df06042fcc7e57cbe3bb

The hidden rate is calculated as total hiden orders / total orders. This order type is used to hide large sized orders (generally), which hides the submitted quantity from the order book. It's also got other nefarious purposes, notably a HFT strategy called hide n' slide.

GME: The first thing to note is that the Hidden rate for GME has increased significantly since the Jan sneeze, and has remained high since. I generally use this metric as a proxy for institutional involvement as there's an exceedingly low probability that retail investors know or bother using hidden orders.

The hidden rate remains high coming into the last quarter, sitting within the 25%-30% band, which to me indicates institutions are still very much active when it comes to trading GME.

Odd Lots

https://preview.redd.it/vbbp9i7r7zv71.png?width=1207&format=png&auto=webp&s=b0ef0bad074f2a8b367545b19dd87d1819958357

Round lots are multiples of 100, odd lots are anything outside that. A previous post I wrote here goes into much more detail about HFTs use them for order shredding and how odd lots relate to the NBBO.

GME: Converse to the hidden rate, I use odd lots as a proxy to gauge retail interest in GME, as the majority of retail are likely buying their shares in odd lots. This allows us to track if there has been any interest lost.

Has there been? Nope. There's been a slight downward trend, but overall it's remained relatively stable at ~40%. Popcorn stock is much lower, but that's due to the fact it's "easier" to buy in multiples of even lots.

Exchange Activity

Below we have two charts, the first goes into GME's trade volume distribution as a % of 100, while the second has a basket of stocks for comparison's sake. *Please note the legend is inverse to the ord

Hello, can someone explain what this line of code implies as a snippet from below code block?

c.length = filestring.length()-c.position;

I understand it as:

int variable "length" in data structure "c" assigned to the variable "filestring"'s length......And then the -c.position; part is really hard to understand to me, does anyone know what it means? Thank you SO much<3

the code block:

#include "Delete.h"
#include <string>
#include <fstream>
#include <iostream>
#include <map>
using namespace std;

struct contact {
    string name, adress, epost, birthday, others, phone;
    int position, length;
};

void Delete(){
    cout <<" Enter the name that you want to delete it"<< endl;
    string nametofind;
    getline(cin, nametofind);

    ifstream efile("contact");
    if(efile.fail())
    {
        cout << "Could not read from contact file, maybe there is no contacts" << endl;
        return;
    }

    string filestring;
    map<string, contact> contacts;
    map<string, int> positions;
    map<string, int> lengths;

    while (!efile.eof())
    {
        contact c;
        c.position = efile.tellg();
        getline(efile, c.name);
        if(c.name.length() == 0) break;
        getline(efile, c.adress);
        getline(efile, c.epost);
        getline(efile, c.birthday);
        getline(efile, c.others);
        getline(efile, c.phone);
        
        filestring+=c.name+"\n"+c.adress+"\n"+c.epost+"\n"+c.birthday+"\n"+c.others+"\n"+c.phone+"\n";
///////////////////////////////////////////////////////////////////////////////
        c.length = filestring.length()-c.position;
//////////////////////////////////////////////////////////////////////////////
        contacts[c.name] = c;
        positions[c.name] = c.position;
        lengths[c.name] = c.length;
    }

    try {
        contact c = contacts.at(nametofind);
        filestring.erase(positions[c.name], lengths[c.name]);
        ofstream ofile("contact");
        if(ofile.fail())
        {
            cout << "Could not open contact file" << endl;
            return;
        }
        ofile << filestring;
        ofile.close();
    } catch(exception e) {

I want to learn data structures and algorithms soon but I am struggling to find a course that is interesting. I have moderate programming experience and have participated on open source projects and created my own projects.

I want to learn competitive programming to get some stuff on my resume and hope to "win" some competitions.

Is it too late to learn competitive programming to get a google code jam tshirt?

Can someone link me some good websites to learn.

EDIT: Why am I downvoted?

I would like to see how LIVE professional poker players notes look like.

How their notes about their session look like?

Hand notes

Win rate notes

Table selection notes

Valuing their own game play for every session A, B, C etc.

How they process and structure all this data once it accumulates? Maybe after every session they review all their live hands and then they would chose the most interesting spots and the one they need to improve on and add them to another data base?

I have been always bad at keeping good records at school and university. In fact at two or 3 classes I was so bad at keeping pace with the lectures and making notes that I completely gave up. Sourced out the one of the best hand writers in class arranged a quick deal for the notes. And literally got better grades then her during tests etc. In fact I would try sometimes helping her during tests because I felt guilty. Without her notes there was no chance of achieving such a great results.

I can't afford same thing to happen in poker. I need to learn how to structure and process my own information otherwise I will never improve.

If by any chance someone would be willing to sale me high quality hand history data notes and every note associated with his poker career for 1 year aprox 800+ hours) how much I should pay him? It should be mainly 2/5 and 5/10 (the stakes I'm aiming at).

Is it really important to know all DataStructures when it comes to Mobile Development(Kotlin)?, i know it’s important in general and helps build good logic and algorithm, i have hands on Data Structures in Java, but the fact should i pursue more or i can make it as an Android Developer without being so good in Data Structures ?

Hi,

I've been unable to catch up what's going on in the course because I'm not that good. I can barely write programs in OOP Java, and I barely recall what log(n) and recursive are. It's still the first week of the semester, but I totally didn't know what was happening in the latter half of the last lecture this week. We were analyzing algorithms and adding up units of time per operation per line of code, then suddenly logs appeared with geometric and...

I think I'm going to do a crash course review of OOP Java in MOOC, I rented out some of the recommended books about OOP Java, I don't want to fail the course and I need to know how to help myself?

Hi. Could you recommend some books for data structures, algorithms and system design in python. I have enrolled for 2 courses on a site. Wanted your suggestions on the books. Looking towards a career change. It would be appreciative if you could help. Thanks.

Hey reddit,

Been programming for a reasonable amount of years and I know Python, R, SQL, Matlab, (basic) C++, (basic) Java + FP Scala amongst others. I recently signed an offer to start working as a data scientist after the summer, for now I'm looking at areas I can improve before I start working.

My background is masters level non-CS so I haven't formally learnt data structures and algorithms so this seems like a good place to begin (other things like testing frameworks and extra cloud skills are also high on my list). I know the basics of the most important data structures (lists, sets, hashmaps, queues) and intuitions about their corresponding time/space complexity. They were needed for some advanced courses (e.g. discrete optimisation w/ C++) together with some sorting / search algorithms but here typically only what was needed was covered.

I do kind of reason about them while writing code but maybe actually formally learning about them would make me a far better programmer. Should I potentially just try and wing it with my current knowledge, what do you think? In the past I've always avoided self-learning these as I believed my time was better spent on extra math + stat than on DS&A.

If you believe it IS a smart idea, can you recommend some resources? I know I could more or less find an 'open sourced' university course and watch the videos and do the exercises but I'm mostly looking for something maybe a tad less rigorous and more hands-on.

I started doing 6.006, and I am halfway through it. It takes me time to understand but once it clicked, man I live for this moment. But this vanishes away once I counter a problem on leetcode for a specific topic I learned. I have this issue, I can't connect between what I learned and how to use it to solve a problem. Is there something I am missing or some tips to help?

I know it’s essential for internships and the major in general. I’m pretty dedicated and I’m wondering how difficult the concepts are.

Last semester for CS111 you could watch any of the professors lectures no matter who you’re actually Professor was. I was wondering if the same would be true for data structures next semester. Thanks!

Do I need to know these (advanced, imo) computer science skills to be a statistical programmer or a biostatistician?

I was on my way to getting a computer science minor in addition to my statistics major and just finished taking the intermediate programming (in java) course last semester. It...kind of kicked my ass.

Object-oriented programming feels so uncomfortable to me, and that course wasn't even considered a weed-out. However, the next course, Data Structures & Algorithms, IS considered a weed-out.

Lordy, if I just barely passed the supposedly "easy" intro class, I'm not feeling hopeful for DS&A, a class that many people fail. Plus, programming in Java generally makes me want to die. For some reason, it just feels so much harder to me compared to programming in R or SAS.

So I am asking for some advice and advisement on whether the computer science minor is worth it to continue, given the future careers I am interested in.

Thanks!

I am a current freshman, and I want to start getting into Leet code, I've done 10 questions so far (all related to arrays and strings) and I want to be able to completely dive into LC, instead of just picking at the array problems.

What are the best resources to teach myself DS & Algo?

How long would it take to learn all the material needed to be able to comfortably jump into LC?

Thank you guys for your time

In JS, array.push() is always O(1) since even if need to 'extend' the current array as the reference to the last item is simply removed.

In Java, there's no array.push() and O(n) as need to shift all the elements to a the new array.

Is there any resources available that show different time complexities for different programming languages?

Happy 2022!

Please contact me asap

I have been searching online to see if C has any built-in data structures like C++ or Java do. For example, both C++ and Java have static arrays, dynamic arrays, linked lists, stacks, queues, hash tables, etc.... When searching for C, I've only been able to find a static array. However, everywhere I read seems to claim that C does have linked lists, stacks, queues etc.... Yet the implementations for these data structures are created manually. So from my understanding C only has integer types, char types, and static arrays. Am I correct?

I am coming from a different university and I took a C class not C++ and that was 3 years ago. The instructor told that there was population of students who didn't achieve 65% so they failed. If you do bad on 2 exams (including the makeup) you also fail the class before you take the final. He also said that he has seen too many students cheat in 2270/1300 and they were given Fs. All submitted work in 2270/1300 goes through a plagiarism checker and they are allowed to do spot interviews on your code . If you have a 55% on all tests and 100% on assignments that's a D+.

I've heard mixed things from ppl. This is a 4 unit class but is the workload equivalent to something like 8 units in your experience?

I see a lot of people, including myself took a long time to solve day 18, so I looked at my overly complicated solution to understand why.

What data structures would have made my life easier? Would a pure function or in-place manipulation been simpler?

Looking at explode, the main complexity is having access to adjacent numbers. In the tree form, its possible, but complex to implement.

The simpler approach is to just convert the tree into an >!adjacency array with depth!<, then it becomes trivial to implement the requirement deep nodes explode to the left and right. Its simply accessing >!previous array element or next array element if they exist!<.

Later in order to reduce a number, we may need to explode/split many times. Notice the requirement is >!we only split after we can no longer explode!<. This means we will attempt to explode >!and continue if it didn't!<. Implementing this requirement with pure functions is annoying as you either check to see if you can explode, then do so. Or you explode and check if the exploded result is different than the input.

Alternatively, if we implement explode >!as an in-place manipulation and have it return true if it changed the input!<, it becomes much simpler to implement reduce down the road.

The code for explode looks like

function explode(array) {
  for (let i = 0; i < array.length; i++) {
    const { value, depth } = array[i];
    if (depth > 4) {
      if (i > 0) { // exploding to the left
        array[i - 1].value += value;
      }
      if (i < array.length - 2) { // exploding to the right
        array[i + 2].value += array[i + 1].value;
      }
      array.splice(i, 2, { value: 0, depth: depth - 1 });
      return true;
    }
  }
  return false;
}

The reduce requirement after adding two arrays becomes

function reducedSum(a, b) {
  const total = addition(a, b);
  while (explode(total) || split(total)); // restarts to explode each time total is modified
  return total;
}

For the [rest of the code see paste](https://topaz.github.io/paste/#XQAAAQAkCAAAAAAAAAA0m0pnuFI8czedv0pOqa8ryE8IvQCOve70v4U063sCoYH8nauxbF6kEjGk/4/gX3d1fUdtITCQC+jx+dl5bJNjKtfTh7iefJoueZs2/piV8ucLGX1CFj5A1IOEn3zlNfTq5arkPp98sq/FsB3kRUsSSh7TOLUELWusKAGNL7qaNR/SOY6f958y5uV3k6J8Ltvpcaj2clIIqInC5qL6CnQUOZLBm5GvdgB7BFUDkEWNK1zFged13+hb4Soq9FqkwzOr8PF/Tt3AkR0ftgC7cuHQrbHSGObbvVxCbHu3Q6qoULVUN6G+EO/Lt5u9At

I have a very brief idea of data structures and algorithms but want to further develop my knowledge In these as I have been told by multiple people who are further in the field they are very important and relevant as I go on, currently I have basic knowledge in both at a beginner level any video/ course recommendations (free only please)