Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[MagmaMcFry]

So you want a normally distributed random variable? Which language are you using? Because there are usually normal  in the standard library. If you don't find any, you can use this method:

Let u, v be two random floats uniformly selected from the unit interval (you do know how to get regular random floating-point random numbers, right?). Let c be your peak, and s be your standard deviation. Then you can get a normally distributed random value by calculating c+s*sqrt(-2*ln(u))*cos(2*pi*v).

[Reelya]

If you need it faster, then you can pre-compute a table which converts the random output to normals, and do linear interpolation to go the precise value. Much faster than doing square roots, cos and logs every time.

Precomputing tables of values when the program starts, and using interpolation, is also good for quick and dirty cos, sqrts etc when you know the input will be in some pre-defined range. Much faster than using your languages math library functions every time, and can be made as accurate as you want without sacrificing speed (but sacrificing memory instead).

Even better if you can trick your inputs to be ints. For example, i read that in Wolfenstein and Doom they don't use 360 degree circles or radian-based circles. They define a circle with 256 "degrees". Then any angle can be expressed in exactly 1 unsigned byte, you don't need any bounds-checking when rotating, since if it goes over 255 it wraps around naturally. And they just have arrays of 256 pre-computed sin and cos values directly stuck in the code, so they can use a direct index lookup instead instead of needing computation or maths functions. In any game that you can look around but doesn't need super accuracy, using 256-degree circles has some really nice benefits.

[Darvi]

Spoiler: This one's for you guys who know greasemonkey/javascript (click to show/hide)

So I made a small script for removing that godawful "Recommended Channels" column out of my YT subscription page.

It worked quite well, except I quickly realised that it only worked after reloading the page because the column uses AJAX to create its content. So I tried to follow stackoverflow's advice and use waitForKeyElements(), except now it doesn't do anything anymore.

Code: [Select]

// ==UserScript==
// @name        NoRecommendations
// @namespace   http://localhost
// @description Removes "recommended" channels from your subscription page.
// @include     https://www.youtube.com/*
// @require     http://ajax.googleapis.com/ajax/libs/jquery/1.7.2/jquery.min.js
// @require     https://gist.github.com/raw/2625891/waitForKeyElements.js
// @grant       GM_addStyle
// ==/UserScript==

function removeElement(){
  "div.branded-page-v2-secondary-col".hide();
}

waitForKeyElements(".branded-page-v2-secondary-col", removeElement);
Anybody able to tell me what I did wrong?

Stop Misquoting Donald Knuth

Spoiler: Obligatory xkcd (click to show/hide)

Did you know that Knuth has an error decorating his wall?

[Orange Wizard]

So for a couple of hours I've been trying to create a RNG that would create a bell curve, the peak at 1, that I could also affect the spread of with a parameter.
So far I've tried doing what are essentially a bunch of dice rolls, but it seems no matter how many "dice" I have the result can vary massively. I've also tried pissing around a bit with powers but nothing much has come of that.

I'm sure there must be a better way but for the life of my can't think what it is, and I really can't progress until I've worked this out.

Python has random.normal random.gauss, both of which do that.

[Thief^]

Graknorke, what language are you using? If it's C++, the (new) C++11 random library has std::normal_distribution to do that kind of thing.

Incidentally, if you're still using rand() in C++ code, STOP. It's one of the worst random generators ever created. Do yourself a favour and learn the new C++ random library.

[hops]

Could anyone explain to me, in general Pseudocode, how I can go about writing a program that create a list containing arrays which have all the possible combinations of orders for an n number of variables?

EDIT: if my wording was confusing, I meant a program that if I, say, put in variable as the strings 'a' and 'b', then it will generate [a, b] and [b, a], and if I put in 'a', 'b', and 'c', then I will get [a, b, c], [a, c, b], [b, a, c], [b, c, a], [c, a, b] and [c, b, a].

[Putnam]

Okay, usually I just accept whatever people are doing and help out with it, but this time I can't help it.

why

[Thief^]

http://www.cplusplus.com/reference/algorithm/next_permutation/

I assume this is for homework, but that's what you'd most likely use in real life.

[Skyrunner]

Or, if you're hipster and use Python.

import itertools
array = [whatever you want to perm]
results = itertools.permutations (array)

(Also a lot less intimidating in syntax)

[hops]

Okay, usually I just accept whatever people are doing and help out with it, but this time I can't help it.

why

I'm trying to check my answer to an NP-hard problem.

[Thief^]

(Also a lot less intimidating in syntax)

I wish the c++ ranges library would hurry up so I could stop writing begin and end in every algorithm call and just pass in the container (like every other language)

[Graknorke]

Graknorke, what language are you using?

VB.net
I know, I know. I wouldn't be if school didn't force me to.

[MagmaMcFry]

Could anyone explain to me, in general Pseudocode, how I can go about writing a program that create a list containing arrays which have all the possible combinations of orders for an n number of variables?

EDIT: if my wording was confusing, I meant a program that if I, say, put in variable as the strings 'a' and 'b', then it will generate [a, b] and [b, a], and if I put in 'a', 'b', and 'c', then I will get [a, b, c], [a, c, b], [b, a, c], [b, c, a], [c, a, b] and [c, b, a].

Code: [Select]

function permlists(elements) {
  if (elements.hasAtMostOneElement()) return [elements];
  [anElement, remainingElements] = pickOneElement(elements);
  result = [];
  for (subpermlist : permlists(remainingElements)) {
    for (insertPosition : everyPossibleInsertPositionOf(subpermlist)) {
      result.append(subpermlist.copy().insertAt(insertPosition, anElement));
    }
  }
  return result;
}


[BlindKitty]

Graknorke, what language are you using?

VB.net
I know, I know. I wouldn't be if school didn't force me to.

.NET random generator always seemed pretty good to me (but I haven't tested it thoroughly), and it uses method Sample() to generate a double x such as 0 <= x < 1. Getting two of those and using Magma McFry's suggested transformation should give you what you need.

Also, use only one instance of Random class to generate all your results, don't seed it (in .NET, it should be seeded automatically with system time), and you should be good to go.

[miauw62]

If you need it faster, then you can pre-compute a table which converts the random output to normals, and do linear interpolation to go the precise value. Much faster than doing square roots, cos and logs every time.

Precomputing tables of values when the program starts, and using interpolation, is also good for quick and dirty cos, sqrts etc when you know the input will be in some pre-defined range. Much faster than using your languages math library functions every time, and can be made as accurate as you want without sacrificing speed (but sacrificing memory instead).

Even better if you can trick your inputs to be ints. For example, i read that in Wolfenstein and Doom they don't use 360 degree circles or radian-based circles. They define a circle with 256 "degrees". Then any angle can be expressed in exactly 1 unsigned byte, you don't need any bounds-checking when rotating, since if it goes over 255 it wraps around naturally. And they just have arrays of 256 pre-computed sin and cos values directly stuck in the code, so they can use a direct index lookup instead instead of needing computation or maths functions. In any game that you can look around but doesn't need super accuracy, using 256-degree circles has some really nice benefits.

The tricks that game devs came up with to go fast never cease to impress me.