Related
I want to implement the Hilbert Transform in C#. From this article I saw that the fastest FFT open source implementation seems to be the FFTW, so I downloaded that example and used it to learn how to use the fftw wrapper for C#.
I have a current signal of 200.000 points which I'm using for testing. Getting the Hilbert transform through the fft is relatively simple:
Compute the fft.
Multiply by 2 all positive frequencies except for the DC and Nyquist components (0 and n/2 + 1, if the sample size is even).
Multiply by 0 all the negative frequencies ([n/2 + 1, n]).
Compute the inverse fft.
This far, I've done all of it. The only problem is the inverse fft. I'm not able to get the same results with fftw than with the ifft from Matlab.
My code
RealArray _input;
ComplexArray _fft;
void ComputeFFT()
{
_fft = new ComplexArray(_length / 2 + 1);
_input.Set(Data);
_plan = Plan.Create1(_length, _input, _fft, Options.Estimate);
_plan.Execute();
}
This far, I've a fft with only the positive frequencies. So I don't need to multiply by zero the negative frequencies: they don't even exist. With the following code, I can get my original signal back:
double[] ComputeIFFT(ComplexArray input)
{
double[] temp = new double[_length];
RealArray output = new RealArray(_length);
_plan = Plan.Create1(_length, input, output, Options.Estimate);
_plan.Execute();
temp = output.ToArray();
for (int i = 0; i < _length; ++i)
{
temp[i] /= _length;
}
return temp;
}
The problem comes when I try to get a complex inverse from the signal.
void ComputeHilbert()
{
double[] fft = FFT.ToArray();
double[] h = new double[_length / 2 + 1];
double[] temp = new double[_length * 2];
bool fftLengthIsOdd = (_length | 1) == 1;
h[0] = 1;
for (int i = 1; i < _length / 2; i++) h[i] = 2;
if (!fftLengthIsOdd) h[(_length / 2)] = 1;
for (int i = 0; i <= _length / 2; i++)
{
temp[2 * i] = fft[2*i] * h[i];
temp[2 * i + 1] = fft[2*i + 1] * h[i];
}
ComplexArray _tempHilbert = new ComplexArray(_length);
_tempHilbert.Set(temp);
_hilbert = ComputeIFFT(_tempHilbert);
_hilbertComputed = true;
}
It's important to note that, when I do apply the ToArray() method on a ComplexArray object, I get as result a double[] with twice as length as the original array, having the real and imaginary parts consecutive. That's it, for a ComplexArray object containing "3 + 1i", I would get a double vector with [3, 1].
So, at this moment, what I have is something like:
[DC Frequency, 2*positive frequencies, Nyquist Frequency, zeros]
If I export this data to Matlab and compute the IFFT, I get the same result as its hilbert(signal).
However, if I try to apply the IFFT provided by fftw, I get weird values from Nyquist Frequency to the end (that is to say, the zeros mess with fftw).
This is the ifft I'm using to do this:
double[] ComputeIFFT(ComplexArray input)
{
double[] temp;
ComplexArray output = new ComplexArray(_length);
_plan = Plan.Create1(_length, input, output, Direction.Backward, Options.Estimate);
_plan.Execute();
temp = output.ToArray();
for (int i = 0; i < _length; ++i)
{
temp[i] /= _length;
}
return temp;
}
So, just to sum it up, my problem is the way I'm using to calculate the ifft. It doesn't seems to work well with zeros. Or maybe Matlab is capable to understand that it has to apply some different approach and I should do it manually, but I don't know how.
Thank you very much for your help in advance, much appreciated!
So the problem was the ComputeIFFT function. In the for loop, I was doing i < _length, but the length of temp array is 2 * _length, because it holds both real and imaginary values.
That's why I only got half of the values right.
The correct code for it is:
double[] ComputeIFFT(ComplexArray input)
{
double[] temp;
ComplexArray output = new ComplexArray(_length);
_plan = Plan.Create1(_length, input, output, Direction.Backward, Options.Estimate);
_plan.Execute();
temp = output.ToArray();
for (int i = 0; i < temp.Length; ++i)
{
temp[i] /= _length;
}
return temp;
}
I hope this will be useful for anyone trying to implement the Hilbert Transform through FFTW in C#.
I'm taking the Coursera machine learning course right now and I cant get my gradient descent linear regression function to minimize. I use: one dependent variable, an intercept, and four values of x and y, therefore the equations are fairly simple. The final value of the Gradient Decent equation varies wildly depending on the initial values of alpha and beta and I cant figure out why.
I've only been coding for about two weeks, so my knowledge is limited to say the least, please keep this in mind if you take the time to help.
using System;
namespace LinearRegression
{
class Program
{
static void Main(string[] args)
{
Random rnd = new Random();
const int N = 4;
//We randomize the inital values of alpha and beta
double theta1 = rnd.Next(0, 100);
double theta2 = rnd.Next(0, 100);
//Values of x, i.e the independent variable
double[] x = new double[N] { 1, 2, 3, 4 };
//VAlues of y, i.e the dependent variable
double[] y = new double[N] { 5, 7, 9, 12 };
double sumOfSquares1;
double sumOfSquares2;
double temp1;
double temp2;
double sum;
double learningRate = 0.001;
int count = 0;
do
{
//We reset the Generalized cost function, called sum of squares
//since I originally used SS to
//determine if the function was minimized
sumOfSquares1 = 0;
sumOfSquares2 = 0;
//Adding 1 to counter for each iteration to keep track of how
//many iterations are completed thus far
count += 1;
//First we calculate the Generalized cost function, which is
//to be minimized
sum = 0;
for (int i = 0; i < (N - 1); i++)
{
sum += Math.Pow((theta1 + theta2 * x[i] - y[i]), 2);
}
//Since we have 4 values of x and y we have 1/(2*N) = 1 /8 = 0.125
sumOfSquares1 = 0.125 * sum;
//Then we calcualte the new alpha value, using the derivative of
//the cost function.
sum = 0;
for (int i = 0; i < (N - 1); i++)
{
sum += theta1 + theta2 * x[i] - y[i];
}
//Since we have 4 values of x and y we have 1/(N) = 1 /4 = 0.25
temp1 = theta1 - learningRate * 0.25 * sum;
//Same for the beta value, it has a different derivative
sum = 0;
for (int i = 0; i < (N - 1); i++)
{
sum += (theta1 + theta2 * x[i]) * x[i] - y[i];
}
temp2 = theta2 - learningRate * 0.25 * sum;
//WE change the values of alpha an beta at the same time, otherwise the
//function wont work
theta1 = temp1;
theta2 = temp2;
//We then calculate the cost function again, with new alpha and beta values
sum = 0;
for (int i = 0; i < (N - 1); i++)
{
sum += Math.Pow((theta1 + theta2 * x[i] - y[i]), 2);
}
sumOfSquares2 = 0.125 * sum;
Console.WriteLine("Alpha: {0:N}", theta1);
Console.WriteLine("Beta: {0:N}", theta2);
Console.WriteLine("GCF Before: {0:N}", sumOfSquares1);
Console.WriteLine("GCF After: {0:N}", sumOfSquares2);
Console.WriteLine("Iterations: {0}", count);
Console.WriteLine(" ");
} while (sumOfSquares2 <= sumOfSquares1 && count < 5000);
//we end the iteration cycle once the generalized cost function
//cannot be reduced any further or after 5000 iterations
Console.ReadLine();
}
}
}
There are two bugs in the code.
First, I assume that you would like to iterate through all the element in the array. So rework the for loop like this: for (int i = 0; i < N; i++)
Second, when updating the theta2 value the summation is not calculated well. According to the update function it should be look like this: sum += (theta1 + theta2 * x[i] - y[i]) * x[i];
Why the final values depend on the initial values?
Because the gradient descent update step is calculated from these values. If the initial values (Starting Point) are too big or too small, then it will be too far away from the final values (Final Value). You could solve this problem by:
Increasing the iteration steps (e.g. 5000 to 50000): gradient descent algorithm has more time to converge.
Decreasing the learning rate (e.g. 0.001 to 0.01): gradient descent update steps are bigger, therefore it converges faster. Note: if the learning rate is too small, then it is possible to step through the global minimum.
The slope (theta2) is around 2.5 and the intercept (theta1) is around 2.3 for the given data. I have created a github project to fix your code and i have also added a shorter solution using LINQ. It is 5 line of codes. If you are curious check it out here.
I'm trying to randomly generate blocks on a flat map and make it so that they don't overlap each other.
I have made a matrix (c# array) of the size of the map (500x500), the blocks have a scale between 1 and 5.
The code works but if a generated block overlaps another one, it is destroyed and not regenerated somewhere else.
Only around 80 of the 1000 blocks I try to generate don't overlap another block.
Here is a picture of the map with around 80 blocks generated, the green squares are blocks
void generateElement(int ratio, int minScale, int maxScale, GameObject g) {
bool elementFound = false;
for (int i = 0; i < ratio * generationDefault; i++) {
GameObject el;
// Randomly generate block size and position
int size = Random.Range(minScale, maxScale + 1);
int x = Random.Range(0, mapSizex + 1 - size);
int y = Random.Range(0, mapSizey + 1 - size);
// Check if there is already an element
for (int j = x; j < x + size; j++)
for (int k = y; k < y + size; k++)
if (map[j][k] != null)
elementFound = true;
if (elementFound)
continue;
else {
el = (GameObject)Instantiate(g, new Vector3(x + (float)size / 2, (float)size / 2, y + (float)size / 2), Quaternion.Euler(0, 0, 0));
el.transform.localScale *= size;
}
// Create element on map array
for (int j = x; j < x + size; j++)
for (int k = y; k < y + size; k++)
if (map[j][k] == null) {
map[j][k] = el.GetComponent<ObjectInterface>();
}
}
}
I thought of 3 possible fixes
I should set the size of the block depending of the place it has.
I should use another randomization algorithm.
I'm not doing this right.
What do you think is the best idea ?
UPDATE
I got the code working much better. I now try to instantiate the blocks multiple times if needed (maximum 5 for the moment) and I fixed the bugs. If there are already many elements on the map, they will not always be instantiated and that's what I wanted, I just have to find the right amount of times it will try to instantiate the block.
I tried instantiating 1280 elements on a 500x500 map. It takes only about 1.5 second and it instantiated 1278/1280 blocks (99.843%).
void generateElement(int ratio, int minScale, int maxScale, GameObject g) {
bool elementFound = false;
int cnt = 0;
// Generate every block
for (int i = 0; i < ratio * generationDefault; i++) {
GameObject el = null;
// Randomly generate block size and position
int size, x, y, tryCnt = 0;
// Try maximum 5 times to generate the block
do {
elementFound = false;
// Randomly set block size and position
size = Random.Range(minScale, maxScale + 1);
x = Random.Range(0, mapSizex + 1 - size);
y = Random.Range(0, mapSizey + 1 - size);
// Check if there is already an element
for (int j = x; j < x + size; j++)
for (int k = y; k < y + size; k++)
if (map[j][k] != null)
elementFound = true;
tryCnt++;
} while (elementFound && tryCnt < 5);
if (tryCnt >= 5 && elementFound) continue;
// Instantiate the block
el = (GameObject)Instantiate(g, new Vector3(x + (float)size / 2, (float)size / 2, y + (float)size / 2), Quaternion.Euler(0, 0, 0));
el.transform.localScale *= size;
// Create element on map array
for (int j = x; j < x + size; j++)
for (int k = y; k < y + size; k++)
if (map[j][k] == null) {
map[j][k] = el.GetComponent<ObjectInterface>();
}
cnt++;
}
print("Instantiated " + cnt + "/" + ratio * generationDefault);
}
This is incredibly difficult to do well.
Here's a quick solution you'll maybe like ... depending on your scene.
actualWidth = 500 //or whatever. assume here is square
// your blocks are up to 5 size
chunkWidth = actualWidth / 5
// it goes without saying, everything here is an int
kChunks = chunkWidth*chunkWidth
List<int> shuf = Enumerable.Range(1,kChunks).OrderBy(r=>Random.value).ToList();
howManyWanted = 1000
shuf = shuf.Take(howManyWanted)
foreach( i in shuf )
x = i % actualWidth
y = i / actualWidth
make block at x y
put block in list allBlocks
HOWEVER ............
...... you'll see that this looks kind of "regular", so do this:
Just randomly perturb all the blocks. Remember, video game programming is about clever tricks!
Ideally, you have to start from the middle and work your way out; in any event you can't just do them in a line. Shuffling is OK. So, do this ..
harmonic = 3 //for example. TRY DIFFERENT VALUES
function rh = Random.Range(1,harmonic) (that's 1 not 0)
function rhPosNeg
n = rh
n = either +n or -n
return n
function onePerturbation
{
allBlocks = allBlocks.OrderBy(r => Random.value) //essential
foreach b in allBlocks
newPotentialPosition = Vector2(rhPosNeg,rhPosNeg)
possible = your function to check if it is possible
to have a block at newPotentialPosition,
however be careful not to check "yourself"
if possible, move block to newPotentialPosition
}
The simplest approach is just run onePerturbation, say, three times. Have a look at it between each run. Also try different values of the harmonic tuning factor.
There are many ways to perturb fields of differently-sized blocks, above is a KISS solution that hopefully looks good for your situation.
Coding note...
How to get sets of unique random numbers.
Just to explain this line of code...
List<int> shuf = Enumerable.Range(1,kChunks).OrderBy(r=>Random.value).ToList();
If you are new to coding: say you want to do this: "get a hundred random numbers, from 1 to million, but with no repeats".
Fortunately, this is a very well known problem with a very simple solution.
The way you get numbers with no repeats, is simply shuffle all the numbers, and then take how many you want off the top.
For example, say you need a random couple of numbers from 1-10 but with no repeats.
So, here's the numbers 1-10 shuffled: 3,8,6,1,2,7,10,9,4,5
Simply take what you need off the front: so, 3, 8, 6 etc.
So to make an example let's say you want twelve numbers, no repeats, from 1 through 75. So the first problem is, you want a List with all the numbers up to 75, but shuffled. In fact you do that like this ..
List<int> shuf = Enumerable.Range(1,75).OrderBy(r=>Random.value).ToList();
So that list is 75 items long. You can check it by saying foreach(int r in shuf) Debug.Log(r);. Next in the example you only want 12 of those numbers. Fortunately there's a List call that does this:
shuf = shuf.Take(12)
So, that's it - you now have 12 numbers, no repeats, all random between 1 and 75. Again you can check with foreach(int r in shuf) Debug.Log(r);
In short, when you want "n" numbers, no repeats, between 1 and Max, all you have to so is this:
List<int> shuf = Enumerable.Range(1,Max).OrderBy(r=>Random.value).ToList();
shuf = shuf.Take(n);
et voilà, you can check the result with foreach(int r in shuf) Debug.Log(r);
I just explain this at length because the question is often asked "how to get random numbers that are unique". This is an "age-old" programming trick and the answer is simply that you shuffle an array of all the integers involved.
Interestingly, if you google this question ("how to get random numbers that are unique") it's one of those rare occasions where google is not much help, because: whenever this question is asked, you get a plethora of keen new programmers (who have not heard the simple trick to do it properly!!) writing out huge long complicated ideas, leading to further confusion and complication.
So that's how you make random numbers with no repeats, fortunately it is trivial.
if (elementFound) continue; will skip out this current loop iteration. You need to wrap the int x=Random..; int y=Random()..; part in a while loop with the condition being while(/* position x/y already occupued*/) { /* generate new valid point */} like this for example:
void generateElement(int ratio, int minScale, int maxScale, GameObject g) {
for (int i = 0; i < ratio * generationDefault; i++) {
GameObject el;
// Randomly generate block size and position
bool elementFound = false;
int size, x, y;
do
{
elementFound = false;
size = Random.Range(minScale, maxScale + 1);
x = Random.Range(0, mapSizex + 1 - size);
y = Random.Range(0, mapSizey + 1 - size);
// Check if there is already an element
for (int j = x; j < x + size; j++)
for (int k = y; k < y + size; k++)
if (map[j][k] != null)
elementFound = true;
} while(elementFound);
el = (GameObject)Instantiate(g, new Vector3(x + (float)size / 2, (float)size / 2, y + (float)size / 2), Quaternion.Euler(0, 0, 0));
el.transform.localScale *= size;
// Create element on map array
for (int j = x; j < x + size; j++)
for (int k = y; k < y + size; k++)
if (map[j][k] == null) {
map[j][k] = el.GetComponent<ObjectInterface>();
}
}
}
You shouldn't be getting that many collisions.
Assuming your blocks were ALL 5 units wide and you're trying to fit them into a grid of 500,500 you would have 100*100 spaces for them at minimum, which gives 10,000 spaces into which to fit 1,000 blocks.
Try playing around with this code:
using System;
using System.Collections.Generic;
namespace ConsoleApplication1
{
class Program
{
static void Main()
{
var result = PlaceNonOverlappingBlocks(1000, 5, 500, 500);
}
static List<Block> PlaceNonOverlappingBlocks(int count, int maxBlockSize, int mapX, int mapY)
{
var map = new bool[mapY, mapX];
var rng = new Random();
var result = new List<Block>(count);
int collisions = 0;
while (count > 0)
{
int size = rng.Next(1, maxBlockSize + 1);
int x = rng.Next(0, mapX - size);
int y = rng.Next(0, mapY - size);
if (fits(map, x, y, size))
{
result.Add(new Block(x, y, size));
addToMap(map, x, y, size);
--count;
}
else
{
if (++collisions> 100000)
throw new InvalidOperationException("Hell has frozen over");
}
}
// This is just for diagnostics, and can be removed.
Console.WriteLine($"There were {collisions} collisions.");
return result;
}
static void addToMap(bool[,] map, int px, int py, int size)
{
for (int x = px; x < px+size; ++x)
for (int y = py; y < py + size; ++y)
map[y, x] = true;
}
static bool fits(bool[,] map, int px, int py, int size)
{
for (int x = px; x < px + size; ++x)
for (int y = py; y < py + size; ++y)
if (map[y, x])
return false;
return true;
}
internal class Block
{
public int X { get; }
public int Y { get; }
public int Size { get; }
public Block(int x, int y, int size)
{
X = x;
Y = y;
Size = size;
}
}
}
}
I implemented this algorithm to solve Leetcode's #28 question "implemment strStr()".
Problem description:
Implement strStr(),
Returns the index of the first occurrence of needle in haystack, or -1 if needle is not part of haystack.
My code was implemented based on http://www.geeksforgeeks.org/searching-for-patterns-set-3-rabin-karp-algorithm/ 's tutoril.
I found out that with different prime number I use to scale down the hash, the function could go wrong.
Here is my code:
public class Solution {
public int StrStr(string haystack, string needle) {
int len = needle.Length;
//2 special case
if (haystack.Length < len) return -1;
if (needle == "") return 0;
//base prime number used for rabin-karp's hash function
int basement = 101;
//prime number used to scale down the hash value
int prime = 101;
//the factor used to multiply with the character to be removed from the hash
int factor = (int)(Math.Pow(basement, needle.Length - 1)) % prime;
//get ascii value of the needle and the initial window
int needleHash = 0;
int windowHash = 0;
byte[] needleBytes = Encoding.ASCII.GetBytes(needle);
byte[] windowBytes = Encoding.ASCII.GetBytes(haystack.Substring(0, len));
//generate hash value for both
for (int i = 0; i < needle.Length; i++)
{
needleHash = (needleHash * basement + needleBytes[i]) % prime;
windowHash = (windowHash * basement + windowBytes[i]) % prime;
}
//loop to find match
bool findMatch = true;
for (int i = 0; i < haystack.Length - len + 1; i++){
//if hash value matches, incase the hash value are not uniq, iterate through needle and window
if(needleHash == windowHash){
findMatch = true;
for (int j = 0; j < len; j++)
{
if (needle[j] != haystack[i + j])
{
findMatch = false;
break;
}
}
if (findMatch == true) return i;
}
//move the sliding window and find the hash value for new window
if (i < haystack.Length - len){
byte removeByte = Encoding.ASCII.GetBytes(haystack.Substring(i, 1))[0];
byte addByte = Encoding.ASCII.GetBytes(haystack.Substring(i + len, 1))[0];
//function of rolling hash
windowHash = ((windowHash - removeByte * factor) * basement + addByte) % prime;
//ensure the window hash to be positive
if(windowHash < 0) windowHash += prime;
}
}
return -1;
}
}
With "prime" set to "101", this code passes all test. But if i change it to other prime number no matter smaller or larger (example: 17, 31, 103), it always failed at test "68/72" which
haystack = "baabbaaaaaaabbaaaaabbabbababaabbabbbbbab
babbbbbbabababaabbbbbaaabbbbabaababababbbaabbbbaaabbaaba
bbbaabaabbabbaaaabababaaabbabbababbabbaaabbbbabbbbabbabbaabbbaa";
needle = "bbaaaababa";
Thus I do believe my code have big issues that I could not detect. Why does that happen?
Two issues with your code:
basement should be set at 256 per the algorithm at the site you used as a reference. This value is the number of characters in the input alphabet.
Your computation of factor is not correct; you're casting to int before calculating the remainder. The Math.Pow operation results in a double value that is larger than Int32.MaxValue. When you cast to int before the modulo operation you truncate this value. You need to perform the modulo with the double value and then cast to int. The line should look like this:
int factor = (int)((Math.Pow(basement, needle.Length - 1)) % prime);
I tested your code with these modifications and the example given and it works for primes of 17, 31, 101 and 103.
I have a datastructure with a field of the float-type. A collection of these structures needs to be sorted by the value of the float. Is there a radix-sort implementation for this.
If there isn't, is there a fast way to access the exponent, the sign and the mantissa.
Because if you sort the floats first on mantissa, exponent, and on exponent the last time. You sort floats in O(n).
Update:
I was quite interested in this topic, so I sat down and implemented it (using this very fast and memory conservative implementation). I also read this one (thanks celion) and found out that you even dont have to split the floats into mantissa and exponent to sort it. You just have to take the bits one-to-one and perform an int sort. You just have to care about the negative values, that have to be inversely put in front of the positive ones at the end of the algorithm (I made that in one step with the last iteration of the algorithm to save some cpu time).
So heres my float radixsort:
public static float[] RadixSort(this float[] array)
{
// temporary array and the array of converted floats to ints
int[] t = new int[array.Length];
int[] a = new int[array.Length];
for (int i = 0; i < array.Length; i++)
a[i] = BitConverter.ToInt32(BitConverter.GetBytes(array[i]), 0);
// set the group length to 1, 2, 4, 8 or 16
// and see which one is quicker
int groupLength = 4;
int bitLength = 32;
// counting and prefix arrays
// (dimension is 2^r, the number of possible values of a r-bit number)
int[] count = new int[1 << groupLength];
int[] pref = new int[1 << groupLength];
int groups = bitLength / groupLength;
int mask = (1 << groupLength) - 1;
int negatives = 0, positives = 0;
for (int c = 0, shift = 0; c < groups; c++, shift += groupLength)
{
// reset count array
for (int j = 0; j < count.Length; j++)
count[j] = 0;
// counting elements of the c-th group
for (int i = 0; i < a.Length; i++)
{
count[(a[i] >> shift) & mask]++;
// additionally count all negative
// values in first round
if (c == 0 && a[i] < 0)
negatives++;
}
if (c == 0) positives = a.Length - negatives;
// calculating prefixes
pref[0] = 0;
for (int i = 1; i < count.Length; i++)
pref[i] = pref[i - 1] + count[i - 1];
// from a[] to t[] elements ordered by c-th group
for (int i = 0; i < a.Length; i++){
// Get the right index to sort the number in
int index = pref[(a[i] >> shift) & mask]++;
if (c == groups - 1)
{
// We're in the last (most significant) group, if the
// number is negative, order them inversely in front
// of the array, pushing positive ones back.
if (a[i] < 0)
index = positives - (index - negatives) - 1;
else
index += negatives;
}
t[index] = a[i];
}
// a[]=t[] and start again until the last group
t.CopyTo(a, 0);
}
// Convert back the ints to the float array
float[] ret = new float[a.Length];
for (int i = 0; i < a.Length; i++)
ret[i] = BitConverter.ToSingle(BitConverter.GetBytes(a[i]), 0);
return ret;
}
It is slightly slower than an int radix sort, because of the array copying at the beginning and end of the function, where the floats are bitwise copied to ints and back. The whole function nevertheless is again O(n). In any case much faster than sorting 3 times in a row like you proposed. I dont see much room for optimizations anymore, but if anyone does: feel free to tell me.
To sort descending change this line at the very end:
ret[i] = BitConverter.ToSingle(BitConverter.GetBytes(a[i]), 0);
to this:
ret[a.Length - i - 1] = BitConverter.ToSingle(BitConverter.GetBytes(a[i]), 0);
Measuring:
I set up some short test, containing all special cases of floats (NaN, +/-Inf, Min/Max value, 0) and random numbers. It sorts exactly the same order as Linq or Array.Sort sorts floats:
NaN -> -Inf -> Min -> Negative Nums -> 0 -> Positive Nums -> Max -> +Inf
So i ran a test with a huge array of 10M numbers:
float[] test = new float[10000000];
Random rnd = new Random();
for (int i = 0; i < test.Length; i++)
{
byte[] buffer = new byte[4];
rnd.NextBytes(buffer);
float rndfloat = BitConverter.ToSingle(buffer, 0);
switch(i){
case 0: { test[i] = float.MaxValue; break; }
case 1: { test[i] = float.MinValue; break; }
case 2: { test[i] = float.NaN; break; }
case 3: { test[i] = float.NegativeInfinity; break; }
case 4: { test[i] = float.PositiveInfinity; break; }
case 5: { test[i] = 0f; break; }
default: { test[i] = test[i] = rndfloat; break; }
}
}
And stopped the time of the different sorting algorithms:
Stopwatch sw = new Stopwatch();
sw.Start();
float[] sorted1 = test.RadixSort();
sw.Stop();
Console.WriteLine(string.Format("RadixSort: {0}", sw.Elapsed));
sw.Reset();
sw.Start();
float[] sorted2 = test.OrderBy(x => x).ToArray();
sw.Stop();
Console.WriteLine(string.Format("Linq OrderBy: {0}", sw.Elapsed));
sw.Reset();
sw.Start();
Array.Sort(test);
float[] sorted3 = test;
sw.Stop();
Console.WriteLine(string.Format("Array.Sort: {0}", sw.Elapsed));
And the output was (update: now ran with release build, not debug):
RadixSort: 00:00:03.9902332
Linq OrderBy: 00:00:17.4983272
Array.Sort: 00:00:03.1536785
roughly more than four times as fast as Linq. That is not bad. But still not yet that fast as Array.Sort, but also not that much worse. But i was really surprised by this one: I expected it to be slightly slower than Linq on very small arrays. But then I ran a test with just 20 elements:
RadixSort: 00:00:00.0012944
Linq OrderBy: 00:00:00.0072271
Array.Sort: 00:00:00.0002979
and even this time my Radixsort is quicker than Linq, but way slower than array sort. :)
Update 2:
I made some more measurements and found out some interesting things: longer group length constants mean less iterations and more memory usage. If you use a group length of 16 bits (only 2 iterations), you have a huge memory overhead when sorting small arrays, but you can beat Array.Sort if it comes to arrays larger than about 100k elements, even if not very much. The charts axes are both logarithmized:
(source: daubmeier.de)
There's a nice explanation of how to perform radix sort on floats here:
http://www.codercorner.com/RadixSortRevisited.htm
If all your values are positive, you can get away with using the binary representation; the link explains how to handle negative values.
By doing some fancy casting and swapping arrays instead of copying this version is 2x faster for 10M numbers as Philip Daubmeiers original with grouplength set to 8. It is 3x faster as Array.Sort for that arraysize.
static public void RadixSortFloat(this float[] array, int arrayLen = -1)
{
// Some use cases have an array that is longer as the filled part which we want to sort
if (arrayLen < 0) arrayLen = array.Length;
// Cast our original array as long
Span<float> asFloat = array;
Span<int> a = MemoryMarshal.Cast<float, int>(asFloat);
// Create a temp array
Span<int> t = new Span<int>(new int[arrayLen]);
// set the group length to 1, 2, 4, 8 or 16 and see which one is quicker
int groupLength = 8;
int bitLength = 32;
// counting and prefix arrays
// (dimension is 2^r, the number of possible values of a r-bit number)
var dim = 1 << groupLength;
int groups = bitLength / groupLength;
if (groups % 2 != 0) throw new Exception("groups must be even so data is in original array at end");
var count = new int[dim];
var pref = new int[dim];
int mask = (dim) - 1;
int negatives = 0, positives = 0;
// counting elements of the 1st group incuding negative/positive
for (int i = 0; i < arrayLen; i++)
{
if (a[i] < 0) negatives++;
count[(a[i] >> 0) & mask]++;
}
positives = arrayLen - negatives;
int c;
int shift;
for (c = 0, shift = 0; c < groups - 1; c++, shift += groupLength)
{
CalcPrefixes();
var nextShift = shift + groupLength;
//
for (var i = 0; i < arrayLen; i++)
{
var ai = a[i];
// Get the right index to sort the number in
int index = pref[( ai >> shift) & mask]++;
count[( ai>> nextShift) & mask]++;
t[index] = ai;
}
// swap the arrays and start again until the last group
var temp = a;
a = t;
t = temp;
}
// Last round
CalcPrefixes();
for (var i = 0; i < arrayLen; i++)
{
var ai = a[i];
// Get the right index to sort the number in
int index = pref[( ai >> shift) & mask]++;
// We're in the last (most significant) group, if the
// number is negative, order them inversely in front
// of the array, pushing positive ones back.
if ( ai < 0) index = positives - (index - negatives) - 1; else index += negatives;
//
t[index] = ai;
}
void CalcPrefixes()
{
pref[0] = 0;
for (int i = 1; i < dim; i++)
{
pref[i] = pref[i - 1] + count[i - 1];
count[i - 1] = 0;
}
}
}
You can use an unsafe block to memcpy or alias a float * to a uint * to extract the bits.
I think your best bet if the values aren't too close and there's a reasonable precision requirement, you can just use the actual float digits before and after the decimal point to do the sorting.
For example, you can just use the first 4 decimals (be they 0 or not) to do the sorting.