Difference between revisions of "User:Skilgannon/KDTree"
Jump to navigation
Jump to search
Skilgannon (talk | contribs) m (cleaning, reverse PrioQueue) |
Skilgannon (talk | contribs) m (pointer stack instead of object stack... prevents loading object contents unless the path is followed) |
||
Line 25: | Line 25: | ||
private int _nodes; | private int _nodes; | ||
private final Node root; | private final Node root; | ||
+ | private final ArrayList<Node> nodeList = new ArrayList<Node>(); | ||
//prevent GC from having to collect _bucketSize*dimensions*8 bytes each time a leaf splits | //prevent GC from having to collect _bucketSize*dimensions*8 bytes each time a leaf splits | ||
Line 37: | Line 38: | ||
//high: 2 * _dimensions * node.index + 2 * dim + 1 | //high: 2 * _dimensions * node.index + 2 * dim + 1 | ||
private final ContiguousDoubleArrayList nodeMinMaxBounds; | private final ContiguousDoubleArrayList nodeMinMaxBounds; | ||
+ | |||
/* | /* | ||
public static void main(String[] args){ | public static void main(String[] args){ | ||
Line 124: | Line 126: | ||
_dimensions = dimensions; | _dimensions = dimensions; | ||
− | //initialise this so that it ends up in 'old' memory | + | //initialise this big so that it ends up in 'old' memory |
nodeMinMaxBounds = new ContiguousDoubleArrayList(512 * 1024 / 8 + 2*_dimensions); | nodeMinMaxBounds = new ContiguousDoubleArrayList(512 * 1024 / 8 + 2*_dimensions); | ||
mem_recycle = new double[_bucketSize*dimensions]; | mem_recycle = new double[_bucketSize*dimensions]; | ||
Line 146: | Line 148: | ||
addNode.expandBounds(location); | addNode.expandBounds(location); | ||
if(location[addNode.splitDim] < addNode.splitVal) | if(location[addNode.splitDim] < addNode.splitVal) | ||
− | addNode = addNode. | + | addNode = nodeList.get(addNode.lessIndex); |
else | else | ||
− | addNode = addNode. | + | addNode = nodeList.get(addNode.moreIndex); |
} | } | ||
addNode.expandBounds(location); | addNode.expandBounds(location); | ||
Line 163: | Line 165: | ||
public ArrayList<SearchResult<T>> nearestNeighbours(double[] searchLocation, int K){ | public ArrayList<SearchResult<T>> nearestNeighbours(double[] searchLocation, int K){ | ||
− | + | //don't actually store the objects so that they aren't a possible cache miss | |
+ | //until we've verified that we need to access them | ||
+ | IntStack stack = new IntStack(); | ||
PrioQueue<T> results = new PrioQueue<T>(K,true); | PrioQueue<T> results = new PrioQueue<T>(K,true); | ||
− | stack.push(root); | + | stack.push(root.index); |
int added = 0; | int added = 0; | ||
while(added < K ) | while(added < K ) | ||
− | added += stack. | + | added += nodeList.get(stack.pop()).search(searchLocation,stack,results); |
while(stack.size() > 0 ){ | while(stack.size() > 0 ){ | ||
− | + | int nodeIndex = stack.pop(); | |
− | if(results.peekPrio() > | + | if(results.peekPrio() > pointRectDist((2*_dimensions)*nodeIndex,searchLocation)) |
− | + | nodeList.get(nodeIndex).search(searchLocation,stack,results); | |
} | } | ||
ArrayList<SearchResult<T>> returnResults = new ArrayList<SearchResult<T>>(K); | ArrayList<SearchResult<T>> returnResults = new ArrayList<SearchResult<T>>(K); | ||
− | + | double[] priorities = results.priorities; | |
+ | Object[] elements = results.elements; | ||
for(int i = 0; i < K; i++){//forward (closest first) | for(int i = 0; i < K; i++){//forward (closest first) | ||
− | SearchResult s = new SearchResult( | + | SearchResult s = new SearchResult(priorities[i],(T)elements[i]); |
returnResults.add(s); | returnResults.add(s); | ||
} | } | ||
return returnResults; | return returnResults; | ||
} | } | ||
− | + | private double pointRectDist(int offset, double[] location){ | |
+ | double distance=0; | ||
+ | double[] array = nodeMinMaxBounds.array; | ||
+ | for(int i = 0; i < location.length; i++,offset += 2){ | ||
+ | |||
+ | double diff = 0; | ||
+ | double bv = array[offset]; | ||
+ | double lv = location[i]; | ||
+ | if(bv > lv) | ||
+ | diff = bv-lv; | ||
+ | else{ | ||
+ | bv=array[offset+1]; | ||
+ | if(lv>bv) | ||
+ | diff = lv-bv; | ||
+ | } | ||
+ | distance += sqr(diff); | ||
+ | } | ||
+ | return distance; | ||
+ | } | ||
//NB! This Priority Queue keeps things with the LOWEST priority. | //NB! This Priority Queue keeps things with the LOWEST priority. | ||
Line 237: | Line 260: | ||
return minPrio; | return minPrio; | ||
} | } | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
} | } | ||
Line 301: | Line 286: | ||
//stem | //stem | ||
− | Node less, more; | + | //Node less, more; |
+ | int lessIndex, moreIndex; | ||
int splitDim; | int splitDim; | ||
double splitVal; | double splitVal; | ||
Line 311: | Line 297: | ||
pointLocations = new ContiguousDoubleArrayList(pointMemory); | pointLocations = new ContiguousDoubleArrayList(pointMemory); | ||
index = _nodes++; | index = _nodes++; | ||
+ | nodeList.add(this); | ||
nodeMinMaxBounds.add(bounds_template); | nodeMinMaxBounds.add(bounds_template); | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
} | } | ||
private final double pointDist(double[] location, int index){ | private final double pointDist(double[] location, int index){ | ||
Line 346: | Line 313: | ||
//returns number of points added to results | //returns number of points added to results | ||
− | private int search(double[] searchLocation, | + | private int search(double[] searchLocation, IntStack stack, PrioQueue<T> results){ |
if(pointLocations == null){ | if(pointLocations == null){ | ||
− | if(searchLocation[splitDim] < splitVal) | + | if(searchLocation[splitDim] < splitVal) |
− | stack. | + | stack.push(moreIndex).push(lessIndex);//less will be popped first |
− | + | else | |
− | + | stack.push(lessIndex).push(moreIndex);//more will be popped first | |
− | else | + | |
− | stack. | ||
− | |||
− | |||
} | } | ||
else{ | else{ | ||
Line 422: | Line 386: | ||
splitVal = nodeMinMaxBounds.array[index*2*_dimensions + 2*splitDim]; | splitVal = nodeMinMaxBounds.array[index*2*_dimensions + 2*splitDim]; | ||
− | less = new Node(mem_recycle);//recycle that memory! | + | Node less = new Node(mem_recycle);//recycle that memory! |
− | more = new Node(); | + | Node more = new Node(); |
+ | lessIndex = less.index; | ||
+ | moreIndex = more.index; | ||
//reduce garbage by factor of _bucketSize by recycling this array | //reduce garbage by factor of _bucketSize by recycling this array | ||
Line 442: | Line 408: | ||
if(less.entries*more.entries == 0){ | if(less.entries*more.entries == 0){ | ||
//one of them was 0, so the split was worthless. throw it away. | //one of them was 0, so the split was worthless. throw it away. | ||
− | |||
− | |||
_nodes -= 2;//recall that bounds memory | _nodes -= 2;//recall that bounds memory | ||
+ | nodeList.remove(moreIndex); | ||
+ | nodeList.remove(lessIndex); | ||
} | } | ||
else{ | else{ | ||
Line 464: | Line 430: | ||
double[] array; | double[] array; | ||
int size; | int size; | ||
− | ContiguousDoubleArrayList(){ | + | ContiguousDoubleArrayList(){this(300);} |
− | + | ContiguousDoubleArrayList(int size){this(new double[size]);} | |
− | + | ContiguousDoubleArrayList(double[] data){array = data;} | |
− | ContiguousDoubleArrayList(int size){ | + | |
− | |||
− | |||
− | ContiguousDoubleArrayList(double[] data){ | ||
− | |||
− | |||
ContiguousDoubleArrayList add(double[] da){ | ContiguousDoubleArrayList add(double[] da){ | ||
− | if(size + da.length > array.length) | + | if(size + da.length > array.length) |
array = Arrays.copyOf(array,(array.length+da.length)*2); | array = Arrays.copyOf(array,(array.length+da.length)*2); | ||
− | + | ||
− | |||
− | |||
System.arraycopy(da,0,array,size,da.length); | System.arraycopy(da,0,array,size,da.length); | ||
size += da.length; | size += da.length; | ||
return this; | return this; | ||
+ | } | ||
+ | } | ||
+ | private static class IntStack{ | ||
+ | int[] array; | ||
+ | int size; | ||
+ | IntStack(){this(64);} | ||
+ | IntStack(int size){this(new int[size]);} | ||
+ | IntStack(int[] data){array = data;} | ||
+ | |||
+ | IntStack push(int i){ | ||
+ | if(size>= array.length) | ||
+ | array = Arrays.copyOf(array,(array.length+1)*2); | ||
+ | |||
+ | array[size++] = i; | ||
+ | return this; | ||
+ | } | ||
+ | int pop(){ | ||
+ | return array[--size]; | ||
+ | } | ||
+ | int size(){ | ||
+ | return size; | ||
} | } | ||
} | } | ||
Line 489: | Line 469: | ||
} | } | ||
− | |||
</syntaxhighlight></code> | </syntaxhighlight></code> |
Revision as of 20:53, 20 July 2013
/*
** KDTree.java by Julian Kent
** Licenced under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License
** See full licencing details here: http://creativecommons.org/licenses/by-nc-sa/3.0/
** For additional licencing rights please contact jkflying@gmail.com
**
** Example usage is given in the main method, as well as benchmarking code against Rednaxela's Gen2 Tree
*/
package jk.mega;
import java.util.ArrayDeque;
import java.util.ArrayList;
import java.util.Arrays;
//import ags.utils.*;
//import ags.utils.dataStructures.*;
public class KDTree<T>{
//use a big bucketSize so that we have less node bounds (for more cache hits) and better splits
private static final int _bucketSize = 50;
private final int _dimensions;
private int _nodes;
private final Node root;
private final ArrayList<Node> nodeList = new ArrayList<Node>();
//prevent GC from having to collect _bucketSize*dimensions*8 bytes each time a leaf splits
private double[] mem_recycle;
//the starting values for bounding boxes, for easy access
private final double[] bounds_template;
//one big self-expanding array to keep all the node bounding boxes so that they stay in cache
// node bounds available at:
//low: 2 * _dimensions * node.index + 2 * dim
//high: 2 * _dimensions * node.index + 2 * dim + 1
private final ContiguousDoubleArrayList nodeMinMaxBounds;
/*
public static void main(String[] args){
int dims = 1;
int size = 2000000;
int testsize = 1;
int k = 40;
int iterations = 1;
System.out.println(
"Config:\n"
+ "No JIT Warmup\n"
+ "Tested on random data.\n"
+ "Training and testing points shared across iterations.\n"
+ "Searches interleaved.");
System.out.println("Num points: " + size);
System.out.println("Num searches: " + testsize);
System.out.println("Dimensions: " + dims);
System.out.println("Num Neighbours: " + k);
System.out.println();
ArrayList<double[]> locs = new ArrayList<double[]>(size);
for(int i = 0; i < size; i++){
double[] loc = new double[dims];
for(int j = 0; j < dims; j++)
loc[j] = Math.random();
locs.add(loc);
}
ArrayList<double[]> testlocs = new ArrayList<double[]>(testsize);
for(int i = 0; i < testsize; i++){
double[] loc = new double[dims];
for(int j = 0; j < dims; j++)
loc[j] = Math.random();
testlocs.add(loc);
}
for(int r = 0; r < iterations; r++){
long t1 = System.nanoTime();
KDTree<double[]> t = new KDTree<double[]>(dims);// This tree
for(int i = 0; i < size; i++){
t.addPoint(locs.get(i),locs.get(i));
}
long t2 = System.nanoTime();
KdTree<double[]> rt = new KdTree.Euclidean<double[]>(dims,null); //Rednaxela Gen2
for(int i = 0; i < size; i++){
rt.addPoint(locs.get(i),locs.get(i));
}
long t3 = System.nanoTime();
long jtn = 0;
long rtn = 0;
long mjtn = 0;
long mrtn = 0;
double dist1 = 0, dist2 = 0;
for(int i = 0; i < testsize; i++){
long t4 = System.nanoTime();
dist1 += t.nearestNeighbours(testlocs.get(i),k).iterator().next().distance;
long t5 = System.nanoTime();
dist2 += rt.nearestNeighbor(testlocs.get(i),k,true).iterator().next().distance;
long t6 = System.nanoTime();
long t7 = System.nanoTime();
jtn += t5 - t4 - (t7 - t6);
rtn += t6 - t5 - (t7 - t6);
mjtn = Math.max(mjtn,t5 - t4 - (t7 - t6));
mrtn = Math.max(mrtn,t6 - t5 - (t7 - t6));
}
System.out.println("Accuracy: " + (Math.abs(dist1-dist2) < 1e-10?"100%":"BROKEN!!!"));
if(Math.abs(dist1-dist2) > 1e-10){
System.out.println("dist1: " + dist1 + " dist2: " + dist2);
}
long jts = t2 - t1;
long rts = t3 - t2;
System.out.println("Iteration: " + (r+1) + "/" + iterations);
System.out.println("This tree add avg: " + jts/size + " ns");
System.out.println("Reds tree add avg: " + rts/size + " ns");
System.out.println("This tree knn avg: " + jtn/testsize + " ns");
System.out.println("Reds tree knn avg: " + rtn/testsize + " ns");
System.out.println("This tree knn max: " + mjtn + " ns");
System.out.println("Reds tree knn max: " + mrtn + " ns");
System.out.println();
}
}
// */
public KDTree(int dimensions){
_dimensions = dimensions;
//initialise this big so that it ends up in 'old' memory
nodeMinMaxBounds = new ContiguousDoubleArrayList(512 * 1024 / 8 + 2*_dimensions);
mem_recycle = new double[_bucketSize*dimensions];
bounds_template = new double[2*_dimensions];
Arrays.fill(bounds_template,Double.NEGATIVE_INFINITY);
for(int i = 0, max = 2*_dimensions; i < max; i+=2)
bounds_template[i] = Double.POSITIVE_INFINITY;
//and.... start!
root = new Node();
}
public int nodes(){
return _nodes;
}
public int addPoint(double[] location, T payload){
Node addNode = root;
//Do a Depth First Search to find the Node where 'location' should be stored
while(addNode.pointLocations == null){
addNode.expandBounds(location);
if(location[addNode.splitDim] < addNode.splitVal)
addNode = nodeList.get(addNode.lessIndex);
else
addNode = nodeList.get(addNode.moreIndex);
}
addNode.expandBounds(location);
int nodeSize = addNode.add(location,payload);
if(nodeSize % _bucketSize == 0)
//try splitting again once every time the node passes a _bucketSize multiple
addNode.split();
return root.entries;
}
public ArrayList<SearchResult<T>> nearestNeighbours(double[] searchLocation, int K){
//don't actually store the objects so that they aren't a possible cache miss
//until we've verified that we need to access them
IntStack stack = new IntStack();
PrioQueue<T> results = new PrioQueue<T>(K,true);
stack.push(root.index);
int added = 0;
while(added < K )
added += nodeList.get(stack.pop()).search(searchLocation,stack,results);
while(stack.size() > 0 ){
int nodeIndex = stack.pop();
if(results.peekPrio() > pointRectDist((2*_dimensions)*nodeIndex,searchLocation))
nodeList.get(nodeIndex).search(searchLocation,stack,results);
}
ArrayList<SearchResult<T>> returnResults = new ArrayList<SearchResult<T>>(K);
double[] priorities = results.priorities;
Object[] elements = results.elements;
for(int i = 0; i < K; i++){//forward (closest first)
SearchResult s = new SearchResult(priorities[i],(T)elements[i]);
returnResults.add(s);
}
return returnResults;
}
private double pointRectDist(int offset, double[] location){
double distance=0;
double[] array = nodeMinMaxBounds.array;
for(int i = 0; i < location.length; i++,offset += 2){
double diff = 0;
double bv = array[offset];
double lv = location[i];
if(bv > lv)
diff = bv-lv;
else{
bv=array[offset+1];
if(lv>bv)
diff = lv-bv;
}
distance += sqr(diff);
}
return distance;
}
//NB! This Priority Queue keeps things with the LOWEST priority.
//If you want highest priority items kept, negate your values
private static class PrioQueue<S>{
Object[] elements;
double[] priorities;
private double minPrio;
private int size;
PrioQueue(int size, boolean prefill){
elements = new Object[size];
priorities = new double[size];
Arrays.fill(priorities,Double.POSITIVE_INFINITY);
if(prefill){
minPrio = Double.POSITIVE_INFINITY;
this.size = size;
}
}
//uses O(log(n)) comparisons and one big shift of size O(N)
//and is MUCH simpler than a heap --> faster on small sets, faster JIT
void addNoGrow(S value, double priority){
int index = searchFor(priority);
int nextIndex = index + 1;
int length = size - index - 1;//remove dependancy on nextIndex
System.arraycopy(elements,index,elements,nextIndex,length);
System.arraycopy(priorities,index,priorities,nextIndex,length);
elements[index]=value;
priorities[index]=priority;
minPrio = priorities[size-1];
}
int searchFor(double priority){
int i = size-1;
int j = 0;
while(i>=j){
int index = (i+j)>>>1;
if( priorities[index] < priority)
j = index+1;
else
i = index-1;
}
return j;
}
double peekPrio(){
return minPrio;
}
}
public static class SearchResult<S>{
public double distance;
public S payload;
SearchResult(double dist, S load){
distance = dist;
payload = load;
}
}
private class Node {
//for accessing bounding box data
// - if trees weren't so unbalanced might be better to use an implicit heap?
int index;
//keep track of size of subtree
int entries;
//leaf
ContiguousDoubleArrayList pointLocations ;
ArrayList<T> pointPayloads = new ArrayList<T>(_bucketSize);
//stem
//Node less, more;
int lessIndex, moreIndex;
int splitDim;
double splitVal;
private Node(){
this(new double[_bucketSize*_dimensions]);
}
private Node(double[] pointMemory){
pointLocations = new ContiguousDoubleArrayList(pointMemory);
index = _nodes++;
nodeList.add(this);
nodeMinMaxBounds.add(bounds_template);
}
private final double pointDist(double[] location, int index){
double[] arr = pointLocations.array;
double distance = 0;
int offset = (index+1)*_dimensions;
for(int i = _dimensions; i-- > 0 ;){
double d;
distance += (d = arr[--offset] - location[i])*d;
}
return distance;
}
//returns number of points added to results
private int search(double[] searchLocation, IntStack stack, PrioQueue<T> results){
if(pointLocations == null){
if(searchLocation[splitDim] < splitVal)
stack.push(moreIndex).push(lessIndex);//less will be popped first
else
stack.push(lessIndex).push(moreIndex);//more will be popped first
}
else{
int updated = 0;
for(int j = entries; j-- > 0;){
double distance = pointDist(searchLocation,j);
if(results.peekPrio() > distance){
updated++;
results.addNoGrow(pointPayloads.get(j),distance);
}
}
return updated;
}
return 0;
}
private void expandBounds(double[] location){
entries++;
int mio = index*2*_dimensions;
for(int i = 0; i < _dimensions;i++){
nodeMinMaxBounds.array[mio] = Math.min(nodeMinMaxBounds.array[mio++],location[i]);
nodeMinMaxBounds.array[mio] = Math.max(nodeMinMaxBounds.array[mio++],location[i]);
}
}
private int add(double[] location, T load){
pointLocations.add(location);
pointPayloads.add(load);
return entries;
}
private void split(){
int offset = index*2*_dimensions;
double diff = 0;
for(int i = 0; i < _dimensions; i++){
double min = nodeMinMaxBounds.array[offset];
double max = nodeMinMaxBounds.array[offset+1];
if(max-min>diff){
double mean = 0;
for(int j = 0; j < entries; j++)
mean += pointLocations.array[i+_dimensions*j];
mean = mean/entries;
double varianceSum = 0;
for(int j = 0; j < entries; j++)
varianceSum += sqr(mean-pointLocations.array[i+_dimensions*j]);
if(varianceSum>diff*entries){
diff = varianceSum/entries;
splitVal = mean;
splitDim = i;
}
}
offset += 2;
}
//kill all the nasties
if(splitVal == Double.POSITIVE_INFINITY)
splitVal = Double.MAX_VALUE;
else if(splitVal == Double.NEGATIVE_INFINITY)
splitVal = Double.MIN_VALUE;
else if(splitVal == nodeMinMaxBounds.array[index*2*_dimensions + 2*splitDim + 1])
splitVal = nodeMinMaxBounds.array[index*2*_dimensions + 2*splitDim];
Node less = new Node(mem_recycle);//recycle that memory!
Node more = new Node();
lessIndex = less.index;
moreIndex = more.index;
//reduce garbage by factor of _bucketSize by recycling this array
double[] pointLocation = new double[_dimensions];
for(int i = 0; i < entries; i++){
System.arraycopy(pointLocations.array,i*_dimensions,pointLocation,0,_dimensions);
T load = pointPayloads.get(i);
if(pointLocation[splitDim] < splitVal){
less.expandBounds(pointLocation);
less.add(pointLocation,load);
}
else{
more.expandBounds(pointLocation);
more.add(pointLocation,load);
}
}
if(less.entries*more.entries == 0){
//one of them was 0, so the split was worthless. throw it away.
_nodes -= 2;//recall that bounds memory
nodeList.remove(moreIndex);
nodeList.remove(lessIndex);
}
else{
//we won't be needing that now, so keep it for the next split to reduce garbage
mem_recycle = pointLocations.array;
pointLocations = null;
pointPayloads.clear();
pointPayloads = null;
}
}
}
private static class ContiguousDoubleArrayList{
double[] array;
int size;
ContiguousDoubleArrayList(){this(300);}
ContiguousDoubleArrayList(int size){this(new double[size]);}
ContiguousDoubleArrayList(double[] data){array = data;}
ContiguousDoubleArrayList add(double[] da){
if(size + da.length > array.length)
array = Arrays.copyOf(array,(array.length+da.length)*2);
System.arraycopy(da,0,array,size,da.length);
size += da.length;
return this;
}
}
private static class IntStack{
int[] array;
int size;
IntStack(){this(64);}
IntStack(int size){this(new int[size]);}
IntStack(int[] data){array = data;}
IntStack push(int i){
if(size>= array.length)
array = Arrays.copyOf(array,(array.length+1)*2);
array[size++] = i;
return this;
}
int pop(){
return array[--size];
}
int size(){
return size;
}
}
private static final double sqr(double d){
return d*d;}
}