songsenand
/
cgal
mirror of https://github.com/CGAL/cgal
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

*** empty log message ***

This commit is contained in:
Hans Tangelder 2002-08-29 09:57:46 +00:00
parent 8e0b01f88a
commit 637f44f617
5 changed files with 65 additions and 38 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
Packages/Spatial_searching/examples/Spatial_searching/Example1.C
View File

@ -1,5 +1,6 @@
// example using nearest_neighbour_iterator for L2
// benchmark example using 10000 data points and 2000 query points
// bucketsize=1
// both generated with Random_points_in_cube_3<Point_3>
// comparing ASPAS to brute force method
@ -33,7 +34,8 @@ typedef K::FT NT;
typedef CGAL::Plane_separator<NT> Separator;
typedef CGAL::Kd_tree_traits_point<Separator,Point> Traits;
typedef CGAL::Nearest_neighbour_L2<Traits,CGAL::Search_nearest_neighbour>::iterator NNN_Iterator;
typedef CGAL::Nearest_neighbour_L2<Traits,
CGAL::Search_nearest_neighbour>::iterator NNN_Iterator;
@ -53,8 +55,10 @@ NT The_squared_distance(Point P, Point Q) {
int bucket_size=1;
NT eps=0.0;
std::cout << "test parameters: d=" << dim << " point_number=" << point_number << std::endl;
std::cout << "query_point_number=" << query_point_number << " bucket_size="
std::cout << "test parameters: d=" << dim << " point_number="
<< point_number << std::endl;
std::cout << "query_point_number=" << query_point_number
<< " bucket_size="
<< bucket_size << " eps=" << eps << std::endl;
// generate 10000 data points
@ -88,13 +92,15 @@ NT The_squared_distance(Point P, Point Q) {
t.stop();
std::cout << "created binary search tree containing" << std::endl
<< point_number << " random points in the 3-dim unit cube in time " << t.time() <<
<< point_number << " random points in the 3-dim unit cube in time "
<< t.time() <<
" seconds " << std::endl;
d.statistics();
// end of building binary search tree
std::vector<Traits::Item_with_distance> nearest_neighbours(query_point_number);
std::vector<Traits::Item_with_distance>
nearest_neighbours(query_point_number);
t.reset(); t.start();
for (int i=0; i < query_point_number; i++) {
@ -114,7 +120,8 @@ NT The_squared_distance(Point P, Point Q) {
// copy data points from vector to list
Vector the_data_points;
the_data_points.reserve(point_number);
std::copy(data_points.begin(),data_points.end(),std::back_inserter(the_data_points));
std::copy(data_points.begin(),data_points.end(),
std::back_inserter(the_data_points));
std::vector<int>
nearest_neighbours_brute_force_index(query_point_number);
@ -146,7 +153,8 @@ NT The_squared_distance(Point P, Point Q) {
if (!(*(nearest_neighbours[i].first)==the_data_points[
nearest_neighbours_brute_force_index[i]])) {
assert(
The_squared_distance(query_points[i],*(nearest_neighbours[i]).first)==
The_squared_distance(query_points[i],
*(nearest_neighbours[i]).first)==
The_squared_distance(query_points[i],
the_data_points[nearest_neighbours_brute_force_index[i]]));
};

25
Packages/Spatial_searching/examples/Spatial_searching/Example2.C
View File

@ -1,5 +1,6 @@
// example using nearest_neighbour_iterator for Linf
// benchmark example using 10000 data points and 2000 query points
// benchmark example using 10000 data points and 2000 query points,
// bucketsize=1
// both generated with Random_points_in_cube_3<Point_3>
// comparing to brute force method
@ -34,7 +35,8 @@ typedef K::FT NT;
typedef CGAL::Plane_separator<NT> Separator;
typedef CGAL::Kd_tree_traits_point<Separator,Point> Traits;
typedef CGAL::Nearest_neighbour_Linf<Traits,CGAL::Search_nearest_neighbour>::iterator NNN_Iterator;
typedef CGAL::Nearest_neighbour_Linf<Traits,
CGAL::Search_nearest_neighbour>::iterator NNN_Iterator;
NT The_Linf_distance(Point P, Point Q) {
NT dist=fabs(P.x()-Q.x());
@ -54,8 +56,10 @@ NT The_Linf_distance(Point P, Point Q) {
int bucket_size=1;
NT eps=0.0;
std::cout << "test parameters: d=" << dim << " point_number=" << point_number << std::endl;
std::cout << "query_point_number=" << query_point_number << " bucket_size="
std::cout << "test parameters: d=" << dim << " point_number="
<< point_number << std::endl;
std::cout << "query_point_number=" << query_point_number
<< " bucket_size="
<< bucket_size << " eps=" << eps << std::endl;
// generate 10000 data points
@ -89,13 +93,14 @@ NT The_Linf_distance(Point P, Point Q) {
t.stop();
std::cout << "created binary search tree containing" << std::endl
<< point_number << " random points in the 3-dim unit cube in time " << t.time() <<
" seconds " << std::endl;
<< point_number << " random points in the 3-dim unit cube in time "
<< t.time() << " seconds " << std::endl;
d.statistics();
// end of building binary search tree
std::vector<Traits::Item_with_distance> nearest_neighbours(query_point_number);
std::vector<Traits::Item_with_distance>
nearest_neighbours(query_point_number);
t.reset(); t.start();
for (int i=0; i < query_point_number; i++) {
@ -113,7 +118,8 @@ NT The_Linf_distance(Point P, Point Q) {
// copy data points from vector to list
Vector the_data_points;
the_data_points.reserve(point_number);
std::copy(data_points.begin(),data_points.end(),std::back_inserter(the_data_points));
std::copy(data_points.begin(),data_points.end(),
std::back_inserter(the_data_points));
std::vector<int>
nearest_neighbours_brute_force_index(query_point_number);
@ -145,7 +151,8 @@ NT The_Linf_distance(Point P, Point Q) {
if (!(*(nearest_neighbours[i].first)==the_data_points[
nearest_neighbours_brute_force_index[i]])) {
assert(
The_Linf_distance(query_points[i],*(nearest_neighbours[i]).first)==
The_Linf_distance(query_points[i],
*(nearest_neighbours[i]).first)==
The_Linf_distance(query_points[i],
the_data_points[nearest_neighbours_brute_force_index[i]]));
};

26
Packages/Spatial_searching/examples/Spatial_searching/Example3.C
View File

@ -1,5 +1,6 @@
// example using nearest_neighbour_iterator for Minkowski_norm
// benchmark example using 10000 data points and 2000 query points, bucketsize=10
// benchmark example using 10000 data points and 2000 query points,
// bucketsize=10
// both generated with Random_points_in_cube_3<Point_3>
#include <CGAL/basic.h>
@ -48,9 +49,10 @@ NNN_Iterator;
std::vector<double> my_weights(dim);
my_weights[0]=0.0; my_weights[1]=2.0; my_weights[2]=5;
std::cout << "test parameters: d=" << dim << " point_number=" << point_number << std::endl;
std::cout << "query_point_number=" << query_point_number << " bucket_size="
<< bucket_size << " eps=" << eps << std::endl;
std::cout << "test parameters: d=" << dim << " point_number="
<< point_number << std::endl;
std::cout << "query_point_number=" << query_point_number
<< " bucket_size=" << bucket_size << " eps=" << eps << std::endl;
// generate 10000 data points
// Prepare a vector for 10000 points.
@ -73,7 +75,8 @@ NNN_Iterator;
query_points.reserve(2000);
// Create 2000 query points within the same cube.
CGAL::copy_n( g, query_point_number, std::back_inserter(query_points));
CGAL::copy_n( g, query_point_number,
std::back_inserter(query_points));
t.reset(); t.start();
@ -85,13 +88,14 @@ NNN_Iterator;
t.stop();
std::cout << "created binary search tree containing" << std::endl
<< point_number << " random points in the 3-dim unit cube in time " << t.time() <<
" seconds " << std::endl;
<< point_number << " random points in the 3-dim unit cube in time "
<< t.time() << " seconds " << std::endl;
d.statistics();
// end of building binary search tree
std::vector<Traits::Item_with_distance> nearest_neighbours(query_point_number);
std::vector<Traits::Item_with_distance>
nearest_neighbours(query_point_number);
Distance_traits tr_dist(the_power,dim,my_weights);
t.reset(); t.start();
@ -110,7 +114,8 @@ NNN_Iterator;
// copy data points from vector to list
Vector the_data_points;
the_data_points.reserve(point_number);
std::copy(data_points.begin(),data_points.end(),std::back_inserter(the_data_points));
std::copy(data_points.begin(),data_points.end(),
std::back_inserter(the_data_points));
std::vector<int>
nearest_neighbours_brute_force_index(query_point_number);
@ -142,7 +147,8 @@ NNN_Iterator;
if (!(*(nearest_neighbours[i].first)==the_data_points[
nearest_neighbours_brute_force_index[i]])) {
assert(
tr_dist.distance(query_points[i],*(nearest_neighbours[i]).first)==
tr_dist.distance(query_points[i],
*(nearest_neighbours[i]).first)==
tr_dist.distance(query_points[i],
the_data_points[nearest_neighbours_brute_force_index[i]]));
};

22
Packages/Spatial_searching/examples/Spatial_searching/Example4.C
View File

@ -1,5 +1,7 @@
// example using nearest_neighbour_iterator for Minkowski_norm with general_distance
// benchmark example using 10000 data points and 2000 query points, bucketsize=10
// example using nearest_neighbour_iterator for Minkowski_norm
// with general_distance
// benchmark example using 10000 data points and 2000 query points,
// bucketsize=10
// both generated with Random_points_in_cube_3<Point_3>
#include <CGAL/basic.h>
@ -48,7 +50,8 @@ NNN_Iterator;
std::vector<double> my_weights(dim);
my_weights[0]=0.0; my_weights[1]=2.0; my_weights[2]=5;
std::cout << "test parameters: d=" << dim << " point_number=" << point_number << std::endl;
std::cout << "test parameters: d=" << dim << " point_number="
<< point_number << std::endl;
std::cout << "query_point_number=" << query_point_number << " bucket_size="
<< bucket_size << " eps=" << eps << std::endl;
@ -85,13 +88,14 @@ NNN_Iterator;
t.stop();
std::cout << "created binary search tree containing" << std::endl
<< point_number << " random points in the 3-dim unit cube in time " << t.time() <<
" seconds " << std::endl;
<< point_number << " random points in the 3-dim unit cube in time "
<< t.time() << " seconds " << std::endl;
d.statistics();
// end of building binary search tree
std::vector<Traits::Item_with_distance> nearest_neighbours(query_point_number);
std::vector<Traits::Item_with_distance>
nearest_neighbours(query_point_number);
Distance_traits tr_dist(the_power,dim,my_weights);
t.reset(); t.start();
@ -110,7 +114,8 @@ NNN_Iterator;
// copy data points from vector to list
Vector the_data_points;
the_data_points.reserve(point_number);
std::copy(data_points.begin(),data_points.end(),std::back_inserter(the_data_points));
std::copy(data_points.begin(),data_points.end(),
std::back_inserter(the_data_points));
std::vector<int>
nearest_neighbours_brute_force_index(query_point_number);
@ -142,7 +147,8 @@ NNN_Iterator;
if (!(*(nearest_neighbours[i].first)==the_data_points[
nearest_neighbours_brute_force_index[i]])) {
assert(
tr_dist.distance(query_points[i],*(nearest_neighbours[i]).first)==
tr_dist.distance(query_points[i],
*(nearest_neighbours[i]).first)==
tr_dist.distance(query_points[i],
the_data_points[nearest_neighbours_brute_force_index[i]]));
};

8
Packages/Spatial_searching/include/CGAL/Nearest_neighbour_Minkowski_norm.h
View File

@ -58,7 +58,7 @@ class Priority_higher
search_nearest = s.Search_nearest();
}
//highest priority is smallest distance
bool operator() (Node_with_distance* n1, Node_with_distance* n2) const
bool operator() (Node_with_distance* n1, Node_with_distance* n2) const
{
if (search_nearest) { return (n1->second > n2->second);}
else {return (n2->second > n1->second);}
@ -78,7 +78,7 @@ class Distance_smaller
}
//highest priority is smallest distance
bool operator() (Item_with_distance* p1, Item_with_distance* p2) const
bool operator() (Item_with_distance* p1, Item_with_distance* p2) const
{
if (search_nearest) {return (p1->second > p2->second);}
else {return (p2->second > p1->second);}
@ -97,8 +97,8 @@ class Distance_smaller
public:
// constructor
Nearest_neighbour_Minkowski_norm(Tree& tree, Item& q, Orthogonal_Distance& tr,
NT Eps=0.0)
Nearest_neighbour_Minkowski_norm(Tree& tree, Item& q,
Orthogonal_Distance& tr, NT Eps=0.0)
{
start = new iterator(tree,q,tr,Eps);
past_the_end = new iterator();