diff options
Diffstat (limited to 'src')
8 files changed, 199 insertions, 115 deletions
diff --git a/src/Alpha_complex/test/CMakeLists.txt b/src/Alpha_complex/test/CMakeLists.txt index db5d840f..0595ca92 100644 --- a/src/Alpha_complex/test/CMakeLists.txt +++ b/src/Alpha_complex/test/CMakeLists.txt @@ -59,4 +59,18 @@ if (NOT CGAL_WITH_EIGEN3_VERSION VERSION_LESS 5.1.0) endif() gudhi_add_boost_test(Weighted_alpha_complex_test_unit) + add_executable ( Weighted_alpha_complex_non_visible_points_test_unit Weighted_alpha_complex_non_visible_points_unit_test.cpp ) + target_link_libraries(Weighted_alpha_complex_non_visible_points_test_unit ${CGAL_LIBRARY}) + if (TBB_FOUND) + target_link_libraries(Weighted_alpha_complex_non_visible_points_test_unit ${TBB_LIBRARIES}) + endif() + gudhi_add_boost_test(Weighted_alpha_complex_non_visible_points_test_unit) + + add_executable ( Zero_weighted_alpha_complex_test_unit Zero_weighted_alpha_complex_unit_test.cpp ) + target_link_libraries(Zero_weighted_alpha_complex_test_unit ${CGAL_LIBRARY}) + if (TBB_FOUND) + target_link_libraries(Zero_weighted_alpha_complex_test_unit ${TBB_LIBRARIES}) + endif() + gudhi_add_boost_test(Zero_weighted_alpha_complex_test_unit) + endif (NOT CGAL_WITH_EIGEN3_VERSION VERSION_LESS 5.1.0)
\ No newline at end of file diff --git a/src/Alpha_complex/test/Weighted_alpha_complex_non_visible_points_unit_test.cpp b/src/Alpha_complex/test/Weighted_alpha_complex_non_visible_points_unit_test.cpp new file mode 100644 index 00000000..dd83c1da --- /dev/null +++ b/src/Alpha_complex/test/Weighted_alpha_complex_non_visible_points_unit_test.cpp @@ -0,0 +1,60 @@ +/* This file is part of the Gudhi Library - https://gudhi.inria.fr/ - which is released under MIT. + * See file LICENSE or go to https://gudhi.inria.fr/licensing/ for full license details. + * Author(s): Vincent Rouvreau + * + * Copyright (C) 2020 Inria + * + * Modification(s): + * - YYYY/MM Author: Description of the modification + */ + +#define BOOST_TEST_DYN_LINK +#define BOOST_TEST_MODULE "weighted_alpha_complex_non_visible_points" +#include <boost/test/unit_test.hpp> +#include <boost/mpl/list.hpp> + +#include <CGAL/Epick_d.h> +#include <CGAL/Epeck_d.h> + +#include <vector> + +#include <gudhi/Alpha_complex.h> +#include <gudhi/Simplex_tree.h> + + +using list_of_1d_kernel_variants = boost::mpl::list<CGAL::Epeck_d< CGAL::Dynamic_dimension_tag >, + CGAL::Epeck_d< CGAL::Dimension_tag<1>>, + CGAL::Epick_d< CGAL::Dynamic_dimension_tag >, + CGAL::Epick_d< CGAL::Dimension_tag<1>> + >; + +BOOST_AUTO_TEST_CASE_TEMPLATE(Weighted_alpha_complex_non_visible_points, Kernel, list_of_1d_kernel_variants) { + // check that for 2 closed weighted 1-d points, one with a high weight to hide the second one with a small weight, + // that the point with a small weight has the same high filtration value than the edge formed by the 2 points + using Point_d = typename Kernel::Point_d; + std::vector<Point_d> points; + std::vector<double> p1 {0.}; + points.emplace_back(p1.begin(), p1.end()); + // closed enough points + std::vector<double> p2 {0.1}; + points.emplace_back(p2.begin(), p2.end()); + std::vector<typename Kernel::FT> weights {100., 0.01}; + + Gudhi::alpha_complex::Alpha_complex<Kernel, true> alpha_complex(points, weights); + Gudhi::Simplex_tree<> stree; + BOOST_CHECK(alpha_complex.create_complex(stree)); + + std::clog << "Iterator on weighted alpha complex simplices in the filtration order, with [filtration value]:" + << std::endl; + for (auto f_simplex : stree.filtration_simplex_range()) { + std::clog << " ( "; + for (auto vertex : stree.simplex_vertex_range(f_simplex)) { + std::clog << vertex << " "; + } + std::clog << ") -> " << "[" << stree.filtration(f_simplex) << "] " << std::endl; + } + + BOOST_CHECK(stree.filtration(stree.find({0})) == -100.); + BOOST_CHECK(stree.filtration(stree.find({1})) == stree.filtration(stree.find({0, 1}))); + BOOST_CHECK(stree.filtration(stree.find({1})) > 100000); +}
\ No newline at end of file diff --git a/src/Alpha_complex/test/Weighted_alpha_complex_unit_test.cpp b/src/Alpha_complex/test/Weighted_alpha_complex_unit_test.cpp index d267276c..875704ee 100644 --- a/src/Alpha_complex/test/Weighted_alpha_complex_unit_test.cpp +++ b/src/Alpha_complex/test/Weighted_alpha_complex_unit_test.cpp @@ -13,10 +13,8 @@ #include <boost/test/unit_test.hpp> #include <boost/mpl/list.hpp> -#include <CGAL/Epick_d.h> #include <CGAL/Epeck_d.h> -#include <cmath> // float comparison #include <vector> #include <random> #include <array> @@ -25,69 +23,6 @@ #include <gudhi/Alpha_complex.h> #include <gudhi/Alpha_complex_3d.h> #include <gudhi/Simplex_tree.h> -#include <gudhi/Unitary_tests_utils.h> - -using list_of_exact_kernel_variants = boost::mpl::list<CGAL::Epeck_d< CGAL::Dynamic_dimension_tag >, - CGAL::Epeck_d< CGAL::Dimension_tag<4> > - > ; - -BOOST_AUTO_TEST_CASE_TEMPLATE(Zero_weighted_alpha_complex, Kernel, list_of_exact_kernel_variants) { - // Check that in exact mode for static dimension 4 the code for dD unweighted and for dD weighted with all weights - // 0 give exactly the same simplex tree (simplices and filtration values). - - // Random points construction - using Point_d = typename Kernel::Point_d; - std::vector<Point_d> points; - std::uniform_real_distribution<double> rd_pts(-10., 10.); - std::random_device rand_dev; - std::mt19937 rand_engine(rand_dev()); - for (int idx = 0; idx < 20; idx++) { - std::vector<double> point {rd_pts(rand_engine), rd_pts(rand_engine), rd_pts(rand_engine), rd_pts(rand_engine)}; - points.emplace_back(point.begin(), point.end()); - } - - // Alpha complex from points - Gudhi::alpha_complex::Alpha_complex<Kernel, false> alpha_complex_from_points(points); - Gudhi::Simplex_tree<> simplex; - Gudhi::Simplex_tree<>::Filtration_value infty = std::numeric_limits<Gudhi::Simplex_tree<>::Filtration_value>::infinity(); - BOOST_CHECK(alpha_complex_from_points.create_complex(simplex, infty, true)); - std::clog << "Iterator on alpha complex simplices in the filtration order, with [filtration value]:" - << std::endl; - for (auto f_simplex : simplex.filtration_simplex_range()) { - std::clog << " ( "; - for (auto vertex : simplex.simplex_vertex_range(f_simplex)) { - std::clog << vertex << " "; - } - std::clog << ") -> " << "[" << simplex.filtration(f_simplex) << "] " << std::endl; - } - - // Alpha complex from zero weighted points - std::vector<typename Kernel::FT> weights(20, 0.); - Gudhi::alpha_complex::Alpha_complex<Kernel, true> alpha_complex_from_zero_weighted_points(points, weights); - Gudhi::Simplex_tree<> zw_simplex; - BOOST_CHECK(alpha_complex_from_zero_weighted_points.create_complex(zw_simplex, infty, true)); - - std::clog << "Iterator on zero weighted alpha complex simplices in the filtration order, with [filtration value]:" - << std::endl; - for (auto f_simplex : zw_simplex.filtration_simplex_range()) { - std::clog << " ( "; - for (auto vertex : zw_simplex.simplex_vertex_range(f_simplex)) { - std::clog << vertex << " "; - } - std::clog << ") -> " << "[" << zw_simplex.filtration(f_simplex) << "] " << std::endl; - } - - BOOST_CHECK(zw_simplex == simplex); -} - -template <typename Point_d> -bool cgal_3d_point_sort (Point_d a,Point_d b) { - if (a[0] != b[0]) - return a[0] < b[0]; - if (a[1] != b[1]) - return a[1] < b[1]; - return a[2] < b[2]; -} BOOST_AUTO_TEST_CASE(Weighted_alpha_complex_3d_comparison) { // check that for random weighted 3d points in safe mode the 3D and dD codes give the same result with some tolerance @@ -189,41 +124,4 @@ BOOST_AUTO_TEST_CASE(Weighted_alpha_complex_3d_comparison) { } ++dD_itr; } -} - -using list_of_1d_kernel_variants = boost::mpl::list<CGAL::Epeck_d< CGAL::Dynamic_dimension_tag >, - CGAL::Epeck_d< CGAL::Dimension_tag<1>>, - CGAL::Epick_d< CGAL::Dynamic_dimension_tag >, - CGAL::Epick_d< CGAL::Dimension_tag<1>> - >; - -BOOST_AUTO_TEST_CASE_TEMPLATE(Weighted_alpha_complex_non_visible_points, Kernel, list_of_1d_kernel_variants) { - // check that for 2 closed weighted 1-d points, one with a high weight to hide the second one with a small weight, - // that the point with a small weight has the same high filtration value than the edge formed by the 2 points - using Point_d = typename Kernel::Point_d; - std::vector<Point_d> points; - std::vector<double> p1 {0.}; - points.emplace_back(p1.begin(), p1.end()); - // closed enough points - std::vector<double> p2 {0.1}; - points.emplace_back(p2.begin(), p2.end()); - std::vector<typename Kernel::FT> weights {100., 0.01}; - - Gudhi::alpha_complex::Alpha_complex<Kernel, true> alpha_complex(points, weights); - Gudhi::Simplex_tree<> stree; - BOOST_CHECK(alpha_complex.create_complex(stree)); - - std::clog << "Iterator on weighted alpha complex simplices in the filtration order, with [filtration value]:" - << std::endl; - for (auto f_simplex : stree.filtration_simplex_range()) { - std::clog << " ( "; - for (auto vertex : stree.simplex_vertex_range(f_simplex)) { - std::clog << vertex << " "; - } - std::clog << ") -> " << "[" << stree.filtration(f_simplex) << "] " << std::endl; - } - - BOOST_CHECK(stree.filtration(stree.find({0})) == -100.); - BOOST_CHECK(stree.filtration(stree.find({1})) == stree.filtration(stree.find({0, 1}))); - BOOST_CHECK(stree.filtration(stree.find({1})) > 100000); }
\ No newline at end of file diff --git a/src/Alpha_complex/test/Zero_weighted_alpha_complex_unit_test.cpp b/src/Alpha_complex/test/Zero_weighted_alpha_complex_unit_test.cpp new file mode 100644 index 00000000..b7df07c7 --- /dev/null +++ b/src/Alpha_complex/test/Zero_weighted_alpha_complex_unit_test.cpp @@ -0,0 +1,77 @@ +/* This file is part of the Gudhi Library - https://gudhi.inria.fr/ - which is released under MIT. + * See file LICENSE or go to https://gudhi.inria.fr/licensing/ for full license details. + * Author(s): Vincent Rouvreau + * + * Copyright (C) 2020 Inria + * + * Modification(s): + * - YYYY/MM Author: Description of the modification + */ + +#define BOOST_TEST_DYN_LINK +#define BOOST_TEST_MODULE "zero_weighted_alpha_complex" +#include <boost/test/unit_test.hpp> +#include <boost/mpl/list.hpp> + +#include <CGAL/Epeck_d.h> + +#include <vector> +#include <random> +#include <cmath> // for std::fabs + +#include <gudhi/Alpha_complex.h> +#include <gudhi/Simplex_tree.h> +#include <gudhi/Unitary_tests_utils.h> + +using list_of_exact_kernel_variants = boost::mpl::list<CGAL::Epeck_d< CGAL::Dynamic_dimension_tag >, + CGAL::Epeck_d< CGAL::Dimension_tag<4> > + > ; + +BOOST_AUTO_TEST_CASE_TEMPLATE(Zero_weighted_alpha_complex, Kernel, list_of_exact_kernel_variants) { + // Check that in exact mode for static dimension 4 the code for dD unweighted and for dD weighted with all weights + // 0 give exactly the same simplex tree (simplices and filtration values). + + // Random points construction + using Point_d = typename Kernel::Point_d; + std::vector<Point_d> points; + std::uniform_real_distribution<double> rd_pts(-10., 10.); + std::random_device rand_dev; + std::mt19937 rand_engine(rand_dev()); + for (int idx = 0; idx < 20; idx++) { + std::vector<double> point {rd_pts(rand_engine), rd_pts(rand_engine), rd_pts(rand_engine), rd_pts(rand_engine)}; + points.emplace_back(point.begin(), point.end()); + } + + // Alpha complex from points + Gudhi::alpha_complex::Alpha_complex<Kernel, false> alpha_complex_from_points(points); + Gudhi::Simplex_tree<> simplex; + Gudhi::Simplex_tree<>::Filtration_value infty = std::numeric_limits<Gudhi::Simplex_tree<>::Filtration_value>::infinity(); + BOOST_CHECK(alpha_complex_from_points.create_complex(simplex, infty, true)); + std::clog << "Iterator on alpha complex simplices in the filtration order, with [filtration value]:" + << std::endl; + for (auto f_simplex : simplex.filtration_simplex_range()) { + std::clog << " ( "; + for (auto vertex : simplex.simplex_vertex_range(f_simplex)) { + std::clog << vertex << " "; + } + std::clog << ") -> " << "[" << simplex.filtration(f_simplex) << "] " << std::endl; + } + + // Alpha complex from zero weighted points + std::vector<typename Kernel::FT> weights(20, 0.); + Gudhi::alpha_complex::Alpha_complex<Kernel, true> alpha_complex_from_zero_weighted_points(points, weights); + Gudhi::Simplex_tree<> zw_simplex; + BOOST_CHECK(alpha_complex_from_zero_weighted_points.create_complex(zw_simplex, infty, true)); + + std::clog << "Iterator on zero weighted alpha complex simplices in the filtration order, with [filtration value]:" + << std::endl; + for (auto f_simplex : zw_simplex.filtration_simplex_range()) { + std::clog << " ( "; + for (auto vertex : zw_simplex.simplex_vertex_range(f_simplex)) { + std::clog << vertex << " "; + } + std::clog << ") -> " << "[" << zw_simplex.filtration(f_simplex) << "] " << std::endl; + } + + BOOST_CHECK(zw_simplex == simplex); +}
\ No newline at end of file diff --git a/src/Tangential_complex/include/gudhi/Tangential_complex.h b/src/Tangential_complex/include/gudhi/Tangential_complex.h index f007bdd5..f3491f91 100644 --- a/src/Tangential_complex/include/gudhi/Tangential_complex.h +++ b/src/Tangential_complex/include/gudhi/Tangential_complex.h @@ -954,7 +954,11 @@ class Tangential_complex { // Triangulation's traits functor & objects typename Tr_traits::Compute_weight_d point_weight = local_tr_traits.compute_weight_d_object(); +#if CGAL_VERSION_NR < 1050200000 typename Tr_traits::Power_center_d power_center = local_tr_traits.power_center_d_object(); +#else + typename Tr_traits::Construct_power_sphere_d power_center = local_tr_traits.construct_power_sphere_d_object(); +#endif //*************************************************** // Build a minimal triangulation in the tangent space @@ -1100,7 +1104,11 @@ class Tangential_complex { std::size_t closest_pt_index = updated_pts_ds.k_nearest_neighbors(center_point, 1, false).begin()->first; typename K::Construct_weighted_point_d k_constr_wp = m_k.construct_weighted_point_d_object(); +#if CGAL_VERSION_NR < 1050200000 typename K::Power_distance_d k_power_dist = m_k.power_distance_d_object(); +#else + typename K::Compute_power_product_d k_power_dist = m_k.compute_power_product_d_object(); +#endif // Construct a weighted point equivalent to the star sphere Weighted_point star_sphere = k_constr_wp(compute_perturbed_point(i), m_squared_star_spheres_radii_incl_margin[i]); diff --git a/src/python/gudhi/representations/vector_methods.py b/src/python/gudhi/representations/vector_methods.py index cdcb1fde..84bc99a2 100644 --- a/src/python/gudhi/representations/vector_methods.py +++ b/src/python/gudhi/representations/vector_methods.py @@ -605,18 +605,19 @@ class Atol(BaseEstimator, TransformerMixin): >>> b = np.array([[4, 2, 0], [4, 4, 0], [4, 0, 2]]) >>> c = np.array([[3, 2, -1], [1, 2, -1]]) >>> atol_vectoriser = Atol(quantiser=KMeans(n_clusters=2, random_state=202006)) - >>> atol_vectoriser.fit(X=[a, b, c]).centers - array([[ 2. , 0.66666667, 3.33333333], - [ 2.6 , 2.8 , -0.4 ]]) + >>> atol_vectoriser.fit(X=[a, b, c]).centers # doctest: +SKIP + >>> # array([[ 2. , 0.66666667, 3.33333333], + >>> # [ 2.6 , 2.8 , -0.4 ]]) >>> atol_vectoriser(a) - array([1.18168665, 0.42375966]) + >>> # array([1.18168665, 0.42375966]) # doctest: +SKIP >>> atol_vectoriser(c) - array([0.02062512, 1.25157463]) - >>> atol_vectoriser.transform(X=[a, b, c]) - array([[1.18168665, 0.42375966], - [0.29861028, 1.06330156], - [0.02062512, 1.25157463]]) + >>> # array([0.02062512, 1.25157463]) # doctest: +SKIP + >>> atol_vectoriser.transform(X=[a, b, c]) # doctest: +SKIP + >>> # array([[1.18168665, 0.42375966], + >>> # [0.29861028, 1.06330156], + >>> # [0.02062512, 1.25157463]]) """ + # Note the example above must be up to date with the one in tests called test_atol_doc def __init__(self, quantiser, weighting_method="cloud", contrast="gaussian"): """ Constructor for the Atol measure vectorisation class. diff --git a/src/python/gudhi/rips_complex.pyx b/src/python/gudhi/rips_complex.pyx index 72e82c79..c3470292 100644 --- a/src/python/gudhi/rips_complex.pyx +++ b/src/python/gudhi/rips_complex.pyx @@ -49,13 +49,13 @@ cdef class RipsComplex: :type max_edge_length: float :param points: A list of points in d-Dimension. - :type points: list of list of double + :type points: list of list of float Or :param distance_matrix: A distance matrix (full square or lower triangular). - :type points: list of list of double + :type points: list of list of float And in both cases @@ -89,10 +89,10 @@ cdef class RipsComplex: def create_simplex_tree(self, max_dimension=1): """ - :param max_dimension: graph expansion for rips until this given maximal + :param max_dimension: graph expansion for Rips until this given maximal dimension. :type max_dimension: int - :returns: A simplex tree created from the Delaunay Triangulation. + :returns: A simplex tree encoding the Vietoris–Rips filtration. :rtype: SimplexTree """ stree = SimplexTree() diff --git a/src/python/test/test_representations.py b/src/python/test/test_representations.py index 43c914f3..cda1a15b 100755 --- a/src/python/test/test_representations.py +++ b/src/python/test/test_representations.py @@ -46,6 +46,32 @@ def test_multiple(): assert d1 == pytest.approx(d2, rel=0.02) +# Test sorted values as points order can be inverted, and sorted test is not documentation-friendly +# Note the test below must be up to date with the Atol class documentation +def test_atol_doc(): + a = np.array([[1, 2, 4], [1, 4, 0], [1, 0, 4]]) + b = np.array([[4, 2, 0], [4, 4, 0], [4, 0, 2]]) + c = np.array([[3, 2, -1], [1, 2, -1]]) + + atol_vectoriser = Atol(quantiser=KMeans(n_clusters=2, random_state=202006)) + # Atol will do + # X = np.concatenate([a,b,c]) + # kmeans = KMeans(n_clusters=2, random_state=202006).fit(X) + # kmeans.labels_ will be : array([1, 0, 1, 0, 0, 1, 0, 0]) + first_cluster = np.asarray([a[0], a[2], b[2]]) + second_cluster = np.asarray([a[1], b[0], b[2], c[0], c[1]]) + + # Check the center of the first_cluster and second_cluster are in Atol centers + centers = atol_vectoriser.fit(X=[a, b, c]).centers + np.isclose(centers, first_cluster.mean(axis=0)).all(1).any() + np.isclose(centers, second_cluster.mean(axis=0)).all(1).any() + + vectorization = atol_vectoriser.transform(X=[a, b, c]) + assert np.allclose(vectorization[0], atol_vectoriser(a)) + assert np.allclose(vectorization[1], atol_vectoriser(b)) + assert np.allclose(vectorization[2], atol_vectoriser(c)) + + def test_dummy_atol(): a = np.array([[1, 2, 4], [1, 4, 0], [1, 0, 4]]) b = np.array([[4, 2, 0], [4, 4, 0], [4, 0, 2]]) |