ML::HDBSCAN::HELPER Namespace Reference

Functions
void	compute_core_dists (const raft::handle_t &handle, const float *X, float *core_dists, size_t m, size_t n, ML::distance::DistanceType metric, int min_samples)
	Compute the core distances for each point in the training matrix. More...
void	compute_inverse_label_map (const raft::handle_t &handle, HDBSCAN::Common::CondensedHierarchy< int, float > &condensed_tree, size_t n_leaves, HDBSCAN::Common::CLUSTER_SELECTION_METHOD cluster_selection_method, rmm::device_uvector< int > &inverse_label_map, bool allow_single_cluster, int max_cluster_size, float cluster_selection_epsilon)
	Compute the map from final, normalize labels to the labels in the CondensedHierarchy. More...

Function Documentation

compute_core_dists()

void ML::HDBSCAN::HELPER::compute_core_dists	(	const raft::handle_t &	handle,
		const float *	X,
		float *	core_dists,
		size_t	m,
		size_t	n,
		ML::distance::DistanceType	metric,
		int	min_samples
	)

Compute the core distances for each point in the training matrix.

Parameters

[in]	handle	raft handle for resource reuse
[in]	X	array (size m, n) on device in row-major format
[out]	core_dists	array (size m, 1) of core distances
	m	number of rows in X
	n	number of columns in X
	metric	distance metric to use
	min_samples	minimum number of samples to use for computing core distances

compute_inverse_label_map()

void ML::HDBSCAN::HELPER::compute_inverse_label_map	(	const raft::handle_t &	handle,
		HDBSCAN::Common::CondensedHierarchy< int, float > &	condensed_tree,
		size_t	n_leaves,
		HDBSCAN::Common::CLUSTER_SELECTION_METHOD	cluster_selection_method,
		rmm::device_uvector< int > &	inverse_label_map,
		bool	allow_single_cluster,
		int	max_cluster_size,
		float	cluster_selection_epsilon
	)

Compute the map from final, normalize labels to the labels in the CondensedHierarchy.

Parameters

[in]	handle	raft handle for resource reuse
[in]	condensed_tree	the Condensed Hierarchy object
[in]	n_leaves	number of leaves in the input data
[in]	cluster_selection_method	cluster selection method
[out]	inverse_label_map	rmm::device_uvector of size 0. It will be resized during the computation
[in]	allow_single_cluster	allow single cluster
[in]	max_cluster_size	max cluster size
[in]	cluster_selection_epsilon	cluster selection epsilon