ML::UMAPParams Class Reference

#include <umapparams.h>

Collaboration diagram for ML::UMAPParams:

Public Types
enum	MetricType { EUCLIDEAN , CATEGORICAL }
enum	graph_build_algo { BRUTE_FORCE_KNN , NN_DESCENT }

Public Attributes
int	n_neighbors = 15
int	n_components = 2
int	n_epochs = 0
float	learning_rate = 1.0
float	min_dist = 0.1
float	spread = 1.0
float	set_op_mix_ratio = 1.0
float	local_connectivity = 1.0
float	repulsion_strength = 1.0
int	negative_sample_rate = 5
float	transform_queue_size = 4.0
int	verbosity = CUML_LEVEL_INFO
float	a = -1.0
float	b = -1.0
float	initial_alpha = 1.0
int	init = 1
graph_build_algo	build_algo = graph_build_algo::BRUTE_FORCE_KNN
nn_index_params	nn_descent_params = {}
int	target_n_neighbors = -1
MetricType	target_metric = CATEGORICAL
float	target_weight = 0.5
uint64_t	random_state = 0
bool	deterministic = true
cuvs::distance::DistanceType	metric = cuvs::distance::DistanceType::L2SqrtExpanded
float	p = 2.0
Internals::GraphBasedDimRedCallback *	callback = nullptr

Member Enumeration Documentation

graph_build_algo

enum ML::UMAPParams::graph_build_algo

Enumerator
BRUTE_FORCE_KNN
NN_DESCENT

MetricType

enum ML::UMAPParams::MetricType

Enumerator
EUCLIDEAN
CATEGORICAL

Member Data Documentation

a

float ML::UMAPParams::a = -1.0

More specific parameters controlling the embedding. If None these values are set automatically as determined by min_dist and spread.

b

float ML::UMAPParams::b = -1.0

More specific parameters controlling the embedding. If None these values are set automatically as determined by min_dist and spread.

build_algo

graph_build_algo ML::UMAPParams::build_algo = graph_build_algo::BRUTE_FORCE_KNN

KNN graph build algorithm

callback

Internals::GraphBasedDimRedCallback* ML::UMAPParams::callback = nullptr

deterministic

bool ML::UMAPParams::deterministic = true

Whether should we use deterministic algorithm. This should be set to true if random_state is provided, otherwise it's false. When it's true, cuml will have higher memory usage but produce stable numeric output.

init

int ML::UMAPParams::init = 1

Embedding initializer algorithm 0 = random layout 1 = spectral layout

initial_alpha

float ML::UMAPParams::initial_alpha = 1.0

Initial learning rate for SGD

learning_rate

float ML::UMAPParams::learning_rate = 1.0

Initial learning rate for the embedding optimization

local_connectivity

float ML::UMAPParams::local_connectivity = 1.0

The local connectivity required – i.e. the number of nearest neighbors that should be assumed to be connected at a local level. The higher this value the more connected the manifold becomes locally. In practice this should be not more than the local intrinsic dimension of the manifold.

metric

cuvs::distance::DistanceType ML::UMAPParams::metric = cuvs::distance::DistanceType::L2SqrtExpanded

min_dist

float ML::UMAPParams::min_dist = 0.1

The effective minimum distance between embedded points. Smaller values will result in a more clustered/clumped embedding where nearby points on the manifold are drawn closer together, while larger values will result on a more even dispersal of points. The value should be set relative to the spread value, which determines the scale at which embedded points will be spread out.

n_components

int ML::UMAPParams::n_components = 2

Number of features in the final embedding

n_epochs

int ML::UMAPParams::n_epochs = 0

Number of epochs to use in the training of the embedding.

n_neighbors

int ML::UMAPParams::n_neighbors = 15

The number of neighbors to use to approximate geodesic distance. Larger numbers induce more global estimates of the manifold that can miss finer detail, while smaller values will focus on fine manifold structure to the detriment of the larger picture.

negative_sample_rate

int ML::UMAPParams::negative_sample_rate = 5

The number of negative samples to select per positive sample in the optimization process. Increasing this value will result in greater repulsive force being applied, greater optimization cost, but slightly more accuracy.

nn_descent_params

nn_index_params ML::UMAPParams::nn_descent_params = {}

p

float ML::UMAPParams::p = 2.0

random_state

uint64_t ML::UMAPParams::random_state = 0

repulsion_strength

float ML::UMAPParams::repulsion_strength = 1.0

Weighting applied to negative samples in low dimensional embedding optimization. Values higher than one will result in greater weight being given to negative samples.

set_op_mix_ratio

float ML::UMAPParams::set_op_mix_ratio = 1.0

Interpolate between (fuzzy) union and intersection as the set operation used to combine local fuzzy simplicial sets to obtain a global fuzzy simplicial sets. Both fuzzy set operations use the product t-norm. The value of this parameter should be between 0.0 and 1.0; a value of 1.0 will use a pure fuzzy union, while 0.0 will use a pure fuzzy intersection.

spread

float ML::UMAPParams::spread = 1.0

The effective scale of embedded points. In combination with min_dist this determines how clustered/clumped the embedded points are.

target_metric

MetricType ML::UMAPParams::target_metric = CATEGORICAL

target_n_neighbors

int ML::UMAPParams::target_n_neighbors = -1

The number of nearest neighbors to use to construct the target simplicial set. If set to -1, use the n_neighbors value.

target_weight

float ML::UMAPParams::target_weight = 0.5

transform_queue_size

float ML::UMAPParams::transform_queue_size = 4.0

For transform operations (embedding new points using a trained model_ this will control how aggressively to search for nearest neighbors. Larger values will result in slower performance but more accurate nearest neighbor evaluation.

verbosity

int ML::UMAPParams::verbosity = CUML_LEVEL_INFO

Control logging level during algorithm execution

---

The documentation for this class was generated from the following file:

• include/cuml/manifold/umapparams.h