base

base

Base class for dimmer package.

Classes

Name	Description
DimmerBase	Base class for local homology stratification determination.

DimmerBase

base.DimmerBase(
    radii=None,
    neighbors=None,
    collapse_edges=False,
    n_jobs=1,
    threshold=None,
    outlier_label=-1,
    tree_type=None,
)

Base class for local homology stratification determination.

This class should not be called directly, but rather GAD or Intersecter.

Parameters

Name	Type	Description	Default
radii	tuple[float, float] | None	The radii of the annular neighborhood at every point. Only one of radii or neighbors should be specified.	None
neighbors	tuple[int, int] | None	The number of neighbors of the annular neighborhood at every point. Only one of radii or neighbors should be specified.	None
collapse_edges	bool | None	The flag determining whether to collapse edges (see giotto-tda’s documentation on VietorisRipsPersistence). The default behavior is False if max_dim < 4; otherwise it is True.	False
n_jobs	int	The number of processors to use.	1
threshold	float | str | None	The real parameter at which persistent features are counted. If "average" is passed, then the average annular thickness will be used for points.	None
outlier_label	float	The dimension to assign to any outliers.	-1
tree_type	str | None	The style of tree to use for nearest neighbors computation. Can be anything acceptable by sklearn.neighbors.RadiusNeighborsTransformer: {"auto", "ball_tree", "kd_tree", "brute"}, default is "auto" .	None

Note: It seems like collapse_edges should be True if the homological dimension is “large”, but what this means in practice is still not clear. This could also cause slow down if a space has strata of many different dimensions.

Methods

Name	Description
predict_dw	Compute GAD classification of each point in a specified degree.

predict_dw

base.DimmerBase.predict_dw(
    X,
    y=None,
    hom_deg=1,
    point_indices=None,
    nbhd_dict=None,
)

Compute GAD classification of each point in a specified degree.

This method computes the local persistent homology of a point cloud in a specified homological degree and counts the points in its persistence diagram. The local neighborhood is specified by self.radii or self.neighbors. The persistent homology of each neighborhood is computed in degree hom_deg. The points in the persistence diagram of lifetime greater than outer_rad - inner_rad are counted. Boundary, manifold, and stratification points are those with 0, 1, and greater than 1, respectively, such points in their diagram when hom_deg is positive. Because of reduced homology, both boundary and manifold points have 0 such points in their persistence diagrams when hom_deg is zero.

Parameters

Name	Type	Description	Default
X	np.ndarray[tuple[int, int], np.dtype[np.float64]]	Array containing n_samples points within ambient Euclidean space of dimension n_features.	required
y	np.ndarray[tuple[int], np.dtype[np.float64]] | None	Ignored.	None
hom_deg	int	The homological degree to compute.	1
point_indices	np.ndarray[tuple[int], np.dtype[np.int64]] | None	Indices of X at which to compute the local cohomology.	None
nbhd_dict	dict[int, np.ndarray[tuple[int], np.dtype[np.int64]]] | None	Dictionary containing the annular neighborhood each key, indexed on self.subset.	None

Returns

Name	Type	Description
strat_points	list[int]	list The indices of X determined to be stratification points.
boundary_points	list[int]	list The indices of X determined to be boundary points.
manifold_points	list[int]	list The indices of X determined to be manifold points.