vectorose.data

Sample datasets for running VectoRose.

This module grants access to the sample data used in the documentation. These datasets may be loaded and accessed as NumPy arrays without requiring any external download.

Warning

This module does not exist as an automatic top-level import in vectorose, and must therefore be explicitly imported.

Classes

SampleData

Interface for accessing sample datasets.

Package Contents

class vectorose.data.SampleData

Bases: enum.Enum

Interface for accessing sample datasets.

This enumeration provides a list of sample datasets provided with VectoRose, as well as a simple interface to load them. The string values associated with each enumerated instance correspond to the base filename for the respective dataset.

Warning

All datasets are assumed to be stored as a NumPy array (*.npy).

Notes

This system is used to ensure the flexibility to easily add new sample data in the future.

CLUSTER_GIRDLE = 'cluster_girdle'

Overlapping cluster and girdle with different magnitudes.

TWO_CLUSTERS = 'two_clusters'

Two clusters with different magnitudes and orientations.

TWISTED_BLOCKS = 'twisted_blocks'

Anisotropy of offset rotated layers of cylinders.

Notes

Computed using Dragonfly 3D World.

load() → numpy.ndarray

Load the current dataset to use with VectoRose.

Returns:

numpy.ndarray – Array of shape (n, d) containing the loaded vectors. If the specific dataset has location coordinates, then d == 6. Otherwise, d == 3. n represents the number of vectors in the dataset.