archetypes.datasets.make_archetypal_dataset#
- archetypes.datasets.make_archetypal_dataset(archetypes, shape, alpha=1.0, noise=0.0, generator=None) -> (<built-in function array>, <class 'list'>)#
Generate a dataset from archetypes.
- Parameters:
- archetypesnp.ndarray
The archetypes.
- shapetuple of int
The shape of the dataset.
- alphafloat, default=1.
The concentration parameter of the Dirichlet distribution.
- noisefloat, default=0.
The standard deviation of the gaussian noise.
- generatorint, Generator instance or None, default=None
Determines random number generation for dataset creation. Pass an int for reproducible output across multiple function calls.
- Returns:
- np.ndarray
The dataset.
- list of np.ndarray
The labels for each dimension.