persim.PersistenceLandscaper¶
- class persim.PersistenceLandscaper(hom_deg: int = 0, start: float = None, stop: float = None, num_steps: int = 500, flatten: bool = False)[source]¶
A scikit-learn transformer for converting persistence diagrams into persistence landscapes.
- Parameters:
hom_deg (int) – Homological degree of persistence landscape.
start (float, optional) – Starting value of approximating grid.
stop (float, optional) – Stopping value of approximating grid.
num_steps (int, optional) – Number of steps of approximating grid.
flatten (bool, optional) – Determines if the resulting values are flattened.
Examples
First instantiate the PersistenceLandscaper:
>>> from persim import PersistenceLandscaper >>> pl = PersistenceLandscaper(hom_deg=0, num_steps=10, flatten=True) >>> print(pl) PersistenceLandscaper(hom_deg=1,num_steps=10)
The fit() method is first called on a list of (-,2) numpy.ndarrays to determine the start and stop parameters of the approximating grid:
>>> ex_dgms = [np.array([[0,3],[1,4]]),np.array([[1,4]])] >>> pl.fit(ex_dgms) PersistenceLandscaper(hom_deg=0, start=0, stop=4, num_steps=10)
The transform() method will then compute the values of the landscape functions on the approximated grid. The flatten flag determines if the output should be a flattened numpy array:
>>> ex_pl = pl.transform(ex_dgms) >>> ex_pl array([0. , 0.44444444, 0.88888889, 1.33333333, 1.33333333, 1.33333333, 1.33333333, 0.88888889, 0.44444444, 0. , 0. , 0. , 0. , 0.44444444, 0.88888889, 0.88888889, 0.44444444, 0. , 0. , 0. ])
- __init__(hom_deg: int = 0, start: float = None, stop: float = None, num_steps: int = 500, flatten: bool = False)[source]¶
Methods
__init__
([hom_deg, start, stop, num_steps, ...])fit
(X[, y])Find optimal start and stop parameters for approximating grid.
fit_transform
(X[, y])Fit to data, then transform it.
get_metadata_routing
()Get metadata routing of this object.
get_params
([deep])Get parameters for this estimator.
set_output
(*[, transform])Set output container.
set_params
(**params)Set the parameters of this estimator.
transform
(X[, y])Construct persistence landscape values.