VineDensity.__init__

VineDensity.__init__(backend=None, batch_size=100, random_state=None)

Vine-copula based density estimator.

Parameters:

backendVinecopBackend or compatible, default=None

Backend instance bundling fit-time controls and an optional pre-specified structure. None resolves to a default VinecopBackend at fit time, which calls from_data() with the nonparametric tll pair family. Pass TorchVinecopBackend for the PyTorch backend.

batch_sizeint, default=100

Number of test points processed per batch when evaluating the density. Higher values trade memory for throughput.

random_stateint, RandomState instance or None, default=None

Seeds the RNG used by stochastic operations (cdf quasi-MC, sample). Resolved via sklearn.utils.check_random_state inside fit.