src.cmfsapy package

Submodules

src.cmfsapy.data module

src.cmfsapy.data.gen_ncube(n, d, realization_id=0)

Generate reproducible pseudo-random datasets given n, d, and id.

Parameters
  • n (int) – number of data points

  • d (int) – dimensionality of the data

  • realization_id (int) – which realization you want from the

Returns

n x d numpy array dataset

src.cmfsapy.theoretical module

src.cmfsapy.theoretical.theoretical_fsa_pdf(w, k, d)

Analitic pdf for Szepesvari-Farahmand dimension estimator

Parameters
  • of float w (numpy.ndarray) – the value of the estimator

  • k (int) – kNN neighborhood size

  • d (int) – the intrinsic dimensionality of the manifold

Returns

pdf value at w

Return type

numpy.ndarray of float

Module contents