depynd.markov_networks._gsmple()

depynd.markov_networks._gsmple(X, lamb=0.0, **kwargs)

Learn the structure of Markov random field with glow-shrink maximum pseudolikelihood estimation (GS-MPLE) [takashina2018structure].

Parameters:
  • X (array, shape (n_samples, n_features)) – Observations of variables.
  • lamb (float) – Threshold for independence tests.
  • kwargs (dict, default None) – Optional parameters for MI estimation.
Returns:

adj – Estimated adjacency matrix of an MRF.
Return type:

array, shape (n_features, n_features)

References

[takashina2018structure]Takashina, Yuya, et al. “Structure Learning of Markov Random Fields through Grow-Shrink Maximum Pseudolikelihood Estimation.” Eighth International Workshop on Statistical Relational AI (2018).