depynd.feature_selection._mrmr()

depynd.feature_selection._mrmr(X, y, lamb, k, **kwargs)

Select effective features in X on predicting y using minimum redundancy maximum relevance feature selection [peng2005feature].

Parameters:
  • X (array-like, shape (n_samples, n_features)) – Observations of feature variables.
  • y (array-like, shape (n_samples)) – Observations of the target variable.
  • lamb (float or None) – Threshold for independence tests. Ignored if k is specified.
  • k (int or None) – Number of selected features.
  • kwargs (dict) – Optional parameters for MI estimation.
Returns:

indices – Indices of the selected features.
Return type:

list

References

[peng2005feature]Peng, Hanchuan, Fuhui Long, and Chris Ding. “Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy.” IEEE Transactions on pattern analysis and machine intelligence 27.8 (2005): 1226-1238.