library.phases.phases_implementation.feature_analysis.feature_selection.manual module

class library.phases.phases_implementation.feature_analysis.feature_selection.manual.LowVariancesFeatureReduction(dataset: Dataset)

Bases: ManualFeatureSelectionFactory

constant_features_reduction()

Removes constant features from the dataset.

fit(threshold: float = 0.01, delete_features: bool = True, save_plots: bool = False, save_path: str = '')

Removes the features with low variance.

class library.phases.phases_implementation.feature_analysis.feature_selection.manual.ManualFeatureSelection(dataset: Dataset)

Bases: object

fit(type: str, threshold: float, delete_features: bool, save_plots: bool, save_path: str)

class library.phases.phases_implementation.feature_analysis.feature_selection.manual.ManualFeatureSelectionFactory(dataset: Dataset)

Bases: ABC

abstractmethod fit(threshold: float, delete_features: bool, save_plots: bool, save_path: str)

class library.phases.phases_implementation.feature_analysis.feature_selection.manual.MutualInformationFeatureReduction(dataset: Dataset)

Bases: ManualFeatureSelectionFactory

fit(threshold: float, delete_features: bool, save_plots: bool = False, save_path: str = '')

class library.phases.phases_implementation.feature_analysis.feature_selection.manual.PCAFeatureReduction(dataset: Dataset)

Bases: ManualFeatureSelectionFactory

fit(threshold: float = 0.95, delete_features: bool = True, save_plots: bool = False, save_path: str = '')

Reduces the number of features using PCA.

class library.phases.phases_implementation.feature_analysis.feature_selection.manual.VIFElimination(dataset: Dataset)

Bases: ManualFeatureSelectionFactory

fit(threshold=10, delete_features: bool = True, save_plots: bool = False, save_path: str = '')

Starts the VIF elimination process. Eliminates in all sets. Note: this is computationally expensive for high-feature datasets.

Parameters:

threshold (float) – The threshold for the VIF.

Return type:

None