skada.SubspaceAlignmentAdapter

skada.SubspaceAlignmentAdapter(n_components=None, random_state=None)

Domain Adaptation Using Subspace Alignment.

See [8] for details.

Parameters:

n_componentsint, default=None

The numbers of components to learn with PCA. If n_components is not set all components are kept:

n_components == min(n_samples, n_features)

random_stateint, RandomState instance or None, default=None

Determines random number generation for dataset creation. Pass an int for reproducible output across multiple function calls.

References

[8]

Basura Fernando et. al. Unsupervised Visual Domain Adaptation Using Subspace Alignment. In IEEE International Conference on Computer Vision, 2013.

Attributes:

`pca_source_`object

The PCA object fitted on the source data.

`pca_target_`object

The PCA object fitted on the target data.