skada.CORALAdapter
- skada.CORALAdapter(reg='auto', assume_centered=False)[source]
Estimator based on Correlation Alignment [1]_.
See [5] for details.
- Parameters:
- reg'auto' or float, default="auto"
The regularization parameter of the covariance estimator. Possible values:
None: no shrinkage).
'auto': automatic shrinkage using the Ledoit-Wolf lemma.
float between 0 and 1: fixed shrinkage parameter.
- assume_centered: bool, default=False
If True, data are not centered before computation.
- Attributes:
- mean_source_: array, shape (n_features,)
- mean_target_: array, shape (n_features,)
- cov_source_inv_sqrt_: array, shape (n_features, n_features)
Inverse of the square root of covariance of the source data with regularization.
- cov_target_sqrt_: array, shape (n_features, n_features)
Square root of covariance of the target data with regularization.
References
[5]Baochen Sun, Jiashi Feng, and Kate Saenko. Correlation Alignment for Unsupervised Domain Adaptation. In Advances in Computer Vision and Pattern Recognition, 2017.
