skada.LinearOTMapping
- skada.LinearOTMapping(base_estimator=None, reg=1.0, bias=True)[source]
Returns a the linear OT mapping method with adapter and estimator.
Uses Gaussian Monge mapping to align source and target domains as proposed in [7].
- Parameters:
- base_estimatorobject, optional (default=None)
The base estimator to fit on the target dataset.
- regfloat, (default=1e-08)
regularization added to the diagonals of covariances.
- bias: bool, optional (default=True)
estimate bias.
- Returns:
- pipelinePipeline
Pipeline containing linear OT mapping adapter and base estimator.
References
[6]N. Courty, R. Flamary, D. Tuia and A. Rakotomamonjy, Optimal Transport for Domain Adaptation, in IEEE Transactions on Pattern Analysis and Machine Intelligence
Examples using skada.LinearOTMapping
Comparison of DA classification methods
Comparison of DA classification methods