skada.EntropicOTMapping
- skada.EntropicOTMapping(base_estimator=None, metric='sqeuclidean', norm=None, max_iter=1000, reg_e=1.0, tol=1e-08)[source]
EntropicOTMapping pipeline with adapter and estimator.
see [6] for details.
- Parameters:
- base_estimatorobject, optional (default=None)
The base estimator to fit on the target dataset.
- reg_efloat, default=1
Entropic regularization parameter.
- metricstr, optional (default="sqeuclidean")
The ground metric for the Wasserstein problem.
- norm{'median', 'max', 'log', 'loglog'} (default=None)
If given, normalize the ground metric to avoid numerical errors that can occur with large metric values.
- max_iterint, float, optional (default=1000)
The minimum number of iteration before stopping the optimization of the Sinkhorn algorithm if it has not converged
- tolfloat, optional (default=10e-9)
The precision required to stop the optimization of the Sinkhorn algorithm.
- Returns:
- pipelinePipeline
Pipeline containing EntropicOTMapping 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
