pottok.OptimalTransportGridSearch

class pottok.OptimalTransportGridSearch(transport_function=<class 'ot.da.MappingTransport'>, params=None, verbose=True)

Initialize Python Optimal Transport suitable for validation.

Parameters

• transport_function (class of ot.da, optional (default=ot.da.MappingTransport)) – from ot.da. e.g ot

• params_ot (dict, optional (default=None)) – parameters of the optimal transport funtion.

• verbose (boolean, optional (default=True)) – Gives informations about the object

__init__(transport_function=<class 'ot.da.MappingTransport'>, params=None, verbose=True)

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__([transport_function, params, verbose])	Initialize self.
assess_transport(Xs_transform[, record, path])	OA comparison before and after OT
assess_transport_circular(Xs_transform[, …])	OA comparison before and after OT
fit_circular([metrics, greater_is_better])	Learn domain adaptation model with circular tuning (fitting).
fit_crossed([cv_ai, cv_ot, classifier, …])	Learn domain adaptation model with crossed tuning (fitting).
load_model(path)	Load model previously saved with SuperLearner.save_model(path).
predict_transfer(data)	Predict model using domain adaptation.
preprocessing(Xs[, ys, Xt, yt, group_s, …])	Stock the input parameters in the object and scaled it if it is asked.
save_model(path)	Save model ‘myModel.npz’ to be loaded later via SuperLearner.load_model(path)
valid_fit_crossed(Xs_transform)	OA comparison before and after OT with Xt_test