Module: tf.contrib.kernel_methods

Ops and estimators that enable explicit kernel methods in TensorFlow.

Classes

class KernelLinearClassifier: Linear classifier using kernel methods as feature preprocessing.

class RandomFourierFeatureMapper: Class that implements Random Fourier Feature Mapping (RFFM) in TensorFlow.

Functions

sparse_multiclass_hinge_loss(...): Adds Ops for computing the multiclass hinge loss.