|View source on GitHub|
Laplacian regularizer for PWL calibration layer.
tfl.pwl_calibration_layer.LaplacianRegularizer( l1=0.0, l2=0.0, is_cyclic=False )
Calibrator Laplacian regularization penalizes the change in the calibration output. It is defined to be:
l1 * ||delta||_1 + l2 * ||delta||_2^2
output_keypoints[1:end] - output_keypoints[0:end-1].
l1: l1 regularization amount as float.
l2: l2 regularization amount as float.
is_cyclic: Whether the first and last keypoints should take the same output value.
__call__( x )
Returns regularization loss.
x: Tensor of shape:
(k, units)which represents weights of PWL calibration layer. First row of weights is bias term. All remaining represent delta in y-value compare to previous point (segment heights).
@classmethod from_config( cls, config )
Creates a regularizer from its config.
This method is the reverse of
capable of instantiating the same regularizer from the config
This method is used by Keras
model_to_estimator, saving and
loading models to HDF5 formats, Keras model cloning, some visualization
utilities, and exporting models to and from JSON.
config: A Python dictionary, typically the output of get_config.
A regularizer instance.
Standard Keras config for serialization.