Missed TensorFlow Dev Summit? Check out the video playlist. Watch recordings

tfl.categorical_calibration_layer.CategoricalCalibrationConstraints

View source on GitHub

Monotonicity and bounds constraints for categorical calibration layer.

tfl.categorical_calibration_layer.CategoricalCalibrationConstraints(
    output_min=None, output_max=None, monotonicities=None
)

Updates the weights of CategoricalCalibration layer to satify bound and monotonicity constraints. The update is an approximate L2 projection into the constrained parameter space.

Args:

  • output_min: Minimum possible output of categorical function.
  • output_max: Maximum possible output of categorical function.
  • monotonicities: Monotonicities of CategoricalCalibration layer.

Methods

__call__

View source

__call__(
    w
)

Applies constraints to w.

get_config

View source

get_config()

Standard Keras config for serialization.