Join us at TensorFlow World, Oct 28-31. Use code TF20 for 20% off select passes. Register now

Module: nsl.configs

View source on GitHub

Configuration classes and APIs for Neural Structured Learning.

Classes

class AdvNeighborConfig: AdvNeighborConfig contains configs for generating adversarial neighbors.

class AdvRegConfig: AdvRegConfig contains configs for adversarial regularization.

class AdvTargetConfig: AdvTargetConfig contains configs for selecting targets to be attacked.

class AdvTargetType: Types of adversarial targeting.

class DecayConfig: DecayConfig contains configs for computing decayed value.

class DecayType: Types of decay.

class DistanceConfig: DistanceConfig contains configs for computing distances.

class DistanceType: Types of distance.

class GraphNeighborConfig: GraphNeighborConfig specifies neighbor attributes for graph regularization.

class GraphRegConfig: GraphRegConfig contains the configuration for graph regularization.

class IntegrationConfig: IntegrationConfig contains configs for computing multimodal integration.

class IntegrationType: Types of integration for multimodal fusion.

class NormType: Types of norms.

class TransformType: Types of nonlinear functions to be applied .

class VirtualAdvConfig: VirtualAdvConfig contains configs for virtual adversarial training.

Functions

make_adv_reg_config(...): Creates AdvRegConfig object.

Other Members

  • DEFAULT_ADVERSARIAL_PARAMS
  • DEFAULT_DISTANCE_PARAMS