1. Epic GamesDeveloper
  3. ULearningAgentsPolicy

ULearningAgentsPolicy

A policy that maps from observations to actions.

API > API/Plugins > API/Plugins/LearningAgents

Inheritance Hierarchy

References

   
Module LearningAgents
Header /Engine/Plugins/Experimental/LearningAgents/Source/LearningAgents/Public/LearningAgentsPolicy.h
Include #include "LearningAgentsPolicy.h"

Syntax

UCLASS (BlueprintType, Blueprintable, Meta=(BlueprintSpawnableComponent))  
class ULearningAgentsPolicy : public ULearningAgentsManagerListener

Remarks

A policy that maps from observations to actions.

Constructors

  Type Name Description
Public function   These constructors/destructors are needed to make forward declarations happy.
Public function    

Destructors

  Type Name Description
Public function Virtual    

Functions

  Type Name Description
Public function UFunction BlueprintCallable, Category void

DecodeAndSampleActions ( const float ActionNoiseScale
)

 Decodes and samples action vectors using the Decoder network.
Public function UFunction BlueprintCallable, Category void -- Blueprint public interface --
Public function UFunction BlueprintCallable, Category void Calling this function will run the underlying neural network on the previously encoded observations to populate the encoded actions.
Public function UFunction BlueprintPure, Category ULearningAgentsNeuralNetwork * Gets the current Encoder Network Asset being used
Public function UE::Learning::FNeuralNetworkFunction & Get a reference to this policy's decoder function object.
Public function UFunction BlueprintPure, Category ULearningAgentsNeuralNetwork * Gets the current Encoder Network Asset being used
Public function UE::Learning::FNeuralNetworkFunction & -- Non-blueprint public interface --
Public function Const UFunction BlueprintCallable, Category, Meta void

GetMemoryState ( TArray< float >& OutMemoryState,
const int32 AgentId
)

 Gets the current memory state for a given agent as represented by an abstract vector learned by the policy.
Public function Const TLearningArrayView< 2, const float > Get a const view to this policy's memory state.
Public function Const UFunction BlueprintPure, Category int32 Gets the size of the memory state
Public function Const TLearningArrayView< 2, const float > Get a const view to this policy's encoded observation vectors.
Public function UFunction BlueprintPure, Category ULearningAgentsNeuralNetwork * Gets the current Policy Network Asset being used
Public function UE::Learning::FNeuralNetworkPolicy & Get a reference to this policy's policy function object.
Public function Const TLearningArrayView< 2, const float > Get a const view to this policy's pre-evaluation memory state.
Public function Const bool Returns true if this policy has encoded observations for the given agent id. Otherwise, false.
Public function Static UFunction BlueprintCallable, Category, Meta ULearningAgentsPolicy *

MakePolicy ( ULearningAgentsManager*& InManager,
ULearningAgentsInteractor*& InInteractor,
TSubclassOf< ULearningAgentsPolicy > Class,
const FName Name,
ULearningAgentsNeuralNetwork* EncoderNeuralNetworkAsset,
ULearningAgentsNeuralNetwork* PolicyNeuralNetworkAsset,
ULearningAgentsNeuralNetwork* DecoderNeuralNetworkAsset,
const bool bReinitializeEncoderNetwork,
const bool bReinitializePolicyNetwork,
const bool bReinitializeDecoderNetwork,
const FLearningAgentsPolicySettings& PolicySettings,
const int32 Seed
)

 Constructs this object to be used with the given agent interactor and policy settings.
Public function Virtual void

OnAgentsAdded_Implementation ( const TArray< int32 >& AgentIds
)

  
Public function Virtual void  
Public function Virtual void

OnAgentsReset_Implementation ( const TArray< int32 >& AgentIds
)

  
Public function UFunction BlueprintCallable, Category void

RunInference ( const float ActionNoiseScale
)

 Calls GatherObservations, EncodeObservations, EvaluatePolicy, DecodeAndSampleActions, PerformActions
Public function UFunction BlueprintCallable, Category, Meta void

SetMemoryState ( const int32 AgentId,
const TArray< float >& InMemoryState
)

 Sets the current memory state for a given agent as represented by an abstract vector learned by the policy.
Public function UFunction BlueprintCallable, Category, Meta void

SetupPolicy ( ULearningAgentsManager*& InManager,
ULearningAgentsInteractor*& InInteractor,
ULearningAgentsNeuralNetwork* EncoderNeuralNetworkAsset,
ULearningAgentsNeuralNetwork* PolicyNeuralNetworkAsset,
ULearningAgentsNeuralNetwork* DecoderNeuralNetworkAsset,
const bool bReinitializeEncoderNetwork,
const bool bReinitializePolicyNetwork,
const bool bReinitializeDecoderNetwork,
const FLearningAgentsPolicySettings& PolicySettings,
const int32 Seed
)

 Initializes this object to be used with the given agent interactor and policy settings.

Ask questions and help your peers Developer Forums

Write your own tutorials or read those from others Learning Library

On this page