unreal.LearningAgentsFlowMatchingTrainerTrainingSettings

class unreal.LearningAgentsFlowMatchingTrainerTrainingSettings(hyperparameters: LearningAgentsFlowMatchingHyperparameters = [], logging_settings: LearningAgentsFlowMatchingLoggingSettings = [], validation_settings: LearningAgentsFlowMatchingValidationSettings = Ellipsis, device: LearningAgentsTrainingDevice = Ellipsis)

Bases: StructBase

The configurable settings for the flow matching training process.

C++ Source:

  • Plugin: LearningAgents

  • Module: LearningAgentsTraining

  • File: LearningAgentsFlowMatchingTrainer.h

Editor Properties: (see get_editor_property/set_editor_property)

  • device (LearningAgentsTrainingDevice): [Read-Write] The device to train on.

  • embed_normalizations_in_weights (bool): [Read-Write] Whether to embed dataset normalizations into encoder and decoder networks. This is recommended to set to true to stabilize trainings. <Important> The normalization option in action/observation schemas must be set to one of the “auto” options.

  • hyperparameters (LearningAgentsFlowMatchingHyperparameters): [Read-Write]

  • logging_settings (LearningAgentsFlowMatchingLoggingSettings): [Read-Write]

  • validation_settings (LearningAgentsFlowMatchingValidationSettings): [Read-Write]

property device: LearningAgentsTrainingDevice

[Read-Write] The device to train on.

Type:

(LearningAgentsTrainingDevice)

property hyperparameters: LearningAgentsFlowMatchingHyperparameters

[Read-Write]

Type:

(LearningAgentsFlowMatchingHyperparameters)

property logging_settings: LearningAgentsFlowMatchingLoggingSettings

[Read-Write]

Type:

(LearningAgentsFlowMatchingLoggingSettings)

property validation_settings: LearningAgentsFlowMatchingValidationSettings

[Read-Write]

Type:

(LearningAgentsFlowMatchingValidationSettings)