VisionLanguageActionOptions

Options & Configuration · AiDotNet.VisionLanguage.Robotics

Base configuration options for Vision-Language-Action (VLA) models.

Inheritance: ModelOptions → GenerativeVLMOptions → VisionLanguageActionOptions

Known derived types: GR00TN1Options, HelixOptions, OctoOptions, PaLMEOptions, PiZeroOptions, RT2Options, ThreeDVLAOptions

For Beginners

These options configure the VisionLanguageAction model. Default values follow the original paper settings.

Constructors

Constructor Summary
VisionLanguageActionOptions() Initializes a new instance with default values.
VisionLanguageActionOptions(VisionLanguageActionOptions other) Initializes a new instance by copying from another instance.

Properties

Property Summary
int ActionDimension { get; set; } Gets or sets the action space dimensionality (e.g., number of joint DOFs).
int ObservationHistory { get; set; } Gets or sets the observation history length (number of past frames).
int PredictionHorizon { get; set; } Gets or sets the maximum action prediction horizon (number of future steps).
string LanguageModelName { get; set; } Gets or sets the language model backbone name.