VideoCLIPOptions

Options & Configuration · AiDotNet.NeuralNetworks.Options

Configuration options for the VideoCLIPNeuralNetwork.

Inheritance: ModelOptions → NeuralNetworkOptions → ModelHyperparameterOptions → VisionLanguageInputOptions → VisionLanguageModelOptions → VideoCLIPOptions

Constructors

Constructor Summary
VideoCLIPOptions() Initializes a new instance of the VideoCLIPOptions class carrying this model’s shipped defaults.

Properties

Property Summary
TemporalAggregationType TemporalAggregation { get; set; } Gets or sets the temporal aggregation.
double FrameRate { get; set; } Gets or sets the nominal sampling rate associated with the supplied video frames.
int NumFrameEncoderLayers { get; set; } Gets or sets the number of native transformer blocks applied to each selected frame.
int NumFrames { get; set; } Gets or sets the number of supplied video frames selected for each clip embedding.
int NumTemporalLayers { get; set; } Gets or sets the number of native temporal-transformer blocks that combine frame features.
int NumTextLayers { get; set; } Gets or sets the number of native text-transformer blocks used to encode a caption.
int TextHiddenDim { get; set; } Gets or sets the number of features in each native text-token representation.

Methods

Method Summary
void Validate() Throws if a value this model requires has been left unset or is not positive.