ArmCognitiveServicesModelFactory.DeploymentSpeculativeDecoding Method

Definition

Namespace:

Azure.ResourceManager.CognitiveServices.Models

Assembly:

Azure.ResourceManager.CognitiveServices.dll

Package:

Azure.ResourceManager.CognitiveServices v1.6.0-beta.3

Source:

ArmCognitiveServicesModelFactory.cs

public static Azure.ResourceManager.CognitiveServices.Models.DeploymentSpeculativeDecoding DeploymentSpeculativeDecoding(Azure.ResourceManager.CognitiveServices.Models.CognitiveServicesAccountDeploymentModel draftModel = default, int? draftTokenCount = default);

static member DeploymentSpeculativeDecoding : Azure.ResourceManager.CognitiveServices.Models.CognitiveServicesAccountDeploymentModel * Nullable<int> -> Azure.ResourceManager.CognitiveServices.Models.DeploymentSpeculativeDecoding

Public Shared Function DeploymentSpeculativeDecoding (Optional draftModel As CognitiveServicesAccountDeploymentModel = Nothing, Optional draftTokenCount As Nullable(Of Integer) = Nothing) As DeploymentSpeculativeDecoding

Parameters

draftModel

CognitiveServicesAccountDeploymentModel

Draft model used to generate speculative decoding tokens.

draftTokenCount

Nullable<Int32>

The number of draft tokens attempted per speculation step.

Returns

DeploymentSpeculativeDecoding

A new DeploymentSpeculativeDecoding instance for mocking.