Table of Contents

Class SupervisedTuningSpec

Namespace
GenerativeAI.Types
Assembly
GenerativeAI.dll

Tuning specification for Supervised Fine-Tuning for first-party models.

public class SupervisedTuningSpec
Inheritance
SupervisedTuningSpec
Inherited Members
Extension Methods

Properties

EvaluationConfig

Optional. Evaluation config for the tuning job.

[JsonPropertyName("evaluationConfig")]
public EvaluationConfig? EvaluationConfig { get; set; }

Property Value

EvaluationConfig

ExportLastCheckpointOnly

Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. Default is false.

[JsonPropertyName("exportLastCheckpointOnly")]
public bool? ExportLastCheckpointOnly { get; set; }

Property Value

bool?

HyperParameters

Optional. Hyperparameters for SFT.

[JsonPropertyName("hyperParameters")]
public SupervisedHyperParameters? HyperParameters { get; set; }

Property Value

SupervisedHyperParameters

TrainingDatasetUri

Required. Training dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.

[JsonPropertyName("trainingDatasetUri")]
public string? TrainingDatasetUri { get; set; }

Property Value

string

TuningMode

Optional. Tuning mode (FULL or PEFT_ADAPTER).

[JsonPropertyName("tuningMode")]
public TuningMode? TuningMode { get; set; }

Property Value

TuningMode?

ValidationDatasetUri

Optional. Validation dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset.

[JsonPropertyName("validationDatasetUri")]
public string? ValidationDatasetUri { get; set; }

Property Value

string