Class LoraTrainingConfig

The Lora training method hyperparameters.

Inheritance

object

TrainingConfig

LoraTrainingConfig

Inherited Members

TrainingConfig.TotalTrainingEpochs

TrainingConfig.LearningRate

TrainingConfig.TrainingBatchSize

TrainingConfig.EarlyStoppingPatience

TrainingConfig.EarlyStoppingThreshold

TrainingConfig.LogModelMetricsIntervalInSteps

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Namespace: Oci.GenerativeaiService.Models

Assembly: OCI.DotNetSDK.Generativeai.dll

Syntax

public class LoraTrainingConfig : TrainingConfig

Properties

LoraAlpha

Declaration

[JsonProperty(PropertyName = "loraAlpha")]
public int? LoraAlpha { get; set; }

Property Value

Type	Description
int?	This parameter represents the scaling factor for the weight matrices in LoRA.

LoraDropout

Declaration

[JsonProperty(PropertyName = "loraDropout")]
public double LoraDropout { get; set; }

Property Value

Type	Description
double	This parameter indicates the dropout probability for LoRA layers.

LoraR

Declaration

[JsonProperty(PropertyName = "loraR")]
public int? LoraR { get; set; }

Property Value

Type	Description
int?	This parameter represents the LoRA rank of the update matrices.