public static class LoraTrainingConfig.Builder extends Object
Constructor and Description |
---|
Builder() |
public LoraTrainingConfig.Builder totalTrainingEpochs(Integer totalTrainingEpochs)
public LoraTrainingConfig.Builder learningRate(Double learningRate)
public LoraTrainingConfig.Builder trainingBatchSize(Integer trainingBatchSize)
public LoraTrainingConfig.Builder earlyStoppingPatience(Integer earlyStoppingPatience)
public LoraTrainingConfig.Builder earlyStoppingThreshold(Double earlyStoppingThreshold)
public LoraTrainingConfig.Builder logModelMetricsIntervalInSteps(Integer logModelMetricsIntervalInSteps)
public LoraTrainingConfig.Builder loraR(Integer loraR)
This parameter represents the LoRA rank of the update matrices.
loraR
- the value to setpublic LoraTrainingConfig.Builder loraAlpha(Integer loraAlpha)
This parameter represents the scaling factor for the weight matrices in LoRA.
loraAlpha
- the value to setpublic LoraTrainingConfig.Builder loraDropout(Double loraDropout)
This parameter indicates the dropout probability for LoRA layers.
loraDropout
- the value to setpublic LoraTrainingConfig build()
public LoraTrainingConfig.Builder copy(LoraTrainingConfig model)
Copyright © 2016–2024. All rights reserved.