com.oracle.bmc.generativeai.model

Class TrainingConfig

java.lang.Object

com.oracle.bmc.http.client.internal.ExplicitlySetBmcModel

com.oracle.bmc.generativeai.model.TrainingConfig

Direct Known Subclasses:

LoraTrainingConfig, TFewTrainingConfig, VanillaTrainingConfig

@Generated(value="OracleSDKGenerator",
           comments="API Version: 20231130")
public class TrainingConfig
extends com.oracle.bmc.http.client.internal.ExplicitlySetBmcModel

The fine-tuning method and hyperparameters used for fine-tuning a custom model.
Note: Objects should always be created or deserialized using the Builder. This model distinguishes fields that are null because they are unset from fields that are explicitly set to null. This is done in the setter methods of the Builder, which maintain a set of all explicitly set fields called Builder#__explicitlySet__. The hashCode() and equals(Object) methods are implemented to take the explicitly set fields into account. The constructor, on the other hand, does not take the explicitly set fields into account (since the constructor cannot distinguish explicit null from unset null).

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	TrainingConfig.TrainingConfigType The fine-tuning method for training a custom model.

Field Summary

Fields inherited from class com.oracle.bmc.http.client.internal.ExplicitlySetBmcModel

EXPLICITLY_SET_FILTER_NAME, EXPLICITLY_SET_PROPERTY_NAME

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	TrainingConfig(Integer totalTrainingEpochs, Double learningRate, Integer trainingBatchSize, Integer earlyStoppingPatience, Double earlyStoppingThreshold, Integer logModelMetricsIntervalInSteps) Deprecated.

Method Summary

Modifier and Type	Method and Description
boolean	equals(Object o)
Integer	getEarlyStoppingPatience() Stop training if the loss metric does not improve beyond ‘early_stopping_threshold’ for this many times of evaluation.
Double	getEarlyStoppingThreshold() How much the loss must improve to prevent early stopping.
Double	getLearningRate() The initial learning rate to be used during training
Integer	getLogModelMetricsIntervalInSteps() Determines how frequently to log model metrics.
Integer	getTotalTrainingEpochs() The maximum number of training epochs to run for.
Integer	getTrainingBatchSize() The batch size used during training.
int	hashCode()
String	toString()
String	toString(boolean includeByteArrayContents) Return a string representation of the object.

Methods inherited from class com.oracle.bmc.http.client.internal.ExplicitlySetBmcModel

markPropertyAsExplicitlySet, wasPropertyExplicitlySet

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

TrainingConfig

@Deprecated
 @ConstructorProperties(value={"totalTrainingEpochs","learningRate","trainingBatchSize","earlyStoppingPatience","earlyStoppingThreshold","logModelMetricsIntervalInSteps"})
protected TrainingConfig(Integer totalTrainingEpochs,
                                                                                                                                                                                                                 Double learningRate,
                                                                                                                                                                                                                 Integer trainingBatchSize,
                                                                                                                                                                                                                 Integer earlyStoppingPatience,
                                                                                                                                                                                                                 Double earlyStoppingThreshold,
                                                                                                                                                                                                                 Integer logModelMetricsIntervalInSteps)

Deprecated.

Method Detail

getTotalTrainingEpochs

public Integer getTotalTrainingEpochs()

The maximum number of training epochs to run for.

Returns:

the value

getLearningRate

public Double getLearningRate()

The initial learning rate to be used during training

Returns:

the value

getTrainingBatchSize

public Integer getTrainingBatchSize()

The batch size used during training.

Returns:

the value

getEarlyStoppingPatience

public Integer getEarlyStoppingPatience()

Stop training if the loss metric does not improve beyond ‘early_stopping_threshold’ for this many times of evaluation.

Returns:

the value

getEarlyStoppingThreshold

public Double getEarlyStoppingThreshold()

How much the loss must improve to prevent early stopping.

Returns:

the value

getLogModelMetricsIntervalInSteps

public Integer getLogModelMetricsIntervalInSteps()

Determines how frequently to log model metrics.

Every step is logged for the first 20 steps and then follows this parameter for log frequency. Set to 0 to disable logging the model metrics.

Returns:

the value

toString

public String toString()

Overrides:

toString in class com.oracle.bmc.http.client.internal.ExplicitlySetBmcModel

toString

public String toString(boolean includeByteArrayContents)

Return a string representation of the object.

Parameters:

includeByteArrayContents - true to include the full contents of byte arrays

Returns:

string representation

equals

public boolean equals(Object o)

Overrides:

equals in class com.oracle.bmc.http.client.internal.ExplicitlySetBmcModel

hashCode

public int hashCode()

Overrides:

hashCode in class com.oracle.bmc.http.client.internal.ExplicitlySetBmcModel