Class GeminiTuningHyperparameters

Namespace

Uralstech.UGemini.Models.Tuning

Hyperparameters controlling the tuning process.

public class GeminiTuningHyperparameters

Inheritance

object

GeminiTuningHyperparameters

Extension Methods

GeminiFieldMaskGenerator.GetFieldMask<T>(T)

Fields

BatchSize

The batch size hyperparameter for tuning. If not set, a default of 4 or 16 will be used based on the number of training examples.

public int BatchSize

Field Value

int

EpochCount

The number of training epochs. An epoch is one pass through the training data. If not set, a default of 5 will be used.

public int EpochCount

Field Value

int

LearningRate

The learning rate hyperparameter for tuning. If not set, a default of 0.001 or 0.0002 will be calculated based on the number of training examples.

public float? LearningRate

Field Value

float?

Remarks

If null, LearningRateMultiplier will be provided.

LearningRateMultiplier

The learning rate multiplier is used to calculate a final learningRate based on the default (recommended) value.

public float? LearningRateMultiplier

Field Value

float?

Remarks

Actual learning rate := learningRateMultiplier * default learning rate Default learning rate is dependent on base model and dataset size. If not set, a default of 1.0 will be used.

If null, LearningRate will be provided.