Class SyntheticTabularFederatedBenchmarkOptions

Namespace

AiDotNet.Configuration

Assembly

AiDotNet.dll

Configuration options for the synthetic federated tabular benchmark suite.

public sealed class SyntheticTabularFederatedBenchmarkOptions

Inheritance

object

SyntheticTabularFederatedBenchmarkOptions

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

This suite generates a deterministic synthetic dataset with non-IID client distributions, allowing benchmark runs without external dataset dependencies.

For Beginners: Synthetic means AiDotNet generates the data automatically instead of reading files from disk. This is useful for CI and quick sanity checks.

Properties

ClassCount

Gets or sets the number of classes for classification tasks (null uses defaults).

public int? ClassCount { get; set; }

Property Value

int?

ClientCount

Gets or sets the number of federated clients to simulate (null uses defaults).

public int? ClientCount { get; set; }

Property Value

int?

DirichletAlpha

Gets or sets the Dirichlet concentration parameter controlling label skew across clients.

public double DirichletAlpha { get; set; }

Property Value

double

Remarks

Smaller values create more heterogeneous (more non-IID) client label distributions.

FeatureCount

Gets or sets the number of input features per sample (null uses defaults).

public int? FeatureCount { get; set; }

Property Value

int?

NoiseStdDev

Gets or sets the noise standard deviation applied to generated targets/scores.

public double NoiseStdDev { get; set; }

Property Value

double

TaskType

Gets or sets the synthetic task type.

public SyntheticTabularTaskType TaskType { get; set; }

Property Value

SyntheticTabularTaskType

TestSamplesPerClient

Gets or sets the number of test samples per client (null uses defaults).

public int? TestSamplesPerClient { get; set; }

Property Value

int?

TrainSamplesPerClient

Gets or sets the number of training samples per client (null uses defaults).

public int? TrainSamplesPerClient { get; set; }

Property Value

int?