StatisticalDriftDetector<T>
Models & Types · AiDotNet.FederatedLearning.DriftDetection
Statistical drift detector: uses Page-Hinkley, ADWIN, or DDM tests on client metrics.
Inheritance: FederatedLearningComponentBase<T> → StatisticalDriftDetector<T>
Implements: IFederatedDriftDetector<T>
For Beginners
This detector monitors each client’s loss or accuracy over rounds and applies proven statistical tests to detect when the distribution has shifted.
How It Works
Methods:
Constructors
| Constructor | Summary |
|---|---|
StatisticalDriftDetector(FederatedDriftOptions options) |
Initializes a new instance of StatisticalDriftDetector. |
Properties
| Property | Summary |
|---|---|
string MethodName { get; } |
Methods
| Method | Summary |
|---|---|
Dictionary<int, double> GetAdaptiveWeights(Dictionary<int, double> originalWeights, DriftReport driftReport) |
|
DriftReport DetectDrift(int round, Dictionary<int, Tensor<T>> clientModels, Tensor<T> globalModel, Dictionary<int, double> clientMetrics) |
|
void Reset() |