ClientSelectionRequest
Models & Types · AiDotNet.Models
Represents a request to select participating clients for a federated learning round.
How It Works
For Beginners: This object packages up the information needed to decide which clients should join the next training round (who is available, how many to pick, and any optional hints like groups).
Constructors
| Constructor | Summary |
|---|---|
ClientSelectionRequest() |
Properties
| Property | Summary |
|---|---|
IReadOnlyDictionary<int, double> ClientAvailabilityProbabilities { get; set; } |
Gets or sets optional client availability probabilities (0.0 to 1.0). |
IReadOnlyDictionary<int, double> ClientPerformanceScores { get; set; } |
Gets or sets optional per-client performance scores (higher is better). |
IReadOnlyDictionary<int, double> ClientWeights { get; set; } |
Gets or sets client weights (typically proportional to sample count). |
IReadOnlyDictionary<int, double[]> ClientEmbeddings { get; set; } |
Gets or sets optional per-client embeddings for cluster-based selection. |
IReadOnlyDictionary<int, string> ClientGroupKeys { get; set; } |
Gets or sets optional group keys for stratified selection. |
IReadOnlyList<int> CandidateClientIds { get; set; } |
Gets or sets the full set of candidate client IDs. |
Random Random { get; set; } |
Gets or sets the random number generator to use for selection. |
double FractionToSelect { get; set; } |
Gets or sets the fraction of clients to select (0.0 to 1.0). |
int RoundNumber { get; set; } |
Gets or sets the round number (0-based) for which selection is being performed. |