TimeBridgeOptions<T>
Options & Configuration · AiDotNet.Models.Options
Configuration options for TimeBridge (Non-Stationarity Matters for Time Series Foundation Models).
Inheritance: ModelOptions → RegressionOptions<T> → TimeSeriesRegressionOptions<T> → TimeBridgeOptions<T>
How It Works
TimeBridge addresses the critical non-stationarity gap in time series foundation models. It introduces a bridge mechanism that preserves and restores non-stationary information (trends, level shifts) that is typically lost during standard normalization.
Reference: “TimeBridge: Non-Stationarity Matters for Long-term Time Series Forecasting”, 2024.
Constructors
| Constructor | Summary |
|---|---|
TimeBridgeOptions() |
Initializes a new instance with default values. |
TimeBridgeOptions(TimeBridgeOptions<T> other) |
Creates a copy of another options instance. |
Properties
| Property | Summary |
|---|---|
FoundationModelSize ModelSize { get; set; } |
Preset size of the foundation model. |
bool UseStationarityGating { get; set; } |
Gets or sets whether to use stationarity gating for adaptive restoration. |
double DropoutRate { get; set; } |
Dropout probability applied during training. |
double LearningRate { get; set; } |
Gets or sets the Adam learning rate used for native training. |
int BridgeDimension { get; set; } |
Gets or sets the dimension of the non-stationarity bridge module. |
int ContextLength { get; set; } |
Number of past time steps the model reads as context. |
int ForecastHorizon { get; set; } |
Number of future time steps to forecast. |
int HiddenDimension { get; set; } |
Width of the model’s hidden representations. |
int IntermediateSize { get; set; } |
Width of the feed-forward layer inside each transformer block. |
int NumHeads { get; set; } |
Number of attention heads. |
int NumLayers { get; set; } |
Number of transformer layers. |
int PatchLength { get; set; } |
Number of time steps in each patch the series is split into. |