FinDiffOptions<T>
Options & Configuration · AiDotNet.Models.Options
Configuration options for FinDiff, a diffusion model specialized for generating realistic financial tabular data with temporal correlation preservation.
Inheritance: ModelOptions → NeuralNetworkOptions → ModelHyperparameterOptions → FinancialNeuralNetworkOptions → RiskModelOptions<T> → SyntheticDataGeneratorOptions<T> → FinDiffOptions<T>
For Beginners
FinDiff generates realistic financial data (stocks, portfolios, etc.) by understanding that financial data has special properties:
- Values change gradually over time (temporal correlation)
- Some values must always be positive (stock prices)
- Volatility clusters (periods of high/low market turbulence)
Example:
How It Works
FinDiff extends standard diffusion models with financial domain knowledge:
- Temporal correlation loss: Preserves autocorrelation and cross-correlation
- Volatility-aware noise: Adapts noise schedule to financial volatility patterns
- Financial constraints: Enforces domain rules (positive prices, valid ranges)
Reference: “Diffusion Models for Financial Tabular Data” (2024)
Constructors
| Constructor | Summary |
|---|---|
FinDiffOptions() |
Properties
| Property | Summary |
|---|---|
bool EnforcePositive { get; set; } |
Gets or sets whether to enforce positive values in generated data. |
double BetaEnd { get; set; } |
Gets or sets the maximum beta for the noise schedule. |
double BetaStart { get; set; } |
Gets or sets the minimum beta for the noise schedule. |
double LearningRate { get; set; } |
Gets or sets the learning rate. |
double TemporalWeight { get; set; } |
Gets or sets the weight for the temporal correlation loss. |
int BatchSize { get; set; } |
Gets or sets the training batch size. |
int Epochs { get; set; } |
Gets or sets the number of training epochs. |
int NumTimesteps { get; set; } |
Gets or sets the number of diffusion timesteps. |
int TimestepEmbeddingDimension { get; set; } |
Gets or sets the timestep embedding dimension. |
int VGMModes { get; set; } |
Gets or sets the number of VGM modes for continuous column transformation. |
int[] MLPDimensions { get; set; } |
Gets or sets the hidden layer dimensions for the denoiser MLP. |