UniTSOptions<T>

Options & Configuration · AiDotNet.Models.Options

Configuration options for UniTS (Unified Time Series Model).

Inheritance: ModelOptions → RegressionOptions<T> → TimeSeriesRegressionOptions<T> → UniTSOptions<T>

For Beginners

UniTS is designed to be a universal time series model:

The Key Insight: Different time series tasks share common patterns. Instead of training separate models, UniTS learns a unified representation that works for all tasks.

Supported Tasks:

  1. Forecasting: Predict future values
  2. Classification: Categorize entire time series
  3. Anomaly Detection: Identify unusual patterns
  4. Imputation: Fill in missing values

Architecture:

  • Multi-scale temporal convolution for local patterns
  • Transformer layers for global dependencies
  • Task-specific heads for different outputs
  • Shared backbone pretrained on diverse datasets

Advantages:

  • One model for multiple tasks (transfer learning)
  • Strong zero-shot performance on new domains
  • Efficient inference (shared computation)

How It Works

UniTS is a unified architecture for multiple time series tasks including forecasting, classification, anomaly detection, and imputation using a single pretrained model.

Reference: Gao et al., “UniTS: A Unified Multi-Task Time Series Model”, 2024. https://arxiv.org/abs/2403.00131

Constructors

Constructor Summary
UniTSOptions() Initializes a new instance of the UniTSOptions class with default values.
UniTSOptions(UniTSOptions<T> other) Initializes a new instance by copying from another instance.

Properties

Property Summary
double DropoutRate { get; set; } Gets or sets the dropout rate for regularization.
int ContextLength { get; set; } Gets or sets the context length (input sequence length).
int ForecastHorizon { get; set; } Gets or sets the forecast horizon (prediction length).
int HiddenDimension { get; set; } Gets or sets the hidden dimension.
int NumClasses { get; set; } Gets or sets the number of classes for classification task.
int NumHeads { get; set; } Gets or sets the number of attention heads.
int NumLayers { get; set; } Gets or sets the number of transformer layers.
int[] ConvKernelSizes { get; set; } Gets or sets the convolution kernel sizes for multi-scale temporal convolution.
string TaskType { get; set; } Gets or sets the task type.