TOTOOptions<T>

Options & Configuration · AiDotNet.Models.Options

Configuration options for TOTO (Datadog’s Time Series Foundation Model for Observability).

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

How It Works

TOTO is Datadog’s domain-specific time series foundation model optimized for IT operations, infrastructure monitoring, and observability. Pre-trained on 1 trillion data points from the Datadog observability platform, it excels at SRE metrics and anomaly detection.

Reference: Datadog, “Introducing Toto: A state-of-the-art time series foundation model”, 2025.

Constructors

Constructor Summary
TOTOOptions() Creates options with the default values.
TOTOOptions(TOTOOptions<T> other) Creates a copy of another options instance.

Properties

Property Summary
FoundationModelSize ModelSize { get; set; } Gets or sets the public size label for this configuration.
double Beta1 { get; set; } Gets or sets AdamW’s first moment coefficient.
double Beta2 { get; set; } Gets or sets AdamW’s second moment coefficient.
double DropoutRate { get; set; } Gets or sets the dropout probability.
double LearningRate { get; set; } Gets or sets AdamW’s peak learning rate.
double WeightDecay { get; set; } Gets or sets AdamW’s decoupled weight decay.
int ContextLength { get; set; } Gets or sets the maximum historical context used for forecasting.
int ForecastHorizon { get; set; } Gets or sets the number of future samples produced by the native forecast head.
int HiddenDimension { get; set; } Gets or sets the transformer embedding dimension.
int IntermediateSize { get; set; } Gets or sets the transformer MLP dimension.
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 PatchLength { get; set; } Gets or sets the causal patch size.
int TotalTrainingSteps { get; set; } Gets or sets the total number of pretraining scheduler steps.
int WarmupSteps { get; set; } Gets or sets the number of linear learning-rate warmup steps.