TSMixerOptions<T>

Options & Configuration · AiDotNet.Models.Options

Configuration options for the TSMixer model.

Inheritance: ModelOptions → TSMixerOptions<T>

For Beginners

These options control how the TSMixer model behaves:

Key settings:

  • NumBlocks: Number of mixer blocks (similar to layers in other models)
  • HiddenDimension: Size of hidden layers in the MLPs
  • FeaturesFirst: Whether to mix features before time (affects performance)
  • UseRevIN: Whether to use reversible instance normalization

TSMixer is simpler than transformer-based models but can be just as effective. It’s faster to train and uses less memory than attention-based approaches.

How It Works

TSMixer is an all-MLP architecture for multivariate time series forecasting that achieves state-of-the-art results using only multilayer perceptrons (MLPs) without attention mechanisms.

Reference: Chen et al., “TSMixer: An All-MLP Architecture for Time Series Forecasting”, TMLR 2023. https://arxiv.org/abs/2303.06053

Constructors

Constructor Summary
TSMixerOptions()

Properties

Property Summary
ILossFunction<T> LossFunction { get; set; } Gets or sets the loss function for training.
bool FeaturesFirst { get; set; } Gets or sets whether to process features before time dimension.
bool UseBatchNorm { get; set; } Gets or sets whether to use batch normalization.
bool UseRevIN { get; set; } Gets or sets whether to use reversible instance normalization (RevIN).
double Dropout { get; set; } Gets or sets the dropout rate.
double FeedForwardExpansion { get; set; } Gets or sets the feedforward expansion factor.
double LearningRate { get; set; } Gets or sets the learning rate.
int HiddenDimension { get; set; } Gets or sets the hidden dimension for MLP layers.
int NumBlocks { get; set; } Gets or sets the number of mixer blocks.
int NumFeatures { get; set; } Gets or sets the number of input features.
int PredictionHorizon { get; set; } Gets or sets the prediction horizon.
int SequenceLength { get; set; } Gets or sets the input sequence length.

Methods

Method Summary
List<string> Validate() Validates the options and returns any validation errors.