DeepAntConfiguration

Options & Configuration · AiDotNet.TimeSeries.AnomalyDetection

The sizes of DeepAnT’s time-series predictor.

Implements: IEquatable<DeepAntConfiguration>

Constructors

Constructor Summary
DeepAntConfiguration(int Window, int Horizon, int Filters, int KernelSize, int PoolSize, int HiddenUnits, double Dropout) The sizes of DeepAnT’s time-series predictor.

Properties

Property Summary
double Dropout { get; set; } The dropout rate after the fully connected layer.
int Filters { get; set; } The filters of each convolution (the paper’s 32).
int FirstLength { get; } The time steps after the first convolution and pooling.
int HiddenUnits { get; set; } The width of the fully connected layer before the output.
int Horizon { get; set; } The prediction window p_w: the values it predicts.
int KernelSize { get; set; } The width of each convolution (‘valid’, stride 1).
int PoolSize { get; set; } The width and stride of each max-pooling.
int SecondLength { get; } The time steps after the second convolution and pooling.
int Window { get; set; } The history window w: the values the predictor reads.

Methods

Method Summary
DeepAntConfiguration <Clone>$()
bool Equals(DeepAntConfiguration other)
bool Equals(object obj)
bool op_Equality(DeepAntConfiguration left, DeepAntConfiguration right)
bool op_Inequality(DeepAntConfiguration left, DeepAntConfiguration right)
int GetHashCode()
string ToString()
void Deconstruct(int Window, int Horizon, int Filters, int KernelSize, int PoolSize, int HiddenUnits, double Dropout)