SSA Spike DetectionΒΆ

The documentation is generated based on the sources available at dotnet/machinelearning and released under MIT License.

Type: datatransform Aliases: SsaSpikeDetector, spike Namespace: Microsoft.ML.Runtime.Data Assembly: Microsoft.ML.TimeSeries.dll Microsoft Documentation: SSA Spike Detection

Description

This transform detects the spikes in a seasonal time-series using Singular Spectrum Analysis (SSA).

Parameters

Name Short name Default Description
confidence cnf   The confidence for spike detection in the range [0, 100].
errorFunction err SignedDifference The function used to compute the error between the expected and the observed value.
name     The name of the new column.
pvalueHistoryLength wnd 100 The size of the sliding window for computing the p-value.
seasonalWindowSize swnd   An upper bound on the largest relevant seasonality in the input time-series.
side   TwoSided The argument that determines whether to detect positive or negative anomalies, or both.
source src   The name of the source column.
trainingWindowSize twnd   The number of points from the beginning of the sequence used for training.