sandy4321/tsml

Java time series machine learning tools in a Weka compatible toolkit

UEA Time Series Classification

Find more info on our website.

A Weka compatible Java toolbox for time series classification, clustering and transformation. Eventually, we would like to support:

Installation

We are looking at getting this on Maven. For now there are two options:

• download the Jar File

* download the source file and include in a project in your favourite IDE you can then construct your own experiment (see BasicExamples.java) or the experimental structure we use (see Experiments.java)

Classifiers

We have implemented the following bespoke classifiers for univariate, equal length time series classification

Distance Based

• DD_DTW
• DTD_C
• ElasticEnsemble
• NN_CID
• SAX_1NN
• SAXVSM
• ProximityForest

Dictionary Based

• BOSS
• BOP
• WEASEL

Spectral Based

• RISE
• CRISE

Shaplet Based

• LearnShapelets
• ShapeletTransformClassifier
• FastShapelets

(to do: recover original ShapeletTree)

Interval Based

• TSF
• TSBF
• LPS

Ensembles

• FlatCote
• HiveCote

We have implemented the following bespoke classifiers for multivariate, equal length time series classification

• NN_ED_D
• NN_ED_I
• NN_DTW_D
• NN_DTW_I
• NN_DTW_A
• MultivariateShapeletTransformClassifier
• ConcatenateClassifier

Clusterers

Currently quite limited. Standard approach would be to perform an unsupervised

• UnsupervisedShapelets

Filters/Transformations ------------SimpleBatchFilters that take an Instances (the set of time series), transforms them and returns a new Instances object

• ACF
• ACF_PACF
• ARMA
• BagOfPatternsFilter
• BinaryTransform
• Clipping
• Correlation
• Cosine
• DerivativeFilter
• Differences
• FFT
• Hilbert
• MatrixProfile
• NormalizeAttribute
• NormalizeCase
• PAA
• PACF
• PowerCepstrum
• PowerSepstrum
• RankOrder
• RunLength
• SAX
• Sine
• SummaryStats