timeseries_classification_data

#TimeSeriesClassificationData

##Description The TimeSeriesClassificationData is the main data structure for recording, labeling, managing, saving, and loading training data for supervised temporal learning problems. Unlike the ClassificationData, in which each sample consists of 1 N dimensional datum, a TimeSeriesClassificationData sample will consist of an N dimensional time series of length M. The length of each time series sample (i.e. M) can be different for each datum in the dataset.

Note, as of GRT revision 347, this replaces the LabelledTimeSeriesClassificationData data structure, however a typedef enables you to still use the LabelledTimeSeriesClassificationData in your old code if needed.

##Suitable Classification Algorithms The TimeSeriesClassificationData data structure should be used to train the following GRT classification algorithms:

• Dynamic Time Warping (DTW)
• Discrete Hidden Markov Model (HMM)
• Continuous Hidden Markov Model (HMM)

##Things To Know There are two key things you should know about the TimeSeriesClassificationData class.

• You should set the number of input dimensions of your dataset before you try and add samples to the training dataset (see the example below for how to do this).
• You can not use the class label of 0 when you add a new sample to your dataset. This is because the class label of 0 is reserved for the special null gesture class. This tutorial provides more information about the null gesture class.

##Example Code TimeSeriesClassificationData