TS_LabelGroups, which populates the
Groupcolumn of the
TimeSeriestable as a categorical labeling of the data (the class labels the classifier will attempt to predict). These group labels are used by a range of analysis functions, including
TS_LabelGroupsassumes that the class labels you are interested in are contained in the
Keywordscolumn of the
TimeSeriestable (set up from your original input file when
HCTSA.mat(specifying the shorthand
'raw'for this default, un-normalized data), corresponding to those labeled as
'parkinsons'and those labeled as
'healthy'are left unlabeled.
TS_LabelGroupscan automatically detect this by running
TS_LabelGroups, but can be achieved by writing a script that does logical operations on calls to
TS_LabelGroupsand saves back to the
TimeSeries.Groupcolumn. Here is an example of combining two labels (male/female and day/night) on fly movement data.
TS_LabelGroupsfunction, and can be cleared by running, e.g.,
HCTSA.mat), then this labeling will remain with the normalized data (e.g., after running