Feature Augmentation of Classifiers Using Learning Time Series Shapelets Transformation for Night Setback Classification of District Heating Substations

Algorithm 1

Learning time series shapelets.

(1)	Input: A time series dataset , where is the number of training instance and is the number of the order in each series. Number of shapelets, length of a shapelet , objective function , learning rate , number of iterations maxIter. Shapelets , classification weights , bias
(2)	for iteration = do
(3)	for i = 1, …, I do
(4)	for k = 1, …, K do
(5)
(6)	for L = 1, …, L do
(7)
(8)	end for
(9)	end for
(10)
(11)	end for
(12)	end for
(13)	Return,