Transformer

PAA

Piecewise Aggregate Approximation Transformer (PAA).

PAA [1] is a dimensionality reduction technique that divides a time series into frames and takes their mean. This implementation offers two variants:

1) the original, which takes the desired number of frames and can set the frame size to a fraction to support cases where the time series cannot be divided into the frames equally. 2) a variant that takes the desired frame size and can decrease the frame size of the last frame to support cases where the time series is not evenly divisible into frames.

Quickstart

python

from sktime.transformations.paa import PAA

estimator = PAA(frames=8, frame_size=0)

Parameters(2)

framesint, optional (default=8, greater equal 1 if frame_size=0): length of transformed time series. Ignored if frame_size is set.
frame_sizeint, optional (default=0, greater equal 0): length of the frames over which the mean is taken. Overrides frames if > 0.

Examples

>>> from numpy import arange
>>> from sktime.transformations.paa import PAA
>>> X = arange (10)
>>> paa = PAA (frames = 3)
>>> paa. fit_transform (X) array([1.2, 4.5, 7.8])
>>> paa = PAA (frame_size = 3) array([1, 4, 7, 9])

References

[1]

Keogh, E., Chakrabarti, K., Pazzani, M., and Mehrotra, S. Dimensionality Reduction for Fast Similarity Search in Large Time Series Databases. Knowledge and Information Systems 3, 263-286 (2001). https://doi.org/10.1007/PL00011669