PiecewiseAggregateApproximation

class pyts.approximation.PiecewiseAggregateApproximation(window_size: int | float | None = 1, output_size: int | float | None = None, overlapping: bool = True)

Piecewise Aggregate Approximation.

Parameters:

window_sizeint, float or None (default = 1)

Length of the sliding window. If float, it represents a percentage of the size of each time series and must be between 0 and 1.

output_sizeint, float or None (default = None)

Size of the returned time series. If float, it represents a percentage of the size of each time series and must be between 0. and 1. Ignored if window_size is not None. It can’t be None if window_size is None. If you want to use output_size over window_size, you must set window_size=None.

overlappingbool (default = True)

When window_size=None, output_size is used to derive the window size; the window size is fixed if overlapping=True and may vary if overlapping=False. Ignored if window_size is specified.

References

[1]

E. Keogh, K. Chakrabarti, M. Pazzani, and S. Mehrotra, “Dimensionality reduction for fast similarity search in large time series databases”. Knowledge and information Systems, 3(3), 263-286 (2001).

Examples

>>> from pyts.approximation import PiecewiseAggregateApproximation
>>> X = [[0, 4, 2, 1, 7, 6, 3, 5],
...      [2, 5, 4, 5, 3, 4, 2, 3]]
>>> transformer = PiecewiseAggregateApproximation(window_size=2)
>>> transformer.transform(X)
array([[2. , 1.5, 6.5, 4. ],
       [3.5, 4.5, 3.5, 2.5]])

Methods

__init__([window_size, output_size, overlapping])
fit([X, y])	Pass.
fit_transform(X[, y])	Fit to data, then transform it.
get_metadata_routing()	Get metadata routing of this object.
get_params([deep])	Get parameters for this estimator.
set_params(**params)	Set the parameters of this estimator.
transform(X)	Reduce the dimensionality of each time series.

__init__(window_size: int | float | None = 1, output_size: int | float | None = None, overlapping: bool = True) → None

fit(X: ArrayLike | None = None, y: ArrayLike | None = None) → Self

Pass.

Parameters:

X

Ignored

y

Ignored

Returns:

self

object

transform(X: ArrayLike) → NDArray[float64]

Reduce the dimensionality of each time series.

Parameters:

Xarray-like, shape = (n_samples, n_timestamps)

Returns:

X_newarray, shape = (n_samples, n_timestamps_new)

Examples using pyts.approximation.PiecewiseAggregateApproximation

Piecewise Aggregate Approximation

Piecewise Aggregate Approximation