tslearn.piecewise.SymbolicAggregateApproximation

class tslearn.piecewise.SymbolicAggregateApproximation(n_segments=1, alphabet_size_avg=5, scale=False)[source]

Symbolic Aggregate approXimation (SAX) transformation.

SAX was originally presented in [1].

Parameters
n_segmentsint (default: 1)

Number of PAA segments to compute

alphabet_size_avgint (default: 5)

Number of SAX symbols to use

scale: bool (default: False)

Whether input data should be scaled for each feature to have zero mean and unit variance across the dataset passed at fit time. Default for this parameter is set to False in version 0.4 to ensure backward compatibility, but is likely to change in a future version.

Attributes
breakpoints_avg_numpy.ndarray of shape (alphabet_size - 1, )

List of breakpoints used to generate SAX symbols

Notes

This method requires a dataset of equal-sized time series.

References

1

J. Lin, E. Keogh, L. Wei, et al. Experiencing SAX: a novel symbolic representation of time series. Data Mining and Knowledge Discovery, 2007. vol. 15(107)

Examples

>>> sax = SymbolicAggregateApproximation(n_segments=3, alphabet_size_avg=2)
>>> data = [[-1., 2., 0.1, -1., 1., -1.], [1., 3.2, -1., -3., 1., -1.]]
>>> sax_data = sax.fit_transform(data)
>>> sax_data.shape
(2, 3, 1)
>>> sax_data
array([[[1],
        [0],
        [1]],

       [[1],
        [0],
        [1]]])
>>> sax.distance_sax(sax_data[0], sax_data[1])  
0.0
>>> sax.distance(data[0], data[1])  
0.0
>>> sax.inverse_transform(sax_data)
array([[[ 0.67448975],
        [ 0.67448975],
        [-0.67448975],
        [-0.67448975],
        [ 0.67448975],
        [ 0.67448975]],

       [[ 0.67448975],
        [ 0.67448975],
        [-0.67448975],
        [-0.67448975],
        [ 0.67448975],
        [ 0.67448975]]])

Methods

distance(ts1, ts2)

Compute distance between SAX representations as defined in [1].

distance_paa(paa1, paa2)

Compute distance between PAA representations as defined in [1].

distance_sax(sax1, sax2)

Compute distance between SAX representations as defined in [1].

fit(X[, y])

Fit a SAX representation.

fit_transform(X[, y])

Fit a SAX representation and transform the data accordingly.

from_hdf5(path)

Load model from a HDF5 file.

from_json(path)

Load model from a JSON file.

from_pickle(path)

Load model from a pickle file.

get_params([deep])

Get parameters for this estimator.

inverse_transform(X)

Compute time series corresponding to given SAX representations.

set_params(**params)

Set the parameters of this estimator.

to_hdf5(path)

Save model to a HDF5 file.

to_json(path)

Save model to a JSON file.

to_pickle(path)

Save model to a pickle file.

transform(X[, y])

Transform a dataset of time series into its SAX representation.

distance(ts1, ts2)[source]

Compute distance between SAX representations as defined in [1].

Parameters
ts1array-like

A time series

ts2array-like

Another time series

Returns
float

SAX distance

References

1(1,2)

J. Lin, E. Keogh, L. Wei, et al. Experiencing SAX: a novel symbolic representation of time series. Data Mining and Knowledge Discovery, 2007. vol. 15(107)

distance_paa(paa1, paa2)[source]

Compute distance between PAA representations as defined in [1].

Parameters
paa1array-like

PAA representation of a time series

paa2array-like

PAA representation of another time series

Returns
float

PAA distance

References

1(1,2)

E. Keogh & M. Pazzani. Scaling up dynamic time warping for datamining applications. SIGKDD 2000, pp. 285–289.

distance_sax(sax1, sax2)[source]

Compute distance between SAX representations as defined in [1].

Parameters
sax1array-like

SAX representation of a time series

sax2array-like

SAX representation of another time series

Returns
float

SAX distance

References

1(1,2)

J. Lin, E. Keogh, L. Wei, et al. Experiencing SAX: a novel symbolic representation of time series. Data Mining and Knowledge Discovery, 2007. vol. 15(107)

fit(X, y=None)[source]

Fit a SAX representation.

Parameters
Xarray-like of shape (n_ts, sz, d)

Time series dataset

Returns
SymbolicAggregateApproximation

self

fit_transform(X, y=None, **fit_params)[source]

Fit a SAX representation and transform the data accordingly.

Parameters
Xarray-like of shape (n_ts, sz, d)

Time series dataset

Returns
numpy.ndarray of integers with shape (n_ts, n_segments, d)

SAX-Transformed dataset

classmethod from_hdf5(path)[source]

Load model from a HDF5 file. Requires h5py http://docs.h5py.org/

Parameters
pathstr

Full path to file.

Returns
Model instance
classmethod from_json(path)[source]

Load model from a JSON file.

Parameters
pathstr

Full path to file.

Returns
Model instance
classmethod from_pickle(path)[source]

Load model from a pickle file.

Parameters
pathstr

Full path to file.

Returns
Model instance
get_params(deep=True)[source]

Get parameters for this estimator.

Parameters
deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns
paramsdict

Parameter names mapped to their values.

inverse_transform(X)[source]

Compute time series corresponding to given SAX representations.

Parameters
Xarray-like of shape (n_ts, sz_sax, d)

A dataset of SAX series.

Returns
numpy.ndarray of shape (n_ts, sz_original_ts, d)

A dataset of time series corresponding to the provided representation.

set_params(**params)[source]

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters
**paramsdict

Estimator parameters.

Returns
selfestimator instance

Estimator instance.

to_hdf5(path)[source]

Save model to a HDF5 file. Requires h5py http://docs.h5py.org/

Parameters
pathstr

Full file path. File must not already exist.

Raises
FileExistsError

If a file with the same path already exists.

to_json(path)[source]

Save model to a JSON file.

Parameters
pathstr

Full file path.

to_pickle(path)[source]

Save model to a pickle file.

Parameters
pathstr

Full file path.

transform(X, y=None)[source]

Transform a dataset of time series into its SAX representation.

Parameters
Xarray-like of shape (n_ts, sz, d)

Time series dataset

Returns
numpy.ndarray of integers with shape (n_ts, n_segments, d)

SAX-Transformed dataset

Examples using tslearn.piecewise.SymbolicAggregateApproximation