AutoVARIMA#

class tslearn.forecasting.AutoVARIMA(max_p=5, max_d=2, max_q=5, default_d_for_non_stationarity=0, seasonal_period=None, max_iter=50, verbose=0)[source]#

Automatically selects the best Vector Autoregressive Integrated Moving Average (VARIMA) model through Hyndman-Khandakar algorithm [1] [2].

Parameters:
max_pint (default: 5)

Maximum order of the AutoRegressive (AR) component to consider.

max_dint (default: 2)

Maximum order of differencing considered to achieve stationarity.

max_qint (default: 5)

Maximum order of the Moving-Average (MA) component to consider.

default_d_for_non_stationarityint or None (default 0)

Used as differentiation order if stationarity cannot be achieved within max_d. If None, an error is raised if stationarity cannot be achieved within max_d.

seasonal_period: int or None (default: None)

Naïve seasonal integration to apply to VARIMA models

max_iterint (default: 50)

The maximum number of iterations to apply to VARIMA models fitting.

verboseint (default 0)

When set to a positive integer, displays logs of the tested VARIMA models selection process. If set to 2 or more, also propagates verbosity to the VARIMA models.

Attributes:
best_estimator_: VARIMA

the fitted VARIMA model.

See also

VARIMA

Vector AutoRegressive Integrated Moving Average (VARIMA) estimator.

Notes

This estimator supports variable length time-series

References

[1]

R. J. Hyndman and G. Athanasopoulos, Forecasting: Principles and Practice. OTexts, 2014. https://otexts.com/fpp3/

[2]

Hyndman, R. J., & Khandakar, Y. (2008). Automatic time series forecasting: The forecast package for R. Journal of Statistical Software, 27(1), 1–22.

Methods

fit(X[, y])

Selects the best Vector Autoregressive Moving Average (VARIMA) model through Hyndman-Khandakar algorithm for the given data.

fit_predict(X[, y, n])

Selects the best Vector Autoregressive Moving Average (VARIMA) model and forecasts n timestamps for the given data.

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_metadata_routing()

Get metadata routing of this object.

get_params([deep])

Get parameters for this estimator.

predict([X, n])

Forecast with the model selected at fitting time.

set_params(**params)

Set the parameters of this estimator.

set_predict_request(*[, n])

Configure whether metadata should be requested to be passed to the predict method.

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.
fit(X, y=None)[source]#

Selects the best Vector Autoregressive Moving Average (VARIMA) model through Hyndman-Khandakar algorithm for the given data.

Parameters:
X: array-like, shape (n_ts, sz, d)

Time-series dataset.

yIgnored
Returns:
self

the fitted estimator

fit_predict(X, y=None, n=1)[source]#

Selects the best Vector Autoregressive Moving Average (VARIMA) model and forecasts n timestamps for the given data.

Parameters:
X: array-like, shape (n_ts, sz, d)

Time-series dataset.

yIgnored
nint (default: 1)

The number of timestamps to forecast, a.k.a. the horizon.

Returns:
array, shape = (n_ts, n, d)

Array of forecasted timestamps

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_metadata_routing()#

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns:
routingMetadataRequest

A MetadataRequest encapsulating routing information.

get_params(deep=True)#

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.

predict(X=None, n=1)[source]#

Forecast with the model selected at fitting time.

Parameters:
Xarray-like, shape (n_ts, sz, d), optional

Time-series dataset to forecast. If None, the fitted data is forecasted otherwise the fitted model is applied to the given data.

nint (default: 1)

The number of timestamps to forecast, a.k.a. the horizon.

Returns:
array, shape = (n_ts, n, d)

Array of forecasted timestamps

set_params(**params)#

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.

set_predict_request(*, n: bool | None | str = '$UNCHANGED$') → AutoVARIMA#

Configure whether metadata should be requested to be passed to the predict method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

  • True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

  • False: metadata is not requested and the meta-estimator will not pass it to predict.

  • None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

  • str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:
nstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for n parameter in predict.

Returns:
selfobject

The updated object.

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.

Examples using tslearn.forecasting.AutoVARIMA#

VARIMA

VARIMA