Skip to content

ForecasterAutoregMultiSeriesCustom

ForecasterAutoregMultiSeriesCustom(regressor, fun_predictors, window_size, name_predictors=None, encoding='ordinal_category', transformer_series=StandardScaler(), transformer_exog=None, weight_func=None, series_weights=None, differentiation=None, dropna_from_series=False, fit_kwargs=None, forecaster_id=None)

Bases: ForecasterBase

This class turns any regressor compatible with the scikit-learn API into a recursive autoregressive (multi-step) forecaster for multiple series with a custom function to create predictors.

Parameters:

Name Type Description Default
regressor regressor or pipeline compatible with the scikit-learn API

An instance of a regressor or pipeline compatible with the scikit-learn API.

required
fun_predictors Callable

Function that receives a time series as input (numpy ndarray) and returns another numpy ndarray with the predictors. The same function is applied to all series.

required
window_size int

Size of the window needed by fun_predictors to create the predictors.

required
name_predictors list

Name of the predictors returned by fun_predictors. If None, predictors are named using the prefix 'custom_predictor_' where i is the index of the position the predictor has in the returned array of fun_predictors.

`None`
encoding str

Encoding used to identify the different series.

  • If 'ordinal', a single column is created with integer values from 0 to n_series - 1.
  • If 'ordinal_category', a single column is created with integer values from 0 to n_series - 1 and the column is transformed into pandas.category dtype so that it can be used as a categorical variable.
  • If 'onehot', a binary column is created for each series. New in version 0.12.0
`'ordinal_category'`
transformer_series (transformer(preprocessor), dict)

An instance of a transformer (preprocessor) compatible with the scikit-learn preprocessing API with methods: fit, transform, fit_transform and inverse_transform. Transformation is applied to each series before training the forecaster. ColumnTransformers are not allowed since they do not have inverse_transform method.

  • If single transformer: it is cloned and applied to all series.
  • If dict of transformers: a different transformer can be used for each series.
`sklearn.preprocessing.StandardScaler`
transformer_exog transformer

An instance of a transformer (preprocessor) compatible with the scikit-learn preprocessing API. The transformation is applied to exog before training the forecaster. inverse_transform is not available when using ColumnTransformers.

`None`
weight_func (Callable, dict)

Function that defines the individual weights for each sample based on the index. For example, a function that assigns a lower weight to certain dates. Ignored if regressor does not have the argument sample_weight in its fit method. See Notes section for more details on the use of the weights.

  • If single function: it is applied to all series.
  • If dict {'series_column_name' : Callable}: a different function can be used for each series, a weight of 1 is given to all series not present in weight_func.
`None`
series_weights dict

Weights associated with each series {'series_column_name' : float}. It is only applied if the regressor used accepts sample_weight in its fit method. See Notes section for more details on the use of the weights.

  • If a dict is provided, a weight of 1 is given to all series not present in series_weights.
  • If None, all levels have the same weight.
`None`
differentiation int

Order of differencing applied to the time series before training the forecaster. If None, no differencing is applied. The order of differentiation is the number of times the differencing operation is applied to a time series. Differencing involves computing the differences between consecutive data points in the series. Differentiation is reversed in the output of predict() and predict_interval(). WARNING: This argument is newly introduced and requires special attention. It is still experimental and may undergo changes. New in version 0.12.0

`None`
dropna_from_series bool

Determine whether NaN detected in the training matrices will be dropped.

  • If True, drop NaNs in X_train and same rows in y_train.
  • If False, leave NaNs in X_train and warn the user. New in version 0.12.0
`False`
fit_kwargs dict

Additional arguments to be passed to the fit method of the regressor.

`None`
forecaster_id (str, int)

Name used as an identifier of the forecaster.

`None`

Attributes:

Name Type Description
regressor regressor or pipeline compatible with the scikit-learn API

An instance of a regressor or pipeline compatible with the scikit-learn API.

fun_predictors Callable

Function that receives a time series as input (numpy ndarray) and returns another numpy ndarray with the predictors. The same function is applied to all series.

source_code_fun_predictors str

Source code of the custom function used to create the predictors.

window_size int

Size of the window needed by fun_predictors to create the predictors.

window_size_diff int

Size of the window extended by the order of differentiation. When using differentiation, the window_size is increased by the order of differentiation so that the predictors can be created correctly.

name_predictors list

Name of the predictors returned by fun_predictors. If None, predictors are named using the prefix 'custom_predictor_' where i is the index of the position the predictor has in the returned array of fun_predictors.

encoding str

Encoding used to identify the different series.

  • If 'ordinal', a single column is created with integer values from 0 to n_series - 1.
  • If 'ordinal_category', a single column is created with integer values from 0 to n_series - 1 and the column is transformed into pandas.category dtype so that it can be used as a categorical variable.
  • If 'onehot', a binary column is created for each series. New in version 0.12.0
encoder preprocessing

Scikit-learn preprocessing encoder used to encode the series. New in version 0.12.0

encoding_mapping dict

Mapping of the encoding used to identify the different series.

transformer_series (transformer(preprocessor), dict)

An instance of a transformer (preprocessor) compatible with the scikit-learn preprocessing API with methods: fit, transform, fit_transform and inverse_transform. Transformation is applied to each series before training the forecaster. ColumnTransformers are not allowed since they do not have inverse_transform method.

  • If single transformer: it is cloned and applied to all series.
  • If dict of transformers: a different transformer can be used for each series.
transformer_series_ dict

Dictionary with the transformer for each series. It is created cloning the objects in transformer_series and is used internally to avoid overwriting.

transformer_exog transformer(preprocessor)

An instance of a transformer (preprocessor) compatible with the scikit-learn preprocessing API. The transformation is applied to exog before training the forecaster. inverse_transform is not available when using ColumnTransformers.

weight_func (Callable, dict)

Function that defines the individual weights for each sample based on the index. For example, a function that assigns a lower weight to certain dates. Ignored if regressor does not have the argument sample_weight in its fit method. See Notes section for more details on the use of the weights.

  • If single function: it is applied to all series.
  • If dict {'series_column_name' : Callable}: a different function can be used for each series, a weight of 1 is given to all series not present in weight_func.
weight_func_ dict

Dictionary with the weight_func for each series. It is created cloning the objects in weight_func and is used internally to avoid overwriting.

source_code_weight_func (str, dict)

Source code of the custom function(s) used to create weights.

series_weights dict

Weights associated with each series {'series_column_name' : float}. It is only applied if the regressor used accepts sample_weight in its fit method. See Notes section for more details on the use of the weights.

  • If a dict is provided, a weight of 1 is given to all series not present in series_weights.
  • If None, all levels have the same weight.
series_weights_ dict

Weights associated with each series.It is created as a clone of series_weights and is used internally to avoid overwriting.

differentiation int

Order of differencing applied to the time series before training the forecaster.

differentiator TimeSeriesDifferentiator

Skforecast object used to differentiate the time series.

differentiator_ dict

Dictionary with the differentiator for each series. It is created cloning the objects in differentiator and is used internally to avoid overwriting.

dropna_from_series bool

Determine whether NaN detected in the training matrices will be dropped.

window_size int

Size of the window needed by fun_predictors to create the predictors.

last_window dict

Last window of training data for each series. It stores the values needed to predict the next step immediately after the training data.

index_type type

Type of index of the input used in training.

index_freq str

Frequency of Index of the input used in training.

training_range dict

First and last values of index of the data used during training for each series.

series_col_names list

Names of the series (levels) provided by the user during training.

series_X_train list

Names of the series (levels) included in the matrix X_train created internally for training. It can be different from series_col_names if some series are dropped during the training process because of NaNs or because they are not present in the training period.

X_train_col_names list

Names of columns of the matrix created internally for training.

included_exog bool

If the forecaster has been trained using exogenous variable/s.

exog_type type

Type of exogenous variable/s used in training.

exog_dtypes dict

Type of each exogenous variable/s used in training. If transformer_exog is used, the dtypes are calculated before the transformation.

exog_col_names list

Names of the exogenous variables used during training.

fit_kwargs dict

Additional arguments to be passed to the fit method of the regressor.

in_sample_residuals dict

Residuals of the model when predicting training data. Only stored up to 1000 values in the form {level: residuals}. If transformer_series is not None, residuals are stored in the transformed scale.

out_sample_residuals dict

Residuals of the model when predicting non-training data. Only stored up to 1000 values in the form {level: residuals}. If transformer_series is not None, residuals are assumed to be in the transformed scale. Use set_out_sample_residuals() method to set values.

fitted bool

Tag to identify if the regressor has been fitted (trained).

creation_date str

Date of creation.

fit_date str

Date of last fit.

skforecast_version str

Version of skforecast library used to create the forecaster.

python_version str

Version of python used to create the forecaster.

forecaster_id (str, int)

Name used as an identifier of the forecaster.

Notes

The weights are used to control the influence that each observation has on the training of the model. ForecasterAutoregMultiseries accepts two types of weights. If the two types of weights are indicated, they are multiplied to create the final weights. The resulting sample_weight cannot have negative values.

  • series_weights : controls the relative importance of each series. If a series has twice as much weight as the others, the observations of that series influence the training twice as much. The higher the weight of a series relative to the others, the more the model will focus on trying to learn that series.
  • weight_func : controls the relative importance of each observation according to its index value. For example, a function that assigns a lower weight to certain dates.
Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
def __init__(
    self,
    regressor: object,
    fun_predictors: Callable, 
    window_size: int,
    name_predictors: Optional[list]=None,
    encoding: str='ordinal_category',
    transformer_series: Optional[Union[object, dict]]=StandardScaler(),
    transformer_exog: Optional[object]=None,
    weight_func: Optional[Union[Callable, dict]]=None,
    series_weights: Optional[dict]=None,
    differentiation: Optional[int]=None,
    dropna_from_series: bool=False,
    fit_kwargs: Optional[dict]=None,
    forecaster_id: Optional[Union[str, int]]=None
) -> None:

    self.regressor                  = regressor
    self.fun_predictors             = fun_predictors
    self.source_code_fun_predictors = None
    self.window_size                = window_size
    self.window_size_diff           = window_size
    self.name_predictors            = name_predictors
    self.encoding                   = encoding
    self.encoder                    = None
    self.encoding_mapping           = {}
    self.transformer_series         = transformer_series
    self.transformer_series_        = None
    self.transformer_exog           = transformer_exog
    self.weight_func                = weight_func
    self.weight_func_               = None
    self.source_code_weight_func    = None
    self.series_weights             = series_weights
    self.series_weights_            = None
    self.differentiation            = differentiation
    self.differentiator             = None
    self.differentiator_            = None
    self.dropna_from_series         = dropna_from_series
    self.last_window                = None
    self.index_type                 = None
    self.index_freq                 = None
    self.training_range             = None
    self.series_col_names           = None
    self.series_X_train             = None
    self.X_train_col_names          = None
    self.included_exog              = False
    self.exog_type                  = None
    self.exog_dtypes                = None
    self.exog_col_names             = None
    self.in_sample_residuals        = None
    self.out_sample_residuals       = None
    self.fitted                     = False
    self.creation_date              = pd.Timestamp.today().strftime('%Y-%m-%d %H:%M:%S')
    self.fit_date                   = None
    self.skforecast_version         = skforecast.__version__
    self.python_version             = sys.version.split(" ")[0]
    self.forecaster_id              = forecaster_id

    if not isinstance(window_size, int):
        raise TypeError(
            f"Argument `window_size` must be an int. Got {type(window_size)}."
        )

    if not isinstance(fun_predictors, Callable):
        raise TypeError(
            f"Argument `fun_predictors` must be a Callable. Got {type(fun_predictors)}."
        )

    self.source_code_fun_predictors = inspect.getsource(fun_predictors)

    self.weight_func, self.source_code_weight_func, self.series_weights = initialize_weights(
        forecaster_name = type(self).__name__, 
        regressor       = regressor, 
        weight_func     = weight_func, 
        series_weights  = series_weights
    )

    if self.differentiation is not None:
        if not isinstance(differentiation, int) or differentiation < 1:
            raise ValueError(
                (f"Argument `differentiation` must be an integer equal to or "
                 f"greater than 1. Got {differentiation}.")
            )
        self.window_size_diff += self.differentiation
        self.differentiator = TimeSeriesDifferentiator(order=self.differentiation)

    self.fit_kwargs = check_select_fit_kwargs(
                          regressor  = regressor,
                          fit_kwargs = fit_kwargs
                      )

    if self.encoding not in ['ordinal', 'ordinal_category', 'onehot']:
        raise ValueError(
            (f"Argument `encoding` must be one of the following values: 'ordinal', "
             f"'ordinal_category', 'onehot'. Got '{self.encoding}'.")
        )

    if self.encoding == 'onehot':
        self.encoder = OneHotEncoder(
                           categories    = 'auto',
                           sparse_output = False,
                           drop          = None,
                           dtype         = int
                       ).set_output(transform='pandas')
    else:
        self.encoder = OrdinalEncoder(
                           categories = 'auto',
                           dtype      = int
                       ).set_output(transform='pandas')

_create_train_X_y_single_series(y, ignore_exog, exog=None)

Create training matrices from univariate time series and exogenous variables. This method does not transform the exog variables.

Parameters:

Name Type Description Default
y pandas Series

Training time series.

required
ignore_exog bool

If True, exog is ignored.

required
exog pandas DataFrame

Exogenous variable/s included as predictor/s.

`None`

Returns:

Name Type Description
X_train_predictors pandas DataFrame

Training values of custom predictors. Shape: (len(y) - self.window_size_diff, )

X_train_exog pandas DataFrame

Training values of exogenous variables. Shape: (len(y) - self.window_size_diff, len(exog.columns))

y_train pandas Series

Values (target) of the time series related to each row of X_train. Shape: (len(y) - self.window_size_diff, )

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
def _create_train_X_y_single_series(
    self,
    y: pd.Series,
    ignore_exog: bool,
    exog: Optional[pd.DataFrame]=None
) -> Tuple[pd.DataFrame, pd.DataFrame, pd.Series]:
    """
    Create training matrices from univariate time series and exogenous
    variables. This method does not transform the exog variables.

    Parameters
    ----------
    y : pandas Series
        Training time series.
    ignore_exog : bool
        If `True`, `exog` is ignored.
    exog : pandas DataFrame, default `None`
        Exogenous variable/s included as predictor/s.

    Returns
    -------
    X_train_predictors : pandas DataFrame
        Training values of custom predictors.
        Shape: (len(y) - self.window_size_diff, )
    X_train_exog : pandas DataFrame
        Training values of exogenous variables.
        Shape: (len(y) - self.window_size_diff, len(exog.columns))
    y_train : pandas Series
        Values (target) of the time series related to each row of `X_train`.
        Shape: (len(y) - self.window_size_diff, )

    """

    series_name = y.name
    fit_transformer = False if self.fitted else True
    y = transform_series(
            series            = y,
            transformer       = self.transformer_series_[series_name],
            fit               = fit_transformer,
            inverse_transform = False
        )

    y_values = y.to_numpy()
    y_index = y.index

    if self.differentiation is not None:
        if not self.fitted:
            y_values = self.differentiator_[series_name].fit_transform(y_values)
        else:
            differentiator = clone(self.differentiator_[series_name])
            y_values = differentiator.fit_transform(y_values)

    X_train_values  = []
    y_train_values  = []
    for i in range(len(y_values) - self.window_size):

        temp_X_index = np.arange(i, self.window_size + i)
        temp_y_index  = self.window_size + i

        X_train_values.append(self.fun_predictors(y=y_values[temp_X_index]))
        y_train_values.append(y_values[temp_y_index])

    X_train_values = np.vstack(X_train_values)
    y_train_values = np.array(y_train_values)

    expected = self.fun_predictors(y_values[:-1])
    observed = X_train_values[-1, :]
    if expected.shape != observed.shape or not np.allclose(expected, observed, equal_nan=True):
        raise ValueError(
            (f"The `window_size` argument ({self.window_size}), declared when "
             f"initializing the forecaster, does not correspond to the window "
             f"used by `{self.fun_predictors.__name__}`.")
        )

    if self.name_predictors is None:
        X_train_predictors_names = [
            f"custom_predictor_{i}" for i in range(X_train_values.shape[1])
        ]
    else:
        if len(self.name_predictors) != X_train_values.shape[1]:
            raise ValueError(
                (f"The length of provided predictors names "
                 f"(`name_predictors`) do not match the number of columns "
                 f"created by `{self.fun_predictors.__name__}`.")
            )
        X_train_predictors_names = self.name_predictors.copy()

    X_train_predictors = pd.DataFrame(
                             data    = X_train_values,
                             columns = X_train_predictors_names,
                             index   = y_index[self.window_size: ]
                         )
    X_train_predictors['_level_skforecast'] = series_name

    if ignore_exog:
        X_train_exog = None
    else:
        if exog is not None:
            # The first `self.window_size` positions have to be removed from exog
            # since they are not in X_train.
            X_train_exog = exog.iloc[self.window_size:, ]
        else:
            X_train_exog = pd.DataFrame(
                               data    = np.nan,
                               columns = ['_dummy_exog_col_to_keep_shape'],
                               index   = y_index[self.window_size: ]
                           )

    y_train = pd.Series(
                  data  = y_train_values,
                  index = y_index[self.window_size: ],
                  name  = 'y'
              )

    if self.differentiation is not None:
        X_train_predictors = X_train_predictors.iloc[self.differentiation: ]
        y_train = y_train.iloc[self.differentiation: ]
        if X_train_exog is not None:
            X_train_exog = X_train_exog.iloc[self.differentiation: ]

    return X_train_predictors, X_train_exog, y_train

_create_train_X_y(series, exog=None, store_last_window=True)

Create training matrices from multiple time series and exogenous variables. See Notes section for more details depending on the type of series and exog.

Parameters:

Name Type Description Default
series pandas DataFrame, dict

Training time series.

required
exog pandas Series, pandas DataFrame, dict

Exogenous variable/s included as predictor/s.

`None`
store_last_window (bool, list)

Whether or not to store the last window of training data.

  • If True, last_window is stored for all series.
  • If list, last_window is stored for the series present in the list.
  • If False, last_window is not stored.
`True`

Returns:

Name Type Description
X_train pandas DataFrame

Training values (predictors).

y_train pandas Series

Values (target) of the time series related to each row of X_train.

series_indexes dict

Dictionary with the index of each series.

series_col_names list

Names of the series (levels) provided by the user during training.

series_X_train list

Names of the series (levels) included in the matrix X_train created internally for training. It can be different from series_col_names if some series are dropped during the training process because of NaNs or because they are not present in the training period.

exog_col_names list

Names of the exogenous variables used during training.

exog_dtypes dict

Type of each exogenous variable/s used in training. If transformer_exog is used, the dtypes are calculated before the transformation.

last_window dict

Last window of training data for each series. It stores the values needed to predict the next step immediately after the training data.

Notes
  • If series is a pandas DataFrame and exog is a pandas Series or DataFrame, each exog is duplicated for each series. Exog must have the same index as series (type, length and frequency).
  • If series is a pandas DataFrame and exog is a dict of pandas Series or DataFrames. Each key in exog must be a column in series and the values are the exog for each series. Exog must have the same index as series (type, length and frequency).
  • If series is a dict of pandas Series, exog must be a dict of pandas Series or DataFrames. The keys in series and exog must be the same. All series and exog must have a pandas DatetimeIndex with the same frequency.
Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
def _create_train_X_y(
    self,
    series: Union[pd.DataFrame, dict],
    exog: Optional[Union[pd.Series, pd.DataFrame, dict]]=None,
    store_last_window: Union[bool, list]=True,
) -> Tuple[pd.DataFrame, pd.Series, dict, list, list, list, dict, dict]:
    """
    Create training matrices from multiple time series and exogenous
    variables. See Notes section for more details depending on the type of
    `series` and `exog`.

    Parameters
    ----------
    series : pandas DataFrame, dict
        Training time series.
    exog : pandas Series, pandas DataFrame, dict, default `None`
        Exogenous variable/s included as predictor/s.
    store_last_window : bool, list, default `True`
        Whether or not to store the last window of training data.

        - If `True`, last_window is stored for all series. 
        - If `list`, last_window is stored for the series present in the list.
        - If `False`, last_window is not stored.

    Returns
    -------
    X_train : pandas DataFrame
        Training values (predictors).
    y_train : pandas Series
        Values (target) of the time series related to each row of `X_train`.
    series_indexes : dict
        Dictionary with the index of each series.
    series_col_names : list
        Names of the series (levels) provided by the user during training.
    series_X_train : list
        Names of the series (levels) included in the matrix `X_train` created
        internally for training. It can be different from `series_col_names` if
        some series are dropped during the training process because of NaNs or
        because they are not present in the training period.
    exog_col_names : list
        Names of the exogenous variables used during training.
    exog_dtypes : dict
        Type of each exogenous variable/s used in training. If `transformer_exog` 
        is used, the dtypes are calculated before the transformation.
    last_window : dict
        Last window of training data for each series. It stores the values 
        needed to predict the next `step` immediately after the training data.

    Notes
    -----
    - If `series` is a pandas DataFrame and `exog` is a pandas Series or 
    DataFrame, each exog is duplicated for each series. Exog must have the
    same index as `series` (type, length and frequency).
    - If `series` is a pandas DataFrame and `exog` is a dict of pandas Series 
    or DataFrames. Each key in `exog` must be a column in `series` and the 
    values are the exog for each series. Exog must have the same index as 
    `series` (type, length and frequency).
    - If `series` is a dict of pandas Series, `exog` must be a dict of pandas
    Series or DataFrames. The keys in `series` and `exog` must be the same.
    All series and exog must have a pandas DatetimeIndex with the same 
    frequency.

    """

    series_dict, series_indexes = check_preprocess_series(series=series)
    input_series_is_dict = isinstance(series, dict)
    series_col_names = list(series_dict.keys())

    if self.fitted and not (series_col_names == self.series_col_names):
        raise ValueError(
            (f"Once the Forecaster has been trained, `series` must have the "
             f"same columns as the series used during training:\n" 
             f" Got      : {series_col_names}\n"
             f" Expected : {self.series_col_names}")
        )

    for k, v in series_dict.items():
        if len(v) < self.window_size_diff + 1:
            raise ValueError(
                (f"Series '{k}' does not have enough values to calculate "
                 f"predictors. It must be at least {self.window_size_diff + 1}.")
            )

    exog_dict = {serie: None for serie in series_col_names}
    exog_col_names = None
    if exog is not None:
        exog_dict, exog_col_names = check_preprocess_exog_multiseries(
                                        input_series_is_dict = input_series_is_dict,
                                        series_indexes       = series_indexes,
                                        series_col_names     = series_col_names,
                                        exog                 = exog,
                                        exog_dict            = exog_dict
                                    )

        if self.fitted:
            if self.exog_col_names is None:
                raise ValueError(
                    ("Once the Forecaster has been trained, `exog` must be `None` "
                     "because no exogenous variables were added during training.")
                )
            else:
                if not set(exog_col_names) == set(self.exog_col_names):
                    raise ValueError(
                        (f"Once the Forecaster has been trained, `exog` must have the "
                         f"same columns as the series used during training:\n" 
                         f" Got      : {exog_col_names}\n"
                         f" Expected : {self.exog_col_names}")
                    )

    if not self.fitted:
        self.transformer_series_ = initialize_transformer_series(
                                       series_col_names = series_col_names,
                                       transformer_series = self.transformer_series
                                   )

    if self.differentiation is None:
        self.differentiator_ = {serie: None for serie in series_col_names}
    else:
        if not self.fitted:
            self.differentiator_ = {serie: clone(self.differentiator) 
                                    for serie in series_col_names}

    series_dict, exog_dict = align_series_and_exog_multiseries(
                                 series_dict          = series_dict,
                                 input_series_is_dict = input_series_is_dict,
                                 exog_dict            = exog_dict
                             )

    # TODO: parallelize
    # ======================================================================
    ignore_exog = True if exog is None else False
    input_matrices = [
        [series_dict[k], exog_dict[k], ignore_exog]
         for k in series_dict.keys()
    ]

    X_train_predictors_buffer = []
    X_train_exog_buffer = []
    y_train_buffer = []
    for matrices in input_matrices:

        X_train_predictors, X_train_exog, y_train = (
            self._create_train_X_y_single_series(
                y           = matrices[0],
                exog        = matrices[1],
                ignore_exog = matrices[2],
            )
        )

        X_train_predictors_buffer.append(X_train_predictors)
        X_train_exog_buffer.append(X_train_exog)
        y_train_buffer.append(y_train)
    # ======================================================================

    X_train = pd.concat(X_train_predictors_buffer, axis=0)
    y_train = pd.concat(y_train_buffer, axis=0)

    if self.fitted:
        encoded_values = self.encoder.transform(X_train[['_level_skforecast']])
    else:
        encoded_values = self.encoder.fit_transform(X_train[['_level_skforecast']])
        for i, code in enumerate(self.encoder.categories_[0]):
            self.encoding_mapping[code] = i

    X_train = pd.concat([
                  X_train.drop(columns='_level_skforecast'),
                  encoded_values
              ], axis=1)

    if self.encoding == 'onehot':
        X_train.columns = X_train.columns.str.replace('_level_skforecast_', '')
    elif self.encoding == 'ordinal_category':
        X_train['_level_skforecast'] = (
            X_train['_level_skforecast'].astype('category')
        )

    del encoded_values

    exog_dtypes = None
    if exog is not None:

        X_train_exog = pd.concat(X_train_exog_buffer, axis=0)
        if '_dummy_exog_col_to_keep_shape' in X_train_exog.columns:
            X_train_exog = (
                X_train_exog.drop(columns=['_dummy_exog_col_to_keep_shape'])
            )

        exog_col_names = X_train_exog.columns.to_list()
        exog_dtypes = get_exog_dtypes(exog=X_train_exog)

        fit_transformer = False if self.fitted else True
        X_train_exog = transform_dataframe(
                           df                = X_train_exog,
                           transformer       = self.transformer_exog,
                           fit               = fit_transformer,
                           inverse_transform = False
                       )

        check_exog_dtypes(X_train_exog, call_check_exog=False)
        if not (X_train_exog.index == X_train.index).all():
            raise ValueError(
                ("Different index for `series` and `exog` after transformation. "
                 "They must be equal to ensure the correct alignment of values.")
            )

        X_train = pd.concat([X_train, X_train_exog], axis=1)

    if y_train.isnull().any():
        mask = y_train.notna().to_numpy()
        y_train = y_train.iloc[mask]
        X_train = X_train.iloc[mask,]
        warnings.warn(
            ("NaNs detected in `y_train`. They have been dropped because the "
             "target variable cannot have NaN values. Same rows have been "
             "dropped from `X_train` to maintain alignment. This is caused by "
             "series with interspersed NaNs."),
             MissingValuesWarning
        )

    if self.dropna_from_series:
        if X_train.isnull().any().any():
            mask = X_train.notna().all(axis=1).to_numpy()
            X_train = X_train.iloc[mask, ]
            y_train = y_train.iloc[mask]
            warnings.warn(
                ("NaNs detected in `X_train`. They have been dropped. If "
                 "you want to keep them, set `forecaster.dropna_from_series = False`. " 
                 "Same rows have been removed from `y_train` to maintain alignment. "
                 "This caused by series with interspersed NaNs."),
                 MissingValuesWarning
            )
    else:
        if X_train.isnull().any().any():
            warnings.warn(
                ("NaNs detected in `X_train`. Some regressors do not allow "
                 "NaN values during training. If you want to drop them, "
                 "set `forecaster.dropna_from_series = True`."),
                 MissingValuesWarning
            )

    if X_train.empty:
        raise ValueError(
            ("All samples have been removed due to NaNs. Set "
             "`forecaster.dropna_from_series = False` or review `exog` values.")
        )

    if self.encoding == 'onehot':
        series_X_train = [
            col for col in series_col_names if X_train[col].sum() > 0
        ]
    else:
        series_X_train = [
            k for k, v in self.encoding_mapping.items()
            if v in X_train['_level_skforecast'].unique()
        ]

    # The last time window of training data is stored so that predictors 
    # in the first iteration of `predict()` can be calculated.
    last_window = None
    if store_last_window:

        series_to_store = (
            series_X_train if store_last_window is True else store_last_window
        )

        series_not_in_series_dict = set(series_to_store) - set(series_X_train)
        if series_not_in_series_dict:
            warnings.warn(
                (f"Series {series_not_in_series_dict} are not present in "
                 f"`series`. No last window is stored for them."),
                IgnoredArgumentWarning
            )
            series_to_store = [s for s in series_to_store 
                               if s not in series_not_in_series_dict]

        if series_to_store:
            last_window = {
                k: v.iloc[-self.window_size_diff:].copy()
                for k, v in series_dict.items()
                if k in series_to_store
            }

    return (
        X_train,
        y_train,
        series_indexes,
        series_col_names,
        series_X_train,
        exog_col_names,
        exog_dtypes,
        last_window,
    )

create_train_X_y(series, exog=None)

Create training matrices from multiple time series and exogenous variables. See Notes section for more details depending on the type of series and exog.

Parameters:

Name Type Description Default
series pandas DataFrame, dict

Training time series.

required
exog pandas Series, pandas DataFrame, dict

Exogenous variable/s included as predictor/s.

`None`

Returns:

Name Type Description
X_train pandas DataFrame

Training values (predictors).

y_train pandas Series

Values (target) of the time series related to each row of X_train.

Notes
  • If series is a pandas DataFrame and exog is a pandas Series or DataFrame, each exog is duplicated for each series. Exog must have the same index as series (type, length and frequency).
  • If series is a pandas DataFrame and exog is a dict of pandas Series or DataFrames. Each key in exog must be a column in series and the values are the exog for each series. Exog must have the same index as series (type, length and frequency).
  • If series is a dict of pandas Series, exogmust be a dict of pandas Series or DataFrames. The keys in series and exog must be the same. All series and exog must have a pandas DatetimeIndex with the same frequency.
Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
def create_train_X_y(
    self,
    series: Union[pd.DataFrame, dict],
    exog: Optional[Union[pd.Series, pd.DataFrame, dict]]=None
) -> Tuple[pd.DataFrame, pd.Series]:
    """
    Create training matrices from multiple time series and exogenous
    variables. See Notes section for more details depending on the type of
    `series` and `exog`.

    Parameters
    ----------
    series : pandas DataFrame, dict
        Training time series.
    exog : pandas Series, pandas DataFrame, dict, default `None`
        Exogenous variable/s included as predictor/s.

    Returns
    -------
    X_train : pandas DataFrame
        Training values (predictors).
    y_train : pandas Series
        Values (target) of the time series related to each row of `X_train`.

    Notes
    -----
    - If `series` is a pandas DataFrame and `exog` is a pandas Series or 
    DataFrame, each exog is duplicated for each series. Exog must have the
    same index as `series` (type, length and frequency).
    - If `series` is a pandas DataFrame and `exog` is a dict of pandas Series 
    or DataFrames. Each key in `exog` must be a column in `series` and the 
    values are the exog for each series. Exog must have the same index as 
    `series` (type, length and frequency).
    - If `series` is a dict of pandas Series, `exog`must be a dict of pandas
    Series or DataFrames. The keys in `series` and `exog` must be the same.
    All series and exog must have a pandas DatetimeIndex with the same 
    frequency.

    """

    output = self._create_train_X_y(
                 series            = series, 
                 exog              = exog, 
                 store_last_window = False
             )

    X_train = output[0]
    y_train = output[1]

    return X_train, y_train

create_sample_weights(series_col_names, X_train)

Crate weights for each observation according to the forecaster's attributes series_weights and weight_func. The resulting weights are product of both types of weights.

Parameters:

Name Type Description Default
series_col_names list

Names of the series (levels) used during training.

required
X_train pandas DataFrame

Dataframe created with the create_train_X_y method, first return.

required

Returns:

Name Type Description
weights numpy ndarray

Weights to use in fit method.

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
 935
 936
 937
 938
 939
 940
 941
 942
 943
 944
 945
 946
 947
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
def create_sample_weights(
    self,
    series_col_names: list,
    X_train: pd.DataFrame
)-> np.ndarray:
    """
    Crate weights for each observation according to the forecaster's attributes
    `series_weights` and `weight_func`. The resulting weights are product of both
    types of weights.

    Parameters
    ----------
    series_col_names : list
        Names of the series (levels) used during training.
    X_train : pandas DataFrame
        Dataframe created with the `create_train_X_y` method, first return.

    Returns
    -------
    weights : numpy ndarray
        Weights to use in `fit` method.

    """

    weights = None
    weights_samples = None
    weights_series = None

    if self.series_weights is not None:
        # Series not present in series_weights have a weight of 1 in all their samples.
        # Keys in series_weights not present in series are ignored.
        series_not_in_series_weights = set(series_col_names) - set(self.series_weights.keys())
        if series_not_in_series_weights:
            warnings.warn(
                (f"{series_not_in_series_weights} not present in `series_weights`. "
                 f"A weight of 1 is given to all their samples."),
                 IgnoredArgumentWarning
            )
        self.series_weights_ = {col: 1. for col in series_col_names}
        self.series_weights_.update(
            (k, v) 
            for k, v in self.series_weights.items() 
            if k in self.series_weights_
        )

        if self.encoding == "onehot":
            weights_series = [
                np.repeat(self.series_weights_[serie], sum(X_train[serie]))
                for serie in series_col_names
            ]
        else:
            weights_series = [
                np.repeat(
                    self.series_weights_[serie],
                    sum(X_train["_level_skforecast"] == self.encoding_mapping[serie]),
                )
                for serie in series_col_names
            ]

        weights_series = np.concatenate(weights_series)

    if self.weight_func is not None:
        if isinstance(self.weight_func, Callable):
            self.weight_func_ = {col: copy(self.weight_func) 
                                 for col in series_col_names}
        else:
            # Series not present in weight_func have a weight of 1 in all their samples
            series_not_in_weight_func = set(series_col_names) - set(self.weight_func.keys())
            if series_not_in_weight_func:
                warnings.warn(
                    (f"{series_not_in_weight_func} not present in `weight_func`. "
                     f"A weight of 1 is given to all their samples."),
                     IgnoredArgumentWarning
                )
            self.weight_func_ = {col: lambda x: np.ones_like(x, dtype=float) 
                                 for col in series_col_names}
            self.weight_func_.update(
                (k, v) 
                for k, v in self.weight_func.items() 
                if k in self.weight_func_
            )

        weights_samples = []
        for key in self.weight_func_.keys():
            if self.encoding == "onehot":
                idx = X_train.index[X_train[key] == 1.0]
            else:
                idx = X_train.index[X_train["_level_skforecast"] == self.encoding_mapping[key]]
            weights_samples.append(self.weight_func_[key](idx))
        weights_samples = np.concatenate(weights_samples)

    if weights_series is not None:
        weights = weights_series
        if weights_samples is not None:
            weights = weights * weights_samples
    else:
        if weights_samples is not None:
            weights = weights_samples

    if weights is not None:
        if np.isnan(weights).any():
            raise ValueError(
                "The resulting `weights` cannot have NaN values."
            )
        if np.any(weights < 0):
            raise ValueError(
                "The resulting `weights` cannot have negative values."
            )
        if np.sum(weights) == 0:
            raise ValueError(
                ("The resulting `weights` cannot be normalized because "
                 "the sum of the weights is zero.")
            )

    return weights

fit(series, exog=None, store_last_window=True, store_in_sample_residuals=True, suppress_warnings=False)

Training Forecaster. See Notes section for more details depending on the type of series and exog.

Additional arguments to be passed to the fit method of the regressor can be added with the fit_kwargs argument when initializing the forecaster.

Parameters:

Name Type Description Default
series pandas DataFrame, dict

Training time series.

required
exog pandas Series, pandas DataFrame, dict

Exogenous variable/s included as predictor/s.

`None`
store_last_window (bool, list)

Whether or not to store the last window of training data.

  • If True, last_window is stored for all series.
  • If list, last_window is stored for the series present in the list.
  • If False, last_window is not stored.
`True`
store_in_sample_residuals bool

If True, in-sample residuals will be stored in the forecaster object after fitting.

`True`
suppress_warnings bool

If True, skforecast warnings will be suppressed during the training process. See skforecast.exceptions.warn_skforecast_categories for more information.

`False`

Returns:

Type Description
None
Notes
  • If series is a pandas DataFrame and exog is a pandas Series or DataFrame, each exog is duplicated for each series. Exog must have the same index as series (type, length and frequency).
  • If series is a pandas DataFrame and exog is a dict of pandas Series or DataFrames. Each key in exog must be a column in series and the values are the exog for each series. Exog must have the same index as series (type, length and frequency).
  • If series is a dict of pandas Series, exogmust be a dict of pandas Series or DataFrames. The keys in series and exog must be the same. All series and exog must have a pandas DatetimeIndex with the same frequency.
Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
def fit(
    self,
    series: Union[pd.DataFrame, dict],
    exog: Optional[Union[pd.Series, pd.DataFrame, dict]]=None,
    store_last_window: Union[bool, list]=True,
    store_in_sample_residuals: bool=True,
    suppress_warnings: bool=False
) -> None:
    """
    Training Forecaster. See Notes section for more details depending on 
    the type of `series` and `exog`.

    Additional arguments to be passed to the `fit` method of the regressor 
    can be added with the `fit_kwargs` argument when initializing the forecaster.

    Parameters
    ----------
    series : pandas DataFrame, dict
        Training time series.
    exog : pandas Series, pandas DataFrame, dict, default `None`
        Exogenous variable/s included as predictor/s.
    store_last_window : bool, list, default `True`
        Whether or not to store the last window of training data.

        - If `True`, last_window is stored for all series. 
        - If `list`, last_window is stored for the series present in the list.
        - If `False`, last_window is not stored.
    store_in_sample_residuals : bool, default `True`
        If `True`, in-sample residuals will be stored in the forecaster object
        after fitting.
    suppress_warnings : bool, default `False`
        If `True`, skforecast warnings will be suppressed during the training 
        process. See skforecast.exceptions.warn_skforecast_categories for more
        information.

    Returns
    -------
    None

    Notes
    -----
    - If `series` is a pandas DataFrame and `exog` is a pandas Series or 
    DataFrame, each exog is duplicated for each series. Exog must have the
    same index as `series` (type, length and frequency).
    - If `series` is a pandas DataFrame and `exog` is a dict of pandas Series 
    or DataFrames. Each key in `exog` must be a column in `series` and the 
    values are the exog for each series. Exog must have the same index as 
    `series` (type, length and frequency).
    - If `series` is a dict of pandas Series, `exog`must be a dict of pandas
    Series or DataFrames. The keys in `series` and `exog` must be the same.
    All series and exog must have a pandas DatetimeIndex with the same 
    frequency.

    """

    set_skforecast_warnings(suppress_warnings, action='ignore')

    # Reset values in case the forecaster has already been fitted.
    self.series_col_names    = None
    self.X_train_col_names   = None
    self.series_X_train      = None
    self.included_exog       = False
    self.exog_type           = None
    self.exog_dtypes         = None
    self.exog_col_names      = None
    self.last_window         = None
    self.in_sample_residuals = None
    self.training_range      = None
    self.index_type          = None
    self.index_freq          = None
    self.fitted              = False

    (
        X_train,
        y_train,
        series_indexes,
        series_col_names,
        series_X_train,
        exog_col_names,
        exog_dtypes,
        last_window
    ) = self._create_train_X_y(
            series=series, exog=exog, store_last_window=store_last_window
    )

    sample_weight = self.create_sample_weights(
                        series_col_names = series_col_names,
                        X_train          = X_train
                    )

    if sample_weight is not None:
        self.regressor.fit(
            X             = X_train,
            y             = y_train,
            sample_weight = sample_weight,
            **self.fit_kwargs
        )
    else:
        self.regressor.fit(X=X_train, y=y_train, **self.fit_kwargs)

    self.series_col_names = series_col_names
    self.series_X_train = series_X_train
    self.X_train_col_names = X_train.columns.to_list()
    self.fitted = True
    self.fit_date = pd.Timestamp.today().strftime('%Y-%m-%d %H:%M:%S')
    self.training_range = {k: v[[0, -1]] for k, v in series_indexes.items()}
    self.index_type = type(series_indexes[series_col_names[0]])
    if isinstance(series_indexes[series_col_names[0]], pd.DatetimeIndex):
        self.index_freq = series_indexes[series_col_names[0]].freqstr
    else: 
        self.index_freq = series_indexes[series_col_names[0]].step

    if exog is not None:
        self.included_exog = True
        self.exog_type = type(exog)
        self.exog_col_names = exog_col_names
        self.exog_dtypes = exog_dtypes

    in_sample_residuals = {}
    if store_in_sample_residuals:

        residuals = y_train - self.regressor.predict(X_train)

        for col in series_X_train:
            if self.encoding == 'onehot':
                in_sample_residuals[col] = residuals.loc[X_train[col] == 1.].to_numpy()
            else:
                encoded_value = self.encoding_mapping[col]
                in_sample_residuals[col] = (
                    residuals.loc[X_train['_level_skforecast'] == encoded_value].to_numpy()
                )
            if len(in_sample_residuals[col]) > 1000:
                # Only up to 1000 residuals are stored
                rng = np.random.default_rng(seed=123)
                in_sample_residuals[col] = rng.choice(
                                               a       = in_sample_residuals[col], 
                                               size    = 1000, 
                                               replace = False
                                           )
    else:
        for col in series_X_train:
            in_sample_residuals[col] = None

    self.in_sample_residuals = in_sample_residuals

    if store_last_window:
        self.last_window = last_window

    set_skforecast_warnings(suppress_warnings, action='default')

_recursive_predict(steps, level, last_window, exog=None)

Predict n steps ahead. It is an iterative process in which, each prediction, is used as a predictor for the next step.

Parameters:

Name Type Description Default
steps int

Number of future steps predicted.

required
level str

Time series to be predicted.

required
last_window numpy ndarray

Series values used to create the predictors needed in the first iteration of the prediction (t + 1).

required
exog numpy ndarray

Exogenous variable/s included as predictor/s.

`None`

Returns:

Name Type Description
predictions numpy ndarray

Predicted values.

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
def _recursive_predict(
    self,
    steps: int,
    level: str,
    last_window: np.ndarray,
    exog: Optional[np.ndarray]=None
) -> np.ndarray:
    """
    Predict n steps ahead. It is an iterative process in which, each prediction,
    is used as a predictor for the next step.

    Parameters
    ----------
    steps : int
        Number of future steps predicted.
    level : str
        Time series to be predicted.
    last_window : numpy ndarray
        Series values used to create the predictors needed in the 
        first iteration of the prediction (t + 1).
    exog : numpy ndarray, default `None`
        Exogenous variable/s included as predictor/s.

    Returns
    -------
    predictions : numpy ndarray
        Predicted values.

    """

    predictions = np.full(shape=steps, fill_value=np.nan)
    level_encoded = np.array([self.encoding_mapping[level]], dtype='float64')

    for i in range(steps):

        X = self.fun_predictors(y=last_window).reshape(1, -1)

        if self.encoding == 'onehot':
            levels_dummies = np.zeros(shape=(1, len(self.series_col_names)), dtype=float)
            levels_dummies[0][self.series_col_names.index(level)] = 1.
            X = np.column_stack((X, levels_dummies.reshape(1, -1)))
        else:
            X = np.column_stack((X, level_encoded))

        if exog is not None:
            X = np.column_stack((X, exog[i, ].reshape(1, -1)))

        with warnings.catch_warnings():
            # Suppress scikit-learn warning: "X does not have valid feature names,
            # but NoOpTransformer was fitted with feature names".
            warnings.simplefilter("ignore")
            prediction = self.regressor.predict(X)
            predictions[i] = prediction.ravel()[0]

        # Update `last_window` values. The first position is discarded and
        # the new prediction is added at the end.
        last_window = np.append(last_window[1:], prediction)

    return predictions

predict(steps, levels=None, last_window=None, exog=None, suppress_warnings=False)

Predict n steps ahead. It is an recursive process in which, each prediction, is used as a predictor for the next step. Only levels whose last window ends at the same datetime index can be predicted together.

Parameters:

Name Type Description Default
steps int

Number of future steps predicted.

required
levels (str, list)

Time series to be predicted. If None all levels whose last window ends at the same datetime index will be predicted together.

`None`
last_window pandas DataFrame

Series values used to create the predictors needed in the first iteration of the prediction (t + 1). If last_window = None, the values stored in self.last_window are used to calculate the initial predictors, and the predictions start right after training data.

`None`
exog pandas Series, pandas DataFrame, dict

Exogenous variable/s included as predictor/s.

`None`
suppress_warnings bool

If True, skforecast warnings will be suppressed during the prediction process. See skforecast.exceptions.warn_skforecast_categories for more information.

`False`

Returns:

Name Type Description
predictions pandas DataFrame

Predicted values, one column for each level.

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
def predict(
    self,
    steps: int,
    levels: Optional[Union[str, list]]=None,
    last_window: Optional[pd.DataFrame]=None,
    exog: Optional[Union[pd.Series, pd.DataFrame, dict]]=None,
    suppress_warnings: bool=False
) -> pd.DataFrame:
    """
    Predict n steps ahead. It is an recursive process in which, each prediction,
    is used as a predictor for the next step. Only levels whose last window
    ends at the same datetime index can be predicted together.

    Parameters
    ----------
    steps : int
        Number of future steps predicted.
    levels : str, list, default `None`
        Time series to be predicted. If `None` all levels whose last window
        ends at the same datetime index will be predicted together.
    last_window : pandas DataFrame, default `None`
        Series values used to create the predictors needed in the 
        first iteration of the prediction (t + 1).
        If `last_window = None`, the values stored in `self.last_window` are
        used to calculate the initial predictors, and the predictions start
        right after training data.
    exog : pandas Series, pandas DataFrame, dict, default `None`
        Exogenous variable/s included as predictor/s.
    suppress_warnings : bool, default `False`
        If `True`, skforecast warnings will be suppressed during the prediction 
        process. See skforecast.exceptions.warn_skforecast_categories for more
        information.

    Returns
    -------
    predictions : pandas DataFrame
        Predicted values, one column for each level.

    """

    set_skforecast_warnings(suppress_warnings, action='ignore')

    input_levels_is_list = False
    if levels is None:
        levels = self.series_X_train
    elif isinstance(levels, str):
        levels = [levels]
    else:
        input_levels_is_list = True

    if last_window is None and self.fitted:
        available_last_windows = set() if self.last_window is None else set(self.last_window.keys())
        not_available_last_window = set(levels) - available_last_windows
        if not_available_last_window:
            warnings.warn(
                (f"Levels {not_available_last_window} are excluded from "
                 f"prediction since they were not stored in `last_window` "
                 f"attribute during training. If you don't want to retrain "
                 f"the Forecaster, provide `last_window` as argument."),
                IgnoredArgumentWarning
            )
            levels = [level for level in levels 
                      if level not in not_available_last_window]

            if not levels:
                raise ValueError(
                    ("No series to predict. None of the series are present in "
                     "`last_window` attribute. Provide `last_window` as argument "
                     "in predict method.")
                )

        last_index_levels = [
            v.index[-1] 
            for k, v in self.last_window.items()
            if k in levels
        ]
        if len(set(last_index_levels)) > 1:
            max_index_levels = max(last_index_levels)
            selected_levels = [
                k
                for k, v in self.last_window.items()
                if k in levels and v.index[-1] == max_index_levels
            ]

            series_excluded_from_last_window = set(levels) - set(selected_levels)
            levels = selected_levels

            if input_levels_is_list and series_excluded_from_last_window:
                warnings.warn(
                    (f"Only series whose last window ends at the same index "
                     f"can be predicted together. Series that do not reach "
                     f"the maximum index, '{max_index_levels}', are excluded "
                     f"from prediction: {series_excluded_from_last_window}."),
                    IgnoredArgumentWarning
                )

        last_window = pd.DataFrame(
            {k: v 
             for k, v in self.last_window.items() 
             if k in levels}
        )

    check_predict_input(
        forecaster_name  = type(self).__name__,
        steps            = steps,
        fitted           = self.fitted,
        included_exog    = self.included_exog,
        index_type       = self.index_type,
        index_freq       = self.index_freq,
        window_size      = self.window_size_diff,
        last_window      = last_window,
        last_window_exog = None,
        exog             = exog,
        exog_type        = self.exog_type,
        exog_col_names   = self.exog_col_names,
        interval         = None,
        alpha            = None,
        max_steps        = None,
        levels           = levels,
        series_col_names = self.series_col_names
    )

    last_window = last_window.iloc[-self.window_size_diff:, ].copy()
    _, last_window_index = preprocess_last_window(
                               last_window   = last_window,
                               return_values = False
                           )
    prediction_index = expand_index(
                           index = last_window_index,
                           steps = steps
                       )

    if exog is not None:
        if isinstance(exog, (pd.Series, pd.DataFrame)):
            if isinstance(exog, pd.DataFrame):
                exog = transform_dataframe(
                           df                = exog,
                           transformer       = self.transformer_exog,
                           fit               = False,
                           inverse_transform = False
                       )
            else:
                exog = transform_series(
                           series            = exog,
                           transformer       = self.transformer_exog,
                           fit               = False,
                           inverse_transform = False
                       )
            check_exog_dtypes(exog=exog)
            exog_values = exog.to_numpy()[:steps]
        else:
            # Empty dataframe to be filled with the exog values of each level
            empty_exog = pd.DataFrame(
                             data    = np.nan,
                             columns = self.exog_col_names,
                             index   = prediction_index
                         )
    else:
        exog_values = None

    predictions = []
    for level in levels:

        last_window_level = transform_series(
                                series            = last_window[level],
                                transformer       = self.transformer_series_[level],
                                fit               = False,
                                inverse_transform = False
                            )
        last_window_values = last_window_level.to_numpy()
        if self.differentiation is not None:
            last_window_values = self.differentiator_[level].fit_transform(last_window_values)

        if isinstance(exog, dict):
            # Fill the empty dataframe with the exog values of each level
            # and transform them if necessary
            exog_level = exog[level]
            if isinstance(exog_level, pd.Series):
                exog_level = exog_level.to_frame()

            exog_level = empty_exog.fillna(exog_level)
            exog_level = transform_dataframe(
                             df                = exog_level,
                             transformer       = self.transformer_exog,
                             fit               = False,
                             inverse_transform = False
                         )

            check_exog_dtypes(
                exog      = exog_level,
                series_id = f"`exog` for series '{level}'"
            )
            exog_values = exog_level.to_numpy()

        preds_level = self._recursive_predict(
                          steps       = steps,
                          level       = level,
                          last_window = last_window_values,
                          exog        = exog_values
                      )

        if self.differentiation is not None:
            preds_level = self.differentiator_[level].inverse_transform_next_window(preds_level)

        preds_level = pd.Series(
                          data  = preds_level,
                          index = prediction_index,
                          name  = level
                      )

        preds_level = transform_series(
                          series            = preds_level,
                          transformer       = self.transformer_series_[level],
                          fit               = False,
                          inverse_transform = True
                      )

        predictions.append(preds_level)

    predictions = pd.concat(predictions, axis=1)

    set_skforecast_warnings(suppress_warnings, action='default')

    return predictions

predict_bootstrapping(steps, levels=None, last_window=None, exog=None, n_boot=500, random_state=123, in_sample_residuals=True, suppress_warnings=False)

Generate multiple forecasting predictions using a bootstrapping process. By sampling from a collection of past observed errors (the residuals), each iteration of bootstrapping generates a different set of predictions. Only levels whose last window ends at the same datetime index can be predicted together. See the Notes section for more information.

Parameters:

Name Type Description Default
steps int

Number of future steps predicted.

required
levels (str, list)

Time series to be predicted. If None all levels whose last window ends at the same datetime index will be predicted together.

`None`
last_window pandas DataFrame

Series values used to create the predictors needed in the first iteration of the prediction (t + 1). If last_window = None, the values stored in self.last_window are used to calculate the initial predictors, and the predictions start right after training data.

`None`
exog pandas Series, pandas DataFrame, dict

Exogenous variable/s included as predictor/s.

`None`
n_boot int

Number of bootstrapping iterations used to estimate predictions.

`500`
random_state int

Sets a seed to the random generator, so that boot predictions are always deterministic.

`123`
in_sample_residuals bool

If True, residuals from the training data are used as proxy of prediction error to create predictions. If False, out of sample residuals are used. In the latter case, the user should have calculated and stored the residuals within the forecaster (see set_out_sample_residuals()).

`True`
suppress_warnings bool

If True, skforecast warnings will be suppressed during the prediction process. See skforecast.exceptions.warn_skforecast_categories for more information.

`False`

Returns:

Name Type Description
boot_predictions dict

Predictions generated by bootstrapping for each level.

Notes

More information about prediction intervals in forecasting: https://otexts.com/fpp3/prediction-intervals.html#prediction-intervals-from-bootstrapped-residuals Forecasting: Principles and Practice (3nd ed) Rob J Hyndman and George Athanasopoulos.

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
def predict_bootstrapping(
    self,
    steps: int,
    levels: Optional[Union[str, list]]=None,
    last_window: Optional[pd.DataFrame]=None,
    exog: Optional[Union[pd.Series, pd.DataFrame, dict]]=None,
    n_boot: int=500,
    random_state: int=123,
    in_sample_residuals: bool=True,
    suppress_warnings: bool=False
) -> dict:
    """
    Generate multiple forecasting predictions using a bootstrapping process. 
    By sampling from a collection of past observed errors (the residuals),
    each iteration of bootstrapping generates a different set of predictions. 
    Only levels whose last window ends at the same datetime index can be 
    predicted together. See the Notes section for more information. 

    Parameters
    ----------
    steps : int
        Number of future steps predicted.
    levels : str, list, default `None`
        Time series to be predicted. If `None` all levels whose last window
        ends at the same datetime index will be predicted together.
    last_window : pandas DataFrame, default `None`
        Series values used to create the predictors needed in the 
        first iteration of the prediction (t + 1).
        If `last_window = None`, the values stored in `self.last_window` are
        used to calculate the initial predictors, and the predictions start
        right after training data.
    exog : pandas Series, pandas DataFrame, dict, default `None`
        Exogenous variable/s included as predictor/s.
    n_boot : int, default `500`
        Number of bootstrapping iterations used to estimate predictions.
    random_state : int, default `123`
        Sets a seed to the random generator, so that boot predictions are always 
        deterministic.
    in_sample_residuals : bool, default `True`
        If `True`, residuals from the training data are used as proxy of
        prediction error to create predictions. If `False`, out of sample 
        residuals are used. In the latter case, the user should have
        calculated and stored the residuals within the forecaster (see
        `set_out_sample_residuals()`).
    suppress_warnings : bool, default `False`
        If `True`, skforecast warnings will be suppressed during the prediction 
        process. See skforecast.exceptions.warn_skforecast_categories for more
        information.

    Returns
    -------
    boot_predictions : dict
        Predictions generated by bootstrapping for each level.
        {level: pandas DataFrame, shape (steps, n_boot)}

    Notes
    -----
    More information about prediction intervals in forecasting:
    https://otexts.com/fpp3/prediction-intervals.html#prediction-intervals-from-bootstrapped-residuals
    Forecasting: Principles and Practice (3nd ed) Rob J Hyndman and George Athanasopoulos.

    """

    set_skforecast_warnings(suppress_warnings, action='ignore')

    if self.fitted:

        input_levels_is_list = False 
        if levels is None:
            levels = self.series_X_train
        elif isinstance(levels, str):
            levels = [levels]
        else:
            input_levels_is_list = True

        if last_window is None:
            available_last_windows = set() if self.last_window is None else set(self.last_window.keys())
            not_available_last_window = set(levels) - available_last_windows
            if not_available_last_window:
                warnings.warn(
                    (f"Levels {not_available_last_window} are excluded from "
                     f"prediction since they were not stored in `last_window` "
                     f"attribute during training. If you don't want to retrain "
                     f"the Forecaster, provide `last_window` as argument."),
                     IgnoredArgumentWarning
                )
                levels = [level for level in levels 
                          if level not in not_available_last_window]

                if not levels:
                    raise ValueError(
                        ("No series to predict. None of the series are present in "
                         "`last_window` attribute. Provide `last_window` as argument "
                         "in predict method.")
                    )

            last_index_levels = [
                v.index[-1] 
                for k, v in self.last_window.items()
                if k in levels
            ]
            if len(set(last_index_levels)) > 1:
                max_index_levels = max(last_index_levels)
                selected_levels = [
                    k
                    for k, v in self.last_window.items()
                    if k in levels and v.index[-1] == max_index_levels
                ]

                series_excluded_from_last_window = set(levels) - set(selected_levels)
                levels = selected_levels

                if input_levels_is_list and series_excluded_from_last_window:
                    warnings.warn(
                        (f"Only series whose last window ends at the same index "
                         f"can be predicted together. Series that do not reach "
                         f"the maximum index, '{max_index_levels}', are excluded "
                         f"from prediction: {series_excluded_from_last_window}."),
                         IgnoredArgumentWarning
                    )

            last_window = pd.DataFrame(
                {k: v 
                 for k, v in self.last_window.items() 
                 if k in levels}
            )

        if in_sample_residuals:
            if not set(levels).issubset(set(self.in_sample_residuals.keys())):
                raise ValueError(
                    (f"Not `forecaster.in_sample_residuals` for levels: "
                     f"{set(levels) - set(self.in_sample_residuals.keys())}.")
                )
            residuals_levels = self.in_sample_residuals
        else:
            if self.out_sample_residuals is None:
                raise ValueError(
                    ("`forecaster.out_sample_residuals` is `None`. Use "
                     "`in_sample_residuals=True` or method "
                     "`set_out_sample_residuals()` before `predict_interval()`, "
                     "`predict_bootstrapping()`,`predict_quantiles()` or "
                     "`predict_dist()`.")
                )
            else:
                if not set(levels).issubset(set(self.out_sample_residuals.keys())):
                    raise ValueError(
                        (f"Not `forecaster.out_sample_residuals` for levels: "
                         f"{set(levels) - set(self.out_sample_residuals.keys())}. "
                         f"Use method `set_out_sample_residuals()`.")
                    )
            residuals_levels = self.out_sample_residuals

        check_residuals = (
            "forecaster.in_sample_residuals" if in_sample_residuals
            else "forecaster.out_sample_residuals"
        )
        for level in levels:
            if (level not in residuals_levels.keys() or 
                residuals_levels[level] is None or 
                len(residuals_levels[level]) == 0):
                raise ValueError(
                    (f"Not available residuals for level '{level}'. "
                     f"Check `{check_residuals}`.")
                )
            elif (any(element is None for element in residuals_levels[level]) or
                  np.any(np.isnan(residuals_levels[level]))):
                raise ValueError(
                    (f"forecaster residuals for level '{level}' contains `None` "
                     f"or `NaNs` values. Check `{check_residuals}`.")
                )

    check_predict_input(
        forecaster_name  = type(self).__name__,
        steps            = steps,
        fitted           = self.fitted,
        included_exog    = self.included_exog,
        index_type       = self.index_type,
        index_freq       = self.index_freq,
        window_size      = self.window_size_diff,
        last_window      = last_window,
        last_window_exog = None,
        exog             = exog,
        exog_type        = self.exog_type,
        exog_col_names   = self.exog_col_names,
        interval         = None,
        alpha            = None,
        max_steps        = None,
        levels           = levels,
        series_col_names = self.series_col_names
    )

    last_window = last_window.iloc[-self.window_size_diff:, ].copy()
    _, last_window_index = preprocess_last_window(
                               last_window   = last_window,
                               return_values = False
                           )
    prediction_index = expand_index(
                           index = last_window_index,
                           steps = steps
                       )

    if exog is not None:
        if isinstance(exog, (pd.Series, pd.DataFrame)):
            if isinstance(exog, pd.DataFrame):
                exog = transform_dataframe(
                           df                = exog,
                           transformer       = self.transformer_exog,
                           fit               = False,
                           inverse_transform = False
                       )
            else:
                exog = transform_series(
                           series            = exog,
                           transformer       = self.transformer_exog,
                           fit               = False,
                           inverse_transform = False
                       )
            check_exog_dtypes(exog=exog)
            exog_values = exog.to_numpy()[:steps]
        else:
            # Empty dataframe to be filled with the exog values of each level
            empty_exog = pd.DataFrame(
                             data    = np.nan,
                             columns = self.exog_col_names,
                             index   = prediction_index
                         )
    else:
        exog_values = None

    boot_predictions = {}
    for level in levels:

        last_window_level = transform_series(
                                series            = last_window[level],
                                transformer       = self.transformer_series_[level],
                                fit               = False,
                                inverse_transform = False
                            )
        last_window_values = last_window_level.to_numpy()
        if self.differentiation is not None:
            last_window_values = self.differentiator_[level].fit_transform(last_window_values)

        if isinstance(exog, dict):
            # Fill the empty dataframe with the exog values of each level
            # and transform them if necessary
            exog_level = exog[level]
            if isinstance(exog_level, pd.Series):
                exog_level = exog_level.to_frame()

            exog_level = empty_exog.fillna(exog_level)
            exog_level = transform_dataframe(
                             df                = exog_level,
                             transformer       = self.transformer_exog,
                             fit               = False,
                             inverse_transform = False
                         )

            check_exog_dtypes(
                exog      = exog_level,
                series_id = f"`exog` for series '{level}'"
            )
            exog_values = exog_level.to_numpy()

        level_boot_predictions = np.full(
                                     shape      = (steps, n_boot),
                                     fill_value = np.nan,
                                     dtype      = float
                                 )
        rng = np.random.default_rng(seed=random_state)
        seeds = rng.integers(low=0, high=10000, size=n_boot)

        residuals = residuals_levels[level]

        for i in range(n_boot):
            # In each bootstraping iteration the initial last_window and exog
            # need to be restored.
            last_window_boot = last_window_values.copy()
            exog_boot = exog_values.copy() if exog is not None else None

            rng = np.random.default_rng(seed=seeds[i])
            sample_residuals = rng.choice(
                                   a       = residuals,
                                   size    = steps,
                                   replace = True
                               )

            for step in range(steps):

                prediction = self._recursive_predict(
                                 steps       = 1,
                                 level       = level,
                                 last_window = last_window_boot,
                                 exog        = exog_boot 
                             )

                prediction_with_residual = prediction + sample_residuals[step]
                level_boot_predictions[step, i] = prediction_with_residual[0]

                last_window_boot = np.append(
                                       last_window_boot[1:],
                                       prediction_with_residual
                                   )
                if exog is not None:
                    exog_boot = exog_boot[1:]

            if self.differentiation is not None:
                level_boot_predictions[:, i] = (
                    self.differentiator_[level].inverse_transform_next_window(level_boot_predictions[:, i])
                )

        level_boot_predictions = pd.DataFrame(
                                     data    = level_boot_predictions,
                                     index   = prediction_index,
                                     columns = [f"pred_boot_{i}" for i in range(n_boot)]
                                 )

        if self.transformer_series_[level]:
            for col in level_boot_predictions.columns:
                level_boot_predictions[col] = transform_series(
                                                  series            = level_boot_predictions[col],
                                                  transformer       = self.transformer_series_[level],
                                                  fit               = False,
                                                  inverse_transform = True
                                              )

        boot_predictions[level] = level_boot_predictions

    set_skforecast_warnings(suppress_warnings, action='default')

    return boot_predictions

predict_interval(steps, levels=None, last_window=None, exog=None, interval=[5, 95], n_boot=500, random_state=123, in_sample_residuals=True, suppress_warnings=False)

Iterative process in which, each prediction, is used as a predictor for the next step and bootstrapping is used to estimate prediction intervals. Both predictions and intervals are returned.

Parameters:

Name Type Description Default
steps int

Number of future steps predicted.

required
levels (str, list)

Time series to be predicted. If None all levels whose last window ends at the same datetime index will be predicted together.

`None`
last_window pandas DataFrame

Series values used to create the predictors needed in the first iteration of the prediction (t + 1). If last_window = None, the values stored in self.last_window are used to calculate the initial predictors, and the predictions start right after training data.

`None`
exog pandas Series, pandas DataFrame, dict

Exogenous variable/s included as predictor/s.

`None`
interval list

Confidence of the prediction interval estimated. Sequence of percentiles to compute, which must be between 0 and 100 inclusive. For example, interval of 95% should be as interval = [2.5, 97.5].

`[5, 95]`
n_boot int

Number of bootstrapping iterations used to estimate prediction intervals.

`500`
random_state int

Sets a seed to the random generator, so that boot predictions are always deterministic.

`123`
in_sample_residuals bool

If True, residuals from the training data are used as proxy of prediction error to create predictions. If False, out of sample residuals are used. In the latter case, the user should have calculated and stored the residuals within the forecaster (see set_out_sample_residuals()).

`True`
suppress_warnings bool

If True, skforecast warnings will be suppressed during the prediction process. See skforecast.exceptions.warn_skforecast_categories for more information.

`False`

Returns:

Name Type Description
predictions pandas DataFrame

Values predicted by the forecaster and their estimated interval.

  • level: predictions.
  • level_lower_bound: lower bound of the interval.
  • level_upper_bound: upper bound of the interval.
Notes

More information about prediction intervals in forecasting: https://otexts.com/fpp2/prediction-intervals.html Forecasting: Principles and Practice (2nd ed) Rob J Hyndman and George Athanasopoulos.

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
def predict_interval(
    self,
    steps: int,
    levels: Optional[Union[str, list]]=None,
    last_window: Optional[pd.DataFrame]=None,
    exog: Optional[Union[pd.Series, pd.DataFrame, dict]]=None,
    interval: list=[5, 95],
    n_boot: int=500,
    random_state: int=123,
    in_sample_residuals: bool=True,
    suppress_warnings: bool=False
) -> pd.DataFrame:
    """
    Iterative process in which, each prediction, is used as a predictor
    for the next step and bootstrapping is used to estimate prediction
    intervals. Both predictions and intervals are returned.

    Parameters
    ----------
    steps : int
        Number of future steps predicted.
    levels : str, list, default `None`
        Time series to be predicted. If `None` all levels whose last window
        ends at the same datetime index will be predicted together.
    last_window : pandas DataFrame, default `None`
        Series values used to create the predictors needed in the 
        first iteration of the prediction (t + 1).
        If `last_window = None`, the values stored in `self.last_window` are
        used to calculate the initial predictors, and the predictions start
        right after training data.
    exog : pandas Series, pandas DataFrame, dict, default `None`
        Exogenous variable/s included as predictor/s.
    interval : list, default `[5, 95]`
        Confidence of the prediction interval estimated. Sequence of 
        percentiles to compute, which must be between 0 and 100 inclusive. 
        For example, interval of 95% should be as `interval = [2.5, 97.5]`.
    n_boot : int, default `500`
        Number of bootstrapping iterations used to estimate prediction 
        intervals.
    random_state : int, default `123`
        Sets a seed to the random generator, so that boot predictions are always 
        deterministic.
    in_sample_residuals : bool, default `True`
        If `True`, residuals from the training data are used as proxy of
        prediction error to create predictions. If `False`, out of sample 
        residuals are used. In the latter case, the user should have
        calculated and stored the residuals within the forecaster (see
        `set_out_sample_residuals()`).
    suppress_warnings : bool, default `False`
        If `True`, skforecast warnings will be suppressed during the prediction 
        process. See skforecast.exceptions.warn_skforecast_categories for more
        information.

    Returns
    -------
    predictions : pandas DataFrame
        Values predicted by the forecaster and their estimated interval.

        - level: predictions.
        - level_lower_bound: lower bound of the interval.
        - level_upper_bound: upper bound of the interval.

    Notes
    -----
    More information about prediction intervals in forecasting:
    https://otexts.com/fpp2/prediction-intervals.html
    Forecasting: Principles and Practice (2nd ed) Rob J Hyndman and
    George Athanasopoulos.

    """

    set_skforecast_warnings(suppress_warnings, action='ignore')

    check_interval(interval=interval)

    preds = self.predict(
                steps             = steps,
                levels            = levels,
                last_window       = last_window,
                exog              = exog,
                suppress_warnings = suppress_warnings
            )

    boot_predictions = self.predict_bootstrapping(
                           steps               = steps,
                           levels              = levels,
                           last_window         = last_window,
                           exog                = exog,
                           n_boot              = n_boot,
                           random_state        = random_state,
                           in_sample_residuals = in_sample_residuals,
                           suppress_warnings   = suppress_warnings
                       )

    interval = np.array(interval)/100
    predictions = []

    for level in preds.columns:
        preds_interval = boot_predictions[level].quantile(q=interval, axis=1).transpose()
        preds_interval.columns = [f'{level}_lower_bound', f'{level}_upper_bound']
        predictions.append(preds[level])
        predictions.append(preds_interval)

    predictions = pd.concat(predictions, axis=1)

    set_skforecast_warnings(suppress_warnings, action='default')

    return predictions

predict_quantiles(steps, levels=None, last_window=None, exog=None, quantiles=[0.05, 0.5, 0.95], n_boot=500, random_state=123, in_sample_residuals=True, suppress_warnings=False)

Calculate the specified quantiles for each step. After generating multiple forecasting predictions through a bootstrapping process, each quantile is calculated for each step.

Parameters:

Name Type Description Default
steps int

Number of future steps predicted.

required
levels (str, list)

Time series to be predicted. If None all levels whose last window ends at the same datetime index will be predicted together.

`None`
last_window pandas DataFrame

Series values used to create the predictors needed in the first iteration of the prediction (t + 1). If last_window = None, the values stored in self.last_window are used to calculate the initial predictors, and the predictions start right after training data.

`None`
exog pandas Series, pandas DataFrame, dict

Exogenous variable/s included as predictor/s.

`None`
quantiles list

Sequence of quantiles to compute, which must be between 0 and 1 inclusive. For example, quantiles of 0.05, 0.5 and 0.95 should be as quantiles = [0.05, 0.5, 0.95].

`[0.05, 0.5, 0.95]`
n_boot int

Number of bootstrapping iterations used to estimate quantiles.

`500`
random_state int

Sets a seed to the random generator, so that boot quantiles are always deterministic.

`123`
in_sample_residuals bool

If True, residuals from the training data are used as proxy of prediction error to create quantiles. If False, out of sample residuals are used. In the latter case, the user should have calculated and stored the residuals within the forecaster (see set_out_sample_residuals()).

`True`
suppress_warnings bool

If True, skforecast warnings will be suppressed during the prediction process. See skforecast.exceptions.warn_skforecast_categories for more information.

`False`

Returns:

Name Type Description
predictions pandas DataFrame

Quantiles predicted by the forecaster.

Notes

More information about prediction intervals in forecasting: https://otexts.com/fpp2/prediction-intervals.html Forecasting: Principles and Practice (2nd ed) Rob J Hyndman and George Athanasopoulos.

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
def predict_quantiles(
    self,
    steps: int,
    levels: Optional[Union[str, list]]=None,
    last_window: Optional[pd.DataFrame]=None,
    exog: Optional[Union[pd.Series, pd.DataFrame, dict]]=None,
    quantiles: list=[0.05, 0.5, 0.95],
    n_boot: int=500,
    random_state: int=123,
    in_sample_residuals: bool=True,
    suppress_warnings: bool=False
) -> pd.DataFrame:
    """
    Calculate the specified quantiles for each step. After generating 
    multiple forecasting predictions through a bootstrapping process, each 
    quantile is calculated for each step.

    Parameters
    ----------
    steps : int
        Number of future steps predicted.
    levels : str, list, default `None`
        Time series to be predicted. If `None` all levels whose last window
        ends at the same datetime index will be predicted together.
    last_window : pandas DataFrame, default `None`
        Series values used to create the predictors needed in the 
        first iteration of the prediction (t + 1).
        If `last_window = None`, the values stored in `self.last_window` are
        used to calculate the initial predictors, and the predictions start
        right after training data.
    exog : pandas Series, pandas DataFrame, dict, default `None`
        Exogenous variable/s included as predictor/s.
    quantiles : list, default `[0.05, 0.5, 0.95]`
        Sequence of quantiles to compute, which must be between 0 and 1 
        inclusive. For example, quantiles of 0.05, 0.5 and 0.95 should be as 
        `quantiles = [0.05, 0.5, 0.95]`.
    n_boot : int, default `500`
        Number of bootstrapping iterations used to estimate quantiles.
    random_state : int, default `123`
        Sets a seed to the random generator, so that boot quantiles are always 
        deterministic.
    in_sample_residuals : bool, default `True`
        If `True`, residuals from the training data are used as proxy of
        prediction error to create quantiles. If `False`, out of sample 
        residuals are used. In the latter case, the user should have
        calculated and stored the residuals within the forecaster (see
        `set_out_sample_residuals()`).
    suppress_warnings : bool, default `False`
        If `True`, skforecast warnings will be suppressed during the prediction 
        process. See skforecast.exceptions.warn_skforecast_categories for more
        information.

    Returns
    -------
    predictions : pandas DataFrame
        Quantiles predicted by the forecaster.

    Notes
    -----
    More information about prediction intervals in forecasting:
    https://otexts.com/fpp2/prediction-intervals.html
    Forecasting: Principles and Practice (2nd ed) Rob J Hyndman and
    George Athanasopoulos.

    """

    set_skforecast_warnings(suppress_warnings, action='ignore')

    check_interval(quantiles=quantiles)

    boot_predictions = self.predict_bootstrapping(
                           steps               = steps,
                           levels              = levels,
                           last_window         = last_window,
                           exog                = exog,
                           n_boot              = n_boot,
                           random_state        = random_state,
                           in_sample_residuals = in_sample_residuals,
                           suppress_warnings   = suppress_warnings
                       )

    predictions = []

    for level in boot_predictions.keys():
        preds_quantiles = boot_predictions[level].quantile(q=quantiles, axis=1).transpose()
        preds_quantiles.columns = [f'{level}_q_{q}' for q in quantiles]
        predictions.append(preds_quantiles)

    predictions = pd.concat(predictions, axis=1)

    set_skforecast_warnings(suppress_warnings, action='default')

    return predictions

predict_dist(steps, distribution, levels=None, last_window=None, exog=None, n_boot=500, random_state=123, in_sample_residuals=True, suppress_warnings=False)

Fit a given probability distribution for each step. After generating multiple forecasting predictions through a bootstrapping process, each step is fitted to the given distribution.

Parameters:

Name Type Description Default
steps int

Number of future steps predicted.

required
distribution Object

A distribution object from scipy.stats. For example scipy.stats.norm.

required
levels (str, list)

Time series to be predicted. If None all levels whose last window ends at the same datetime index will be predicted together.

`None`
last_window pandas DataFrame

Series values used to create the predictors needed in the first iteration of the prediction (t + 1). If last_window = None, the values stored in self.last_window are used to calculate the initial predictors, and the predictions start right after training data.

`None`
exog pandas Series, pandas DataFrame, dict

Exogenous variable/s included as predictor/s.

`None`
n_boot int

Number of bootstrapping iterations used to estimate predictions.

`500`
random_state int

Sets a seed to the random generator, so that boot predictions are always deterministic.

`123`
in_sample_residuals bool

If True, residuals from the training data are used as proxy of prediction error to create predictions. If False, out of sample residuals are used. In the latter case, the user should have calculated and stored the residuals within the forecaster (see set_out_sample_residuals()).

`True`
suppress_warnings bool

If True, skforecast warnings will be suppressed during the prediction process. See skforecast.exceptions.warn_skforecast_categories for more information.

`False`

Returns:

Name Type Description
predictions pandas DataFrame

Distribution parameters estimated for each step and level.

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
def predict_dist(
    self,
    steps: int,
    distribution: object,
    levels: Optional[Union[str, list]]=None,
    last_window: Optional[pd.DataFrame]=None,
    exog: Optional[Union[pd.Series, pd.DataFrame, dict]]=None,
    n_boot: int=500,
    random_state: int=123,
    in_sample_residuals: bool=True,
    suppress_warnings: bool=False
) -> pd.DataFrame:
    """
    Fit a given probability distribution for each step. After generating 
    multiple forecasting predictions through a bootstrapping process, each 
    step is fitted to the given distribution.

    Parameters
    ----------
    steps : int
        Number of future steps predicted.
    distribution : Object
        A distribution object from scipy.stats. For example scipy.stats.norm.
    levels : str, list, default `None`
        Time series to be predicted. If `None` all levels whose last window
        ends at the same datetime index will be predicted together.
    last_window : pandas DataFrame, default `None`
        Series values used to create the predictors needed in the 
        first iteration of the prediction (t + 1).
        If `last_window = None`, the values stored in `self.last_window` are
        used to calculate the initial predictors, and the predictions start
        right after training data.
    exog : pandas Series, pandas DataFrame, dict, default `None`
        Exogenous variable/s included as predictor/s.
    n_boot : int, default `500`
        Number of bootstrapping iterations used to estimate predictions.
    random_state : int, default `123`
        Sets a seed to the random generator, so that boot predictions are always 
        deterministic.
    in_sample_residuals : bool, default `True`
        If `True`, residuals from the training data are used as proxy of
        prediction error to create predictions. If `False`, out of sample 
        residuals are used. In the latter case, the user should have
        calculated and stored the residuals within the forecaster (see
        `set_out_sample_residuals()`).
    suppress_warnings : bool, default `False`
        If `True`, skforecast warnings will be suppressed during the prediction 
        process. See skforecast.exceptions.warn_skforecast_categories for more
        information.

    Returns
    -------
    predictions : pandas DataFrame
        Distribution parameters estimated for each step and level.

    """

    set_skforecast_warnings(suppress_warnings, action='ignore')

    boot_samples = self.predict_bootstrapping(
                       steps               = steps,
                       levels              = levels,
                       last_window         = last_window,
                       exog                = exog,
                       n_boot              = n_boot,
                       random_state        = random_state,
                       in_sample_residuals = in_sample_residuals,
                       suppress_warnings   = suppress_warnings
                   )

    param_names = [
        p for p in inspect.signature(distribution._pdf).parameters if not p == "x"
    ] + ["loc", "scale"]
    predictions = []

    for level in boot_samples.keys():
        param_values = np.apply_along_axis(
            lambda x: distribution.fit(x), axis=1, arr=boot_samples[level]
        )
        level_param_names = [f'{level}_{p}' for p in param_names]

        pred_level = pd.DataFrame(
                         data    = param_values,
                         columns = level_param_names,
                         index   = boot_samples[level].index
                     )

        predictions.append(pred_level)

    predictions = pd.concat(predictions, axis=1)

    set_skforecast_warnings(suppress_warnings, action='default')

    return predictions

set_params(params)

Set new values to the parameters of the scikit learn model stored in the forecaster.

Parameters:

Name Type Description Default
params dict

Parameters values.

required

Returns:

Type Description
None
Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
def set_params(
    self, 
    params: dict
) -> None:
    """
    Set new values to the parameters of the scikit learn model stored in the
    forecaster.

    Parameters
    ----------
    params : dict
        Parameters values.

    Returns
    -------
    None

    """

    self.regressor = clone(self.regressor)
    self.regressor.set_params(**params)

set_fit_kwargs(fit_kwargs)

Set new values for the additional keyword arguments passed to the fit method of the regressor.

Parameters:

Name Type Description Default
fit_kwargs dict

Dict of the form {"argument": new_value}.

required

Returns:

Type Description
None
Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
def set_fit_kwargs(
    self, 
    fit_kwargs: dict
) -> None:
    """
    Set new values for the additional keyword arguments passed to the `fit` 
    method of the regressor.

    Parameters
    ----------
    fit_kwargs : dict
        Dict of the form {"argument": new_value}.

    Returns
    -------
    None

    """

    self.fit_kwargs = check_select_fit_kwargs(self.regressor, fit_kwargs=fit_kwargs)

set_out_sample_residuals(residuals, append=True, transform=True, random_state=123)

Set new values to the attribute out_sample_residuals. Out of sample residuals are meant to be calculated using observations that did not participate in the training process.

Parameters:

Name Type Description Default
residuals dict

Dictionary of numpy ndarrays with the residuals of each level in the form {level: residuals}. If len(residuals) > 1000, only a random sample of 1000 values are stored. Keys must be the same as levels.

required
append bool

If True, new residuals are added to the once already stored in the attribute out_sample_residuals. Once the limit of 1000 values is reached, no more values are appended. If False, out_sample_residuals is overwritten with the new residuals.

`True`
transform bool

If True, new residuals are transformed using self.transformer_series.

`True`
random_state int

Sets a seed to the random sampling for reproducible output.

`123`

Returns:

Type Description
None
Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
def set_out_sample_residuals(
    self, 
    residuals: dict,
    append: bool=True,
    transform: bool=True,
    random_state: int=123
)-> None:
    """
    Set new values to the attribute `out_sample_residuals`. Out of sample
    residuals are meant to be calculated using observations that did not
    participate in the training process.

    Parameters
    ----------
    residuals : dict
        Dictionary of numpy ndarrays with the residuals of each level in the
        form {level: residuals}. If len(residuals) > 1000, only a random 
        sample of 1000 values are stored. Keys must be the same as `levels`.
    append : bool, default `True`
        If `True`, new residuals are added to the once already stored in the
        attribute `out_sample_residuals`. Once the limit of 1000 values is
        reached, no more values are appended. If False, `out_sample_residuals`
        is overwritten with the new residuals.
    transform : bool, default `True`
        If `True`, new residuals are transformed using self.transformer_series.
    random_state : int, default `123`
        Sets a seed to the random sampling for reproducible output.

    Returns
    -------
    None

    """

    if not isinstance(residuals, dict) or not all(isinstance(x, np.ndarray) for x in residuals.values()):
        raise TypeError(
            (f"`residuals` argument must be a dict of numpy ndarrays in the form "
             "`{level: residuals}`. " 
             f"Got {type(residuals)}.")
        )

    if not self.fitted:
        raise NotFittedError(
            ("This forecaster is not fitted yet. Call `fit` with appropriate "
             "arguments before using `set_out_sample_residuals()`.")
        )

    if self.out_sample_residuals is None:
        self.out_sample_residuals = {level: None for level in self.series_col_names}

    if not set(self.out_sample_residuals.keys()).issubset(set(residuals.keys())):
        warnings.warn(
            (
                f"Only residuals of levels "
                f"{set(self.out_sample_residuals.keys()).intersection(set(residuals.keys()))} "
                f"are updated."
            ), IgnoredArgumentWarning
        )

    residuals = {
        key: value 
        for key, value in residuals.items() 
        if key in self.out_sample_residuals.keys()
    }

    for level, value in residuals.items():

        residuals_level = value

        if not transform and self.transformer_series_[level] is not None:
            warnings.warn(
                (f"Argument `transform` is set to `False` but forecaster was "
                 f"trained using a transformer {self.transformer_series_[level]} "
                 f"for level {level}. Ensure that the new residuals are "
                 f"already transformed or set `transform=True`.")
            )

        if transform and self.transformer_series_ and self.transformer_series_[level]:
            warnings.warn(
                (f"Residuals will be transformed using the same transformer used "
                 f"when training the forecaster for level {level} : "
                 f"({self.transformer_series_[level]}). Ensure that the new "
                 f"residuals are on the same scale as the original time series.")
            )
            residuals_level = transform_series(
                                  series            = pd.Series(residuals_level, name='residuals'),
                                  transformer       = self.transformer_series_[level],
                                  fit               = False,
                                  inverse_transform = False
                              ).to_numpy()

        if len(residuals_level) > 1000:
            rng = np.random.default_rng(seed=random_state)
            residuals_level = rng.choice(a=residuals_level, size=1000, replace=False)

        if append and self.out_sample_residuals[level] is not None:
            free_space = max(0, 1000 - len(self.out_sample_residuals[level]))
            if len(residuals_level) < free_space:
                residuals_level = np.hstack((
                                      self.out_sample_residuals[level],
                                      residuals_level
                                  ))
            else:
                residuals_level = np.hstack((
                                      self.out_sample_residuals[level],
                                      residuals_level[:free_space]
                                  ))

        self.out_sample_residuals[level] = residuals_level

get_feature_importances(sort_importance=True)

Return feature importances of the regressor stored in the forecaster. Only valid when regressor stores internally the feature importances in the attribute feature_importances_ or coef_.

Parameters:

Name Type Description Default
sort_importance bool

If True, sorts the feature importances in descending order.

True

Returns:

Name Type Description
feature_importances pandas DataFrame

Feature importances associated with each predictor.

Source code in skforecast\ForecasterAutoregMultiSeriesCustom\ForecasterAutoregMultiSeriesCustom.py
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
def get_feature_importances(
    self,
    sort_importance: bool=True
) -> pd.DataFrame:
    """
    Return feature importances of the regressor stored in the
    forecaster. Only valid when regressor stores internally the feature
    importances in the attribute `feature_importances_` or `coef_`.

    Parameters
    ----------
    sort_importance: bool, default `True`
        If `True`, sorts the feature importances in descending order.

    Returns
    -------
    feature_importances : pandas DataFrame
        Feature importances associated with each predictor.

    """

    if not self.fitted:
        raise NotFittedError(
            ("This forecaster is not fitted yet. Call `fit` with appropriate "
             "arguments before using `get_feature_importances()`.")
        )

    if isinstance(self.regressor, Pipeline):
        estimator = self.regressor[-1]
    else:
        estimator = self.regressor

    if hasattr(estimator, 'feature_importances_'):
        feature_importances = estimator.feature_importances_
    elif hasattr(estimator, 'coef_'):
        feature_importances = estimator.coef_
    else:
        warnings.warn(
            (f"Impossible to access feature importances for regressor of type "
             f"{type(estimator)}. This method is only valid when the "
             f"regressor stores internally the feature importances in the "
             f"attribute `feature_importances_` or `coef_`.")
        )
        feature_importances = None

    if feature_importances is not None:
        feature_importances = pd.DataFrame({
                                  'feature': self.X_train_col_names,
                                  'importance': feature_importances
                              })
        if sort_importance:
            feature_importances = feature_importances.sort_values(
                                      by='importance', ascending=False
                                  )

    return feature_importances