For each data transformer in pipeline transform data. Then transformed data is passed to next transformer. Parameters ------

Convert a point along the time index into an integer index ranging from (0, len(series)-1) inclusive. Parameters ---------- p

Return a new series with added hierarchy. Parameters ---------- hierarchy A dictionary mapping components to a li

Returns the inner metric function `func` which bypasses the decorators `multi_ts_support` and `multivariate_support`. It significantly decreases p

Generates names for the parameters of the Likelihood.

Return a new series with the cumulative sum along the time axis. Returns ------- TimeSeries A new series, with th

Return a new series where everything after (and in-/excluding) the provided time `split_point` was dropped. The timestamp may not be in the s

Create a list of ``TimeSeries`` grouped by a selection of columns from a `DataFrame`. One column (or the DataFrame index) has to represent th

Get the number of time steps with a given frequency `freq` between `end` and `start`. Works for both integers and time stamps. * if `end`, `s

Return a new univariate series with a selected component. This drops the hierarchy (if any), and retains only the relevant static covariates

Return a new series similar to this one but with new `values`. Parameters ---------- values A Numpy array with ne

Computes a score/metric between anomaly scores against true anomalies. `anomalies` and `pred_scores` must have the same shape. `anomalies` mu

Returns the estimator that forecasts the step ``horizon`` of the target component ``target_dim``. For probabilistic models fitting quantiles,

For each data transformer in the pipeline, inverse-transform data. Then inverse transformed data is passed to the next transformer. T

Sets the value of the specified option. Available Options: - display.[max_rows, max_cols] - plotting.use_darts_style Parameter

Helper function that creates a `linear_timeseries` with a specified number of components. To help distinguish each component from one

Return a new series with `values` prepended to this series along the time axis (added to the beginning). This adds time steps to the beginnin

Create ``__all__``, ``__getattr__``, and ``__dir__`` for a lazy package. Parameters ---------- lazy_imports Mapping of public att

Beta distribution. https://en.wikipedia.org/wiki/Beta_distribution - Univariate continuous distribution. - Support:

Add static covariates to the features' table for SKLearnModels. If `uses_static_covariates=True`, all target series used in `fit()` and `pred

Find the best hyper-parameters among a given set using a grid search. This function has 3 modes of operation: Expanding window mode,

Mean Absolute Percentage Error (MAPE). For the true series :math:`y` and predicted series :math:`\\hat{y}` of length :math:`T`, it is computed as

Return a deterministic series with the standard deviation of each component computed over the samples of the stochastic series. This

Computes the anomaly score between `series` and `pred_series`, and returns the score of an agnostic threshold metric. Parameters

Generates/extracts time index (or integer index) for covariates at model inference / prediction time. Parameters ----------

If lags were specified in a component-wise manner, they are contained in self.component_lags and the values in self.lags should be ig

Returns encoded index for all past and/or future covariates for inference/prediction. Which covariates are generated depends on the parameters

Each subclass must implement a method to encode the covariates index for training. Parameters ---------- target T

Sequentially applies the Kalman filter on the provided series of observations. Parameters ---------- series

Compute and return gaps in the series. Works only on deterministic time series (1 sample). Parameters ---------- mod

Root Mean Squared Error (RMSE). For the true series :math:`y` and predicted series :math:`\\hat{y}` of length :math:`T`, it is computed per c

Returns a list of series, where all series have been intersected along the time index. Parameters ---------- series sequence of `

Extract params from model creation to set up PLForecastingModule (the actual torch.nn.Module)

extract params from model creation to set up TorchForecastingModels

Returns encoded index for all past and/or future covariates for training. Which covariates are generated depends on the parameters used at mod

Generates/extracts time index (or integer index) for covariates at model training time. Parameters ---------- target

Return a new series with added metadata. Parameters ---------- metadata A dictionary with metadata to be added to

Builds a forecast time series starting after the end of the training time series, with the correct time index (or after the end of th

Helper to compare series differenced by `Diff`. Parameters ---------- series1 First `TimeSeries`

Return a new series where everything before (and in-/excluding) the provided time `split_point` was dropped. The timestamp may not be in the

Each subclass must implement a method to encode the covariates index for prediction. Parameters ---------- n The

Fit/train the model on one or multiple series. Parameters ---------- series TimeSeries or Sequence[TimeS

Fits the scorer on the two (sequences of) series. The function `diff_fn` passed as a parameter to the scorer, will transform `pred_series` an

Display a SHAP "Force Plot" for one target and one horizon. It shows SHAP values of all input features with an additive force layout

Parameters ---------- attribute The attribute of the underlying pd.DatetimeIndex from for which to apply cyclic

Helper function that creates the lagged features 'block' of a specific `series` (i.e. either `target_series`, `past_covariates`, or `

Extracts trend and seasonality from a TimeSeries instance using `statsmodels.tsa`. Parameters ---------- ts The series to de

static cov and metadata are not transferred when performing value comparisons. Output is a `np.ndarray`

Creates an array of shape `(n_timesteps * n_samples, n_components)` from either a `TimeSeries` or the `array_values` of a `TimeSeries

Utility function that parallelise the execution of a function over an Iterator Parameters ---------- iterator (Iterator[Tuple])

Return the sliced values of the series where the time index was intersected with the `other` series. This method is in general *not* symmetri

Return a slice of the deterministic time series where NaN-containing entries at the beginning and the end were removed. No entries af

Checks whether the given series has a given seasonality. Analyzes the given TimeSeries instance for seasonality of the given period while tak

Get the `Likelihood` object for `SKLearnModel`. Parameters ---------- likelihood The likelihood name. Must be one of `available_l

Extracts components specified by `component_mask` from `series` Parameters ---------- series input TimeS

Explains the last forecast of a foreground series (or background, if foreground is not provided) and returns a :class:`ShapSingleExpl

Fills missing values in the provided time series Parameters ---------- series The time series for which to fill missing valu

Formats a dictionary as a string, showing at most `max_items` items. Keys longer than `pad` are truncated with '...'. Values longer than `max_

Whether the series has the same time index as the `other` series. Parameters ---------- other the other series

extracts encoders from parameters at model creation

Forecasts values for `n` time steps after the end of the series. Parameters ---------- n : int Forecast horizon -

Extracts moving window views of an `x` array along a specified `axis`, where each window is of length `window_len` and consecutive windows ar

Reshapes the 2D array returned by `stack_samples` back into an array of shape `(n_timesteps, n_components, n_samples)`; this 'undoes'

Extract values from another series or array and check for compatible shapes.

Check if *obj* is an xarray type without importing xarray.

Marks a grid cell as active. Parameters ---------- elem Tuple of grid cell index (column, row)

Return an ``xarray.DataArray`` representation of the series. Parameters ---------- copy Whether to return a copy

Helper method to compare all the entries between two historical forecasts

Mean Absolute Error (MAE). For the true series :math:`y` and predicted series :math:`\\hat{y}` of length :math:`T`, it is computed per compon

Context manager to temporarily set options in the `with` statement context. Available Options: - display.[max_rows, max_cols] - plo

Return a new series, which is a multiple of this series such that the first value is `value_at_first_step`. Note: Numerical errors can appear

Reset one or more options to their default value. Available Options: - display.[max_rows, max_cols] - plotting.use_darts_style

If `ts` is a Sequence with only a single series, return the single series as TimeSeries. Parameters ---------- ts None, a single