Plot the results generated by an anomaly model. The plot will be composed of the following: - the actual series itself with the output of

Adds back components previously removed by `component_mask` in `apply_component_mask` method. Parameters ----------

Compute the accuracy of the anomaly scores computed by the model. Predicts the `series` with the forecasting model, and applies the scorer(s)

Create a ``TimeSeries`` from the JSON String representation of a ``TimeSeries``. The JSON String representation can be generated with :func:`

Creates a binary univariate TimeSeries with index `time_index` that equals 1 at every index that lies within (or equals) a selected country's

Gives the index paths from `series1` to `series2`. Returns ------- np.ndarray of shape `(len(path), 2)` An array

This method allows for predicting with a specific :class:`darts.utils.data.TorchInferenceDataset` instance. These datasets implement

Display a SHAP "Summary Plot" for each horizon and each component dimension of the target. On each summary plot, SHAP values of each

Return a new series with the last `size` points. Parameters ---------- size : int, default: 5 number of points to

Validates that the predict sample matches a sample that the model was trained on. For models relying on `TorchTrainingDataset` and `TorchInfe

Return a new series with one (or more) additional component(s) that contain an attribute of the series' time index. The additional co

Expands a np.ndarray to `ndim` dimensions (if not already satisfied).

Fit the underlying forecasting model (if applicable) and the fittable scorers, if any. Train the forecasting model (if not already fitted and

Initializes the Kalman filter using the N4SID algorithm. Parameters ---------- series The series of outp

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

Provides four tests for granger non causality of 2 time series using :func:`statsmodels.tsa.stattools.grangercausalitytests`. See [1]_.

Return a new series with the first `size` points. Parameters ---------- size : int, default 5 number of points

Return a new series with the median computed over the specified `axis`. If we reduce over time (``axis=0``), the series will have length one

Return a 2-D array of shape (time, component), containing the series' values for one sample taken uniformly at random from all samples.

Write the deterministic series to a CSV file. For a list of parameters, refer to the documentation of :func:`pandas.DataFrame.to_csv()` [1]_.

Return a JSON string representation of the deterministic series. At the moment this function works only on deterministic time series (i.e., m

Helper that extracts and returns the explainability result attribute for a specified horizon and component from the input attribute.

Verify that all static covariates are numeric.

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

Creates the SHAP input for the given series and covariates, by following the logic of the model's prediction / inference dataset and

Gives the total distance between pair-wise elements in the two series after warping. Returns ------- float The to

Compute a metric for the anomaly scores computed by the model. Predicts the `series` with the filtering model, and applies the scorer(s) on t

Fits/trains the model using the provided list of features time series and the target time series. Parameters ----------

Formats bytes as human-readable text. Parameters ---------- nbytes The number of bytes to format. precision The numbe

Create a ``TimeSeries`` from an `xarray.DataArray`. The dimensions of the DataArray have to be (time, component, sample), in this order. The

Generate categorical series with specific probabilities for each class

instantiates the SequentialEncoder object based on self._model_encoder_settings and parameter ``add_encoders`` used at model creation

Creates a collapsible HTML section. Parameters ---------- title The title of the section. content The content of the

Generates sampled or direct likelihood parameter predictions. Parameters ---------- model The Darts `SKL

Generates formatted quantile interval names, optionally added to a component name. Parameters ---------- q_interval A tuple or mu

Compute anomaly score(s) for the given series. Predicts the given target time series with the forecasting model, and applies the scorer(s)

Cast module precision (float32 by default) to another precision.

Truncates a key string to `max_len`, adding '...' if truncated. Parameters ---------- key The key to truncate. max_len

Base class for sklearn wrapper (e.g. `SKLearnModel`) likelihoods. Parameters ---------- likelihood_type A pre-def

PyTorch module implementing the basic building block of the N-BEATS architecture. The blocks produce outputs of size (target_length, nr_param

Applies a transformation over the time dimension of a sequence based on the `PyTorch implementation of TSMixer <https://github.com/ditschuk/py

Global Naive Drift Model. The model generates forecasts for each `series` as described below: - take the slope `m` from each target

Applies the `diff_fn` to two sequences of time series. Converts two time series into 1. Each series-pair in series and pred_series must:

Avoid pitfalls: `encode_train()` or `encode_inference()` can be called multiple times or chained. Exclude any encoded components from `covaria

Extract deterministic values from `series` (quantile=0.5 if `series` is probabilistic).

Converts a frequency to number of days required by Facebook Prophet Parameters ---------- freq frequency string o

Transforms a window-wise anomaly score into a point-wise anomaly score. When using a window of size `W`, a scorer will return an ano

Returns the existing `component_mask` or generates a new component mask based on `columns`.

Get output time index and components based on a aggregation `axis` and potential new column name `new_cname`.

Find the calibration start point (CSP) (for historical forecasts on calibration set). - If `start=None`, the CSP is also `None` (all possible hfc

Returns the reduction function either from user kwargs or metric default. Optionally performs sanity checks for presence of `axis` parameter, and

Convert the `series_idx` to a Sequence[int]. Note: the validity of the entries in series_idx is checked in _get_params().

Checks whether the TimeSeries `ts` is seasonal with period `m` or not. If `m` is None, we work under the assumption that there is a unique s

Computes a loss from a `model_output`, which represents the parameters of a given probability distribution for every ground truth val

Helper function called to retrieve the name of the features and labels arrays created with `create_lagged_data()`. The order of the features

Return a new series with dropped components (columns). Parameters ---------- col_names String or list of strings

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

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

Returns the :class:`TFTExplainabilityResult <darts.explainability.explainability_result.TFTExplainabilityResult>` result for all series in

Fits transformer to a (sequence of) `TimeSeries` by calling the user-implemented `ts_fit` method. The fitted parameters returned by `ts_fit`

Train the model with a specific :class:`darts.utils.data.TorchTrainingDataset` instance. These datasets implement a PyTorch ``Dataset

Returns TimeSeries objects that can be used for testing impact of categorical covariates. Details: - series is a univariate

Last values of the potentially multivariate series. Returns ------- numpy.ndarray The last values of every compon