Multi-head attention forward pass. Args: hidden_states : Input tensor of shape [batch_size, seq_len, d_model] mask :

RNN Module forward. Parameters ---------- x_in Tuple of Tensors containing the features of the input sequence. Th

Parameters ---------- x input tensor of shape batch x (optional) time x categoricals in the order of ``variable_n

BlockRNN Module forward. Parameters ---------- x_in Tuple of Tensors containing the features of the input sequenc

TSMixer model forward pass. Parameters ---------- x_in comes as Tuple `(x_past, x_future, x_static)` where `x_pas

TFT model forward pass. Parameters ---------- x_in comes as tuple `(x_past, x_future, x_static)` where `x_past` i

Forward pass returning quantile predictions shaped ``(batch, time, n_targets, n_quantiles)``. During training with fine-tuning enabled, all 9

TiDE model forward pass. Parameters ---------- x_in comes as tuple `(x_past, x_future, x_static)` where `x_past` i

TimesFM 2.5 model forward pass. Parameters ---------- x_in comes as a tuple `(x_past, x_future, x_static)` where

Same as :meth:`torch.nn.Module.forward`. Parameters ---------- x_in ``(x_past, x_future, x_static)`` the past, fu

Chronos-2 model forward pass. Parameters ---------- x_in comes as a tuple `(x_past, x_future, x_static)` where `x

Run the backbone forward pass (matches ``PatchTSTFMModel.forward``). Returns ------- quantile_predictions Raw (no

PatchTST-FM model forward pass adapted for Darts interface. Parameters ---------- x_in Comes as tuple `(x_past, x

PyTorch-native forward pass. Parameters ---------- x_in comes as tuple `(x_past, x_future, x_static)` where `x_pa

x_in comes as tuple `(x, x_future, x_static)` where `x` is the past target, past covariates and historic future covar

Naive model forward pass. Parameters ---------- x_in comes as tuple `(x_past, x_future, x_static)` where `x_past`

x_in comes as tuple `(x_past, x_future, x_static)` where `x_past` is the input/past chunk and `x_future` is the outpu

The frequency of the series. A ``pandas.DateOffset`` if the series is indexed with a ``pandas.DatetimeIndex``. An integer (step size)

The string representation of the series' frequency.

Returns the future covariates component names generated by `SequentialEncoder.future_encoders`. Only available after calling `SequentialEncode

Returns the future covariates encoders

Returns the future transformer object

generates expected start and end times for the corresponding covariates.

Returns causal mask to apply for self-attention layer.

Returns one or several explanations for a given component. Parameters ---------- component Optionally, t

Returns one or several ``TimeSeries`` representing the explanations for a given horizon and component. Parameters --

Returns the ``TimeSeries`` representing the explanation for a given component. The components of the ``TimeSeries`` correspond to th

Returns the ``TimeSeries`` representing the feature values for a given component. The components of the ``TimeSeries`` correspond to

Returns the underlying ``shap.Explanation`` object for a given component. Parameters ---------- component

Convert a point into a ``pandas.Timestamp`` (if datetime-indexed) or integer (if integer-indexed). Parameters ---------- poin

Whether the series is indexed with a ``pandas.DatetimeIndex`` (otherwise it is indexed with an ``pandas.RangeIndex``).

Whether the series contains a hierarchy.

Whether the series contains metadata.

Whether the series is indexed with an ``pandas.RangeIndex`` (otherwise it is indexed with a ``pandas.DatetimeIndex``).

Whether the series contains static covariates.

The hierarchy of this series. If defined, the hierarchy is given as a dictionary. The keys are the individual components and values are the

Interval Non-Conformity Score for Quantile Regression (INCS_QR). INCS_QR gives the absolute error to the closest predicted quantile interval boun