diff --git a/.github/workflows/merge.yml b/.github/workflows/merge.yml
index d8b5ae732f..87bc82f63b 100644
--- a/.github/workflows/merge.yml
+++ b/.github/workflows/merge.yml
@@ -87,7 +87,7 @@ jobs:
runs-on: ubuntu-latest
strategy:
matrix:
- example-name: [00-quickstart.ipynb, 01-multi-time-series-and-covariates.ipynb, 02-data-processing.ipynb, 03-FFT-examples.ipynb, 04-RNN-examples.ipynb, 05-TCN-examples.ipynb, 06-Transformer-examples.ipynb, 07-NBEATS-examples.ipynb, 08-DeepAR-examples.ipynb, 09-DeepTCN-examples.ipynb, 10-Kalman-filter-examples.ipynb, 11-GP-filter-examples.ipynb, 12-Dynamic-Time-Warping-example.ipynb, 13-TFT-examples.ipynb, 15-static-covariates.ipynb, 16-hierarchical-reconciliation.ipynb, 18-TiDE-examples.ipynb, 19-EnsembleModel-examples.ipynb, 20-RegressionModel-examples.ipynb]
+ example-name: [00-quickstart.ipynb, 01-multi-time-series-and-covariates.ipynb, 02-data-processing.ipynb, 03-FFT-examples.ipynb, 04-RNN-examples.ipynb, 05-TCN-examples.ipynb, 06-Transformer-examples.ipynb, 07-NBEATS-examples.ipynb, 08-DeepAR-examples.ipynb, 09-DeepTCN-examples.ipynb, 10-Kalman-filter-examples.ipynb, 11-GP-filter-examples.ipynb, 12-Dynamic-Time-Warping-example.ipynb, 13-TFT-examples.ipynb, 15-static-covariates.ipynb, 16-hierarchical-reconciliation.ipynb, 18-TiDE-examples.ipynb, 19-EnsembleModel-examples.ipynb, 20-RegressionModel-examples.ipynb, 21-TSMixer-examples.ipynb]
steps:
- name: "1. Clone repository"
uses: actions/checkout@v2
diff --git a/.gitignore b/.gitignore
index 453913f0b7..1e3939db7f 100644
--- a/.gitignore
+++ b/.gitignore
@@ -16,6 +16,7 @@ runs/
htmlcov
coverage.xml
.darts
+darts_logs/
docs_env
.DS_Store
.gradle
diff --git a/CHANGELOG.md b/CHANGELOG.md
index b2e6da8fd9..258086b94e 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -9,14 +9,15 @@ but cannot always guarantee backwards compatibility. Changes that may **break co
### For users of the library:
**Improved**
-- πππ Improvements to metrics, historical forecasts, backtest, and residuals through major refactor. The refactor includes optimization of multiple process and improvemenets to consistency, reliability, and the documentation. Some of these necessary changes come at the cost of breaking changes. [#2284](https://github.com/unit8co/darts/pull/2284) by [Dennis Bader](https://github.com/dennisbader).
+- ππ New forecasting model: `TSMixerModel` as proposed in [this paper](https://arxiv.org/abs/2303.06053). An MLP based model that combines temporal, static and cross-sectional feature information using stacked mixing layers. [#1807](https://https://github.com/unit8co/darts/pull/001), by [Dennis Bader](https://github.com/dennisbader) and [Cristof Rojas](https://github.com/cristof-r).
+- ππ Improvements to metrics, historical forecasts, backtest, and residuals through major refactor. The refactor includes optimization of multiple process and improvemenets to consistency, reliability, and the documentation. Some of these necessary changes come at the cost of breaking changes. [#2284](https://github.com/unit8co/darts/pull/2284) by [Dennis Bader](https://github.com/dennisbader).
- Metrics:
- - Optimized all metrics, which now run >20 times faster than before for univariate series, and >>20 times for multivariate series. This boosts direct metric computations as well as backtesting and residuals computation!
+ - Optimized all metrics, which now run **> n * 20 times faster** than before for series with `n` components/columns. This boosts direct metric computations as well as backtesting and residuals computation!
- Added new metrics:
- Time aggregated metric `merr()` (Mean Error)
- Time aggregated scaled metrics `rmsse()`, and `msse()`: The (Root) Mean Squared Scaled Error.
- "Per time step" metrics that return a metric score per time step: `err()` (Error), `ae()` (Absolute Error), `se()` (Squared Error), `sle()` (Squared Log Error), `ase()` (Absolute Scaled Error), `sse` (Squared Scaled Error), `ape()` (Absolute Percentage Error), `sape()` (symmetric Absolute Percentage Error), `arre()` (Absolute Ranged Relative Error), `ql` (Quantile Loss)
- - All scaled metrics now accept `insample` series that can be overlapping into `pred_series` (before that had to end exactly one step before `pred_series`). Darts will handle the correct time extraction for you.
+ - All scaled metrics now accept `insample` series that can be overlapping into `pred_series` (before they had to end exactly one step before `pred_series`). Darts will handle the correct time extraction for you.
- Improvements to the documentation:
- Added a summary list of all metrics to the [metrics documentation page](https://unit8co.github.io/darts/generated_api/darts.metrics.html)
- Standardized the documentation of each metric (added formula, improved return documentation, ...)
diff --git a/README.md b/README.md
index 1786c968a6..4d482214cc 100644
--- a/README.md
+++ b/README.md
@@ -255,6 +255,7 @@ on bringing more models and features.
| [DLinearModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.dlinear.html#darts.models.forecasting.dlinear.DLinearModel) | [DLinear paper](https://arxiv.org/pdf/2205.13504.pdf) | π© π© | π© π© π© | π© π© | π© |
| [NLinearModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.nlinear.html#darts.models.forecasting.nlinear.NLinearModel) | [NLinear paper](https://arxiv.org/pdf/2205.13504.pdf) | π© π© | π© π© π© | π© π© | π© |
| [TiDEModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.tide_model.html#darts.models.forecasting.tide_model.TiDEModel) | [TiDE paper](https://arxiv.org/pdf/2304.08424.pdf) | π© π© | π© π© π© | π© π© | π© |
+| [TSMixerModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.tsmixer_model.html#darts.models.forecasting.tsmixer_model.TSMixerModel) | [TSMixer paper](https://arxiv.org/pdf/2303.06053.pdf), [PyTorch Implementation](https://github.com/ditschuk/pytorch-tsmixer) | π© π© | π© π© π© | π© π© | π© |
| **Ensemble Models**
([GlobalForecastingModel](https://unit8co.github.io/darts/userguide/covariates.html#global-forecasting-models-gfms)): Model support is dependent on ensembled forecasting models and the ensemble model itself | | | | | |
| [NaiveEnsembleModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.baselines.html#darts.models.forecasting.baselines.NaiveEnsembleModel) | | π© π© | π© π© π© | π© π© | π© |
| [RegressionEnsembleModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.regression_ensemble_model.html#darts.models.forecasting.regression_ensemble_model.RegressionEnsembleModel) | | π© π© | π© π© π© | π© π© | π© |
diff --git a/darts/models/__init__.py b/darts/models/__init__.py
index edcca507ea..3409aaa2ab 100644
--- a/darts/models/__init__.py
+++ b/darts/models/__init__.py
@@ -51,6 +51,7 @@
from darts.models.forecasting.tft_model import TFTModel
from darts.models.forecasting.tide_model import TiDEModel
from darts.models.forecasting.transformer_model import TransformerModel
+ from darts.models.forecasting.tsmixer_model import TSMixerModel
except ModuleNotFoundError:
logger.warning(
"Support for Torch based models not available. "
diff --git a/darts/models/forecasting/__init__.py b/darts/models/forecasting/__init__.py
index 9fa591ca27..37a50aa4bc 100644
--- a/darts/models/forecasting/__init__.py
+++ b/darts/models/forecasting/__init__.py
@@ -46,6 +46,7 @@
- :class:`~darts.models.forecasting.dlinear.DLinearModel`
- :class:`~darts.models.forecasting.nlinear.NLinearModel`
- :class:`~darts.models.forecasting.tide_model.TiDEModel`
+ - :class:`~darts.models.forecasting.tsmixer_model.TSMixerModel`
Ensemble Models (`GlobalForecastingModel `_)
- :class:`~darts.models.forecasting.baselines.NaiveEnsembleModel`
- :class:`~darts.models.forecasting.regression_ensemble_model.RegressionEnsembleModel`
diff --git a/darts/models/forecasting/tsmixer_model.py b/darts/models/forecasting/tsmixer_model.py
new file mode 100644
index 0000000000..0e53080739
--- /dev/null
+++ b/darts/models/forecasting/tsmixer_model.py
@@ -0,0 +1,846 @@
+"""
+Time-Series Mixer (TSMixer)
+---------------------------
+"""
+
+# The inner layers (``nn.Modules``) and the ``TimeBatchNorm2d`` were provided by a PyTorch implementation
+# of TSMixer: https://github.com/ditschuk/pytorch-tsmixer
+#
+# The License of pytorch-tsmixer v0.2.0 from https://github.com/ditschuk/pytorch-tsmixer/blob/main/LICENSE,
+# accessed Thursday, March 21st, 2024:
+# 'The MIT License
+#
+# Copyright 2023 Konstantin Ditschuneit
+#
+# Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
+# associated documentation files (the βSoftwareβ), to deal in the Software without restriction,
+# including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense,
+# and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so,
+# subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in all copies or substantial
+# portions of the Software.
+# '
+
+from typing import Callable, Optional, Tuple, Union
+
+import torch
+from torch import nn
+
+from darts.logging import get_logger, raise_log
+from darts.models.components import layer_norm_variants
+from darts.models.forecasting.pl_forecasting_module import (
+ PLMixedCovariatesModule,
+ io_processor,
+)
+from darts.models.forecasting.torch_forecasting_model import MixedCovariatesTorchModel
+from darts.utils.torch import MonteCarloDropout
+
+MixedCovariatesTrainTensorType = Tuple[
+ torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor
+]
+
+logger = get_logger(__name__)
+
+ACTIVATIONS = [
+ "ReLU",
+ "RReLU",
+ "PReLU",
+ "ELU",
+ "Softplus",
+ "Tanh",
+ "SELU",
+ "LeakyReLU",
+ "Sigmoid",
+ "GELU",
+]
+
+NORMS = [
+ "LayerNorm",
+ "LayerNormNoBias",
+ "TimeBatchNorm2d",
+]
+
+
+def _time_to_feature(x: torch.Tensor) -> torch.Tensor:
+ """Converts a time series Tensor to a feature Tensor."""
+ return x.permute(0, 2, 1)
+
+
+class TimeBatchNorm2d(nn.BatchNorm2d):
+ def __init__(self, *args, **kwargs):
+ """A batch normalization layer that normalizes over the last two dimensions of a Tensor."""
+ super().__init__(num_features=1)
+
+ def forward(self, x: torch.Tensor) -> torch.Tensor:
+ # `x` has shape (batch_size, time, features)
+ if x.ndim != 3:
+ raise_log(
+ ValueError(
+ f"Expected 3D input Tensor, but got {x.ndim}D Tensor" " instead."
+ ),
+ logger=logger,
+ )
+ # apply 2D batch norm over reshape input_data `(batch_size, 1, timepoints, features)`
+ output = super().forward(x.unsqueeze(1))
+ # reshape back to (batch_size, timepoints, features)
+ return output.squeeze(1)
+
+
+class _FeatureMixing(nn.Module):
+ def __init__(
+ self,
+ sequence_length: int,
+ input_dim: int,
+ output_dim: int,
+ ff_size: int,
+ activation: Callable[[torch.Tensor], torch.Tensor],
+ dropout: float,
+ normalize_before: bool,
+ norm_type: nn.Module,
+ ) -> None:
+ """A module for feature mixing with flexibility in normalization and activation based on the
+ `PyTorch implementation of TSMixer `_.
+
+ This module provides options for batch normalization before or after mixing
+ features, uses dropout for regularization, and allows for different activation
+ functions.
+
+ Parameters
+ ----------
+ sequence_length
+ The length of the input sequences.
+ input_dim
+ The number of input channels to the module.
+ output_dim
+ The number of output channels from the module.
+ ff_size
+ The dimension of the feed-forward network internal to the module.
+ activation
+ The activation function used within the feed-forward network.
+ dropout
+ The dropout probability used for regularization.
+ normalize_before
+ A boolean indicating whether to apply normalization before
+ the rest of the operations.
+ norm_type
+ The type of normalization to use.
+ """
+ super().__init__()
+
+ self.projection = (
+ nn.Linear(input_dim, output_dim)
+ if input_dim != output_dim
+ else nn.Identity()
+ )
+ self.norm_before = (
+ norm_type((sequence_length, input_dim))
+ if normalize_before
+ else nn.Identity()
+ )
+ self.fc1 = nn.Linear(input_dim, ff_size)
+ self.activation = activation
+ self.dropout1 = MonteCarloDropout(dropout)
+ self.fc2 = nn.Linear(ff_size, output_dim)
+ self.dropout2 = MonteCarloDropout(dropout)
+ self.norm_after = (
+ norm_type((sequence_length, output_dim))
+ if not normalize_before
+ else nn.Identity()
+ )
+
+ def forward(self, x: torch.Tensor) -> torch.Tensor:
+ x_proj = self.projection(x)
+ x = self.norm_before(x)
+ x = self.fc1(x)
+ x = self.activation(x)
+ x = self.dropout1(x)
+ x = self.fc2(x)
+ x = self.dropout2(x)
+ x = x_proj + x
+ x = self.norm_after(x)
+ return x
+
+
+class _TimeMixing(nn.Module):
+ def __init__(
+ self,
+ sequence_length: int,
+ input_dim: int,
+ activation: Callable,
+ dropout: float,
+ normalize_before: bool,
+ norm_type: nn.Module,
+ ) -> None:
+ """Applies a transformation over the time dimension of a sequence based on the
+ `PyTorch implementation of TSMixer `_.
+
+ This module applies a linear transformation followed by an activation function
+ and dropout over the sequence length of the input feature torch.Tensor after converting
+ feature maps to the time dimension and then back.
+
+ Parameters
+ ----------
+ sequence_length
+ The length of the sequences to be transformed.
+ input_dim
+ The number of input channels to the module.
+ activation
+ The activation function to be used after the linear
+ transformation.
+ dropout
+ The dropout probability to be used after the activation function.
+ normalize_before
+ Whether to apply normalization before or after feature mixing.
+ norm_type
+ The type of normalization to use.
+ """
+ super().__init__()
+ self.normalize_before = normalize_before
+ self.norm_before = (
+ norm_type((sequence_length, input_dim))
+ if normalize_before
+ else nn.Identity()
+ )
+ self.activation = activation
+ self.dropout = MonteCarloDropout(dropout)
+ self.fc1 = nn.Linear(sequence_length, sequence_length)
+ self.norm_after = (
+ norm_type((sequence_length, input_dim))
+ if not normalize_before
+ else nn.Identity()
+ )
+
+ def forward(self, x: torch.Tensor) -> torch.Tensor:
+ # permute the feature dim with the time dim
+ x_temp = self.norm_before(x)
+ x_temp = _time_to_feature(x_temp)
+ x_temp = self.activation(self.fc1(x_temp))
+ x_temp = self.dropout(x_temp)
+ # permute back the time dim with the feature dim
+ x_temp = x + _time_to_feature(x_temp)
+ x_temp = self.norm_after(x_temp)
+ return x_temp
+
+
+class _ConditionalMixerLayer(nn.Module):
+ def __init__(
+ self,
+ sequence_length: int,
+ input_dim: int,
+ output_dim: int,
+ static_cov_dim: int,
+ ff_size: int,
+ activation: Callable,
+ dropout: float,
+ normalize_before: bool,
+ norm_type: nn.Module,
+ ) -> None:
+ """Conditional mix layer combining time and feature mixing with static context based on the
+ `PyTorch implementation of TSMixer `_.
+
+ This module combines time mixing and conditional feature mixing, where the latter
+ is influenced by static features. This allows the module to learn representations
+ that are influenced by both dynamic and static features.
+
+ Parameters
+ ----------
+ sequence_length
+ The length of the input sequences.
+ input_dim
+ The number of input channels of the dynamic features.
+ output_dim
+ The number of output channels after feature mixing.
+ static_cov_dim
+ The number of channels in the static feature input.
+ ff_size
+ The inner dimension of the feedforward network used in feature mixing.
+ activation
+ The activation function used in both mixing operations.
+ dropout
+ The dropout probability used in both mixing operations.
+ normalize_before
+ Whether to apply normalization before or after mixing.
+ norm_type
+ The type of normalization to use.
+ """
+ super().__init__()
+
+ mixing_input = input_dim
+ if static_cov_dim != 0:
+ self.feature_mixing_static = _FeatureMixing(
+ sequence_length=sequence_length,
+ input_dim=static_cov_dim,
+ output_dim=output_dim,
+ ff_size=ff_size,
+ activation=activation,
+ dropout=dropout,
+ normalize_before=normalize_before,
+ norm_type=norm_type,
+ )
+ mixing_input += output_dim
+ else:
+ self.feature_mixing_static = None
+
+ self.time_mixing = _TimeMixing(
+ sequence_length=sequence_length,
+ input_dim=mixing_input,
+ activation=activation,
+ dropout=dropout,
+ normalize_before=normalize_before,
+ norm_type=norm_type,
+ )
+ self.feature_mixing = _FeatureMixing(
+ sequence_length=sequence_length,
+ input_dim=mixing_input,
+ output_dim=output_dim,
+ ff_size=ff_size,
+ activation=activation,
+ dropout=dropout,
+ normalize_before=normalize_before,
+ norm_type=norm_type,
+ )
+
+ def forward(
+ self, x: torch.Tensor, x_static: Optional[torch.Tensor]
+ ) -> torch.Tensor:
+ if self.feature_mixing_static is not None:
+ x_static_mixed = self.feature_mixing_static(x_static)
+ x = torch.cat([x, x_static_mixed], dim=-1)
+ x = self.time_mixing(x)
+ x = self.feature_mixing(x)
+ return x
+
+
+class _TSMixerModule(PLMixedCovariatesModule):
+ def __init__(
+ self,
+ input_dim: int,
+ output_dim: int,
+ past_cov_dim: int,
+ future_cov_dim: int,
+ static_cov_dim: int,
+ nr_params: int,
+ hidden_size: int,
+ ff_size: int,
+ num_blocks: int,
+ activation: str,
+ dropout: float,
+ norm_type: Union[str, nn.Module],
+ normalize_before: bool,
+ **kwargs,
+ ) -> None:
+ """
+ Initializes the TSMixer module for use within a Darts forecasting model.
+
+ Parameters
+ ----------
+ input_dim
+ Number of input target features.
+ output_dim
+ Number of output target features.
+ past_cov_dim
+ Number of past covariate features.
+ future_cov_dim
+ Number of future covariate features.
+ static_cov_dim
+ Number of static covariate features (number of target features
+ (or 1 if global static covariates) * number of static covariate features).
+ nr_params
+ The number of parameters of the likelihood (or 1 if no likelihood is used).
+ hidden_size
+ Hidden state size of the TSMixer.
+ ff_size
+ Dimension of the feedforward network internal to the module.
+ num_blocks
+ Number of mixer blocks.
+ activation
+ Activation function to use.
+ dropout
+ Dropout rate for regularization.
+ norm_type
+ Type of normalization to use.
+ normalize_before
+ Whether to apply normalization before or after mixing.
+ """
+ super().__init__(**kwargs)
+ self.input_dim = input_dim
+ self.output_dim = output_dim
+ self.future_cov_dim = future_cov_dim
+ self.static_cov_dim = static_cov_dim
+ self.nr_params = nr_params
+
+ if activation not in ACTIVATIONS:
+ raise_log(
+ ValueError(
+ f"Invalid `activation={activation}`. Must be on of {ACTIVATIONS}."
+ ),
+ logger=logger,
+ )
+ activation = getattr(nn, activation)()
+
+ if isinstance(norm_type, str):
+ if norm_type not in NORMS:
+ raise_log(
+ ValueError(
+ f"Invalid `norm_type={norm_type}`. Must be on of {NORMS}."
+ ),
+ logger=logger,
+ )
+ if norm_type == "TimeBatchNorm2d":
+ norm_type = TimeBatchNorm2d
+ else:
+ norm_type = getattr(layer_norm_variants, norm_type)
+ else:
+ norm_type = norm_type
+
+ mixer_params = {
+ "ff_size": ff_size,
+ "activation": activation,
+ "dropout": dropout,
+ "norm_type": norm_type,
+ "normalize_before": normalize_before,
+ }
+
+ self.fc_hist = nn.Linear(self.input_chunk_length, self.output_chunk_length)
+ self.feature_mixing_hist = _FeatureMixing(
+ sequence_length=self.output_chunk_length,
+ input_dim=input_dim + past_cov_dim + future_cov_dim,
+ output_dim=hidden_size,
+ **mixer_params,
+ )
+ if future_cov_dim:
+ self.feature_mixing_future = _FeatureMixing(
+ sequence_length=self.output_chunk_length,
+ input_dim=future_cov_dim,
+ output_dim=hidden_size,
+ **mixer_params,
+ )
+ else:
+ self.feature_mixing_future = None
+ self.conditional_mixer = self._build_mixer(
+ prediction_length=self.output_chunk_length,
+ num_blocks=num_blocks,
+ hidden_size=hidden_size,
+ future_cov_dim=future_cov_dim,
+ static_cov_dim=static_cov_dim,
+ **mixer_params,
+ )
+ self.fc_out = nn.Linear(hidden_size, output_dim * nr_params)
+
+ @staticmethod
+ def _build_mixer(
+ prediction_length: int,
+ num_blocks: int,
+ hidden_size: int,
+ future_cov_dim: int,
+ static_cov_dim: int,
+ **kwargs,
+ ) -> nn.ModuleList:
+ """Build the mixer blocks for the model."""
+ # the first block takes `x` consisting of concatenated features with size `hidden_size`:
+ # - historic features
+ # - optional future features
+ input_dim_block = hidden_size * (1 + int(future_cov_dim > 0))
+
+ mixer_layers = nn.ModuleList()
+ for _ in range(num_blocks):
+ layer = _ConditionalMixerLayer(
+ input_dim=input_dim_block,
+ output_dim=hidden_size,
+ sequence_length=prediction_length,
+ static_cov_dim=static_cov_dim,
+ **kwargs,
+ )
+ mixer_layers.append(layer)
+ # after the first block, `x` consists of previous block output with size `hidden_size`
+ input_dim_block = hidden_size
+ return mixer_layers
+
+ @io_processor
+ def forward(
+ self,
+ x_in: Tuple[torch.Tensor, Optional[torch.Tensor], Optional[torch.Tensor]],
+ ) -> torch.Tensor:
+ # x_hist contains the historical time series data and the historical
+ """TSMixer model forward pass.
+
+ Parameters
+ ----------
+ x_in
+ comes as Tuple `(x_past, x_future, x_static)` where `x_past` is the input/past chunk and
+ `x_future` is the output/future chunk. Input dimensions are `(batch_size, time_steps,
+ components)`.
+
+ Returns
+ -------
+ torch.torch.Tensor
+ The output Tensorof shape `(batch_size, output_chunk_length, output_dim, nr_params)`.
+ """
+ # B: batch size
+ # L: input chunk length
+ # T: output chunk length
+ # C: target components
+ # P: past cov features
+ # F: future cov features
+ # S: static cov features
+ # H = C + P + F: historic features
+ # H_S: hidden Size
+ # N_P: likelihood parameters
+
+ # `x`: (B, L, H), `x_future`: (B, T, F), `x_static`: (B, C or 1, S)
+ x, x_future, x_static = x_in
+
+ # swap feature and time dimensions (B, L, H) -> (B, H, L)
+ x = _time_to_feature(x)
+ # linear transformations to horizon (B, H, L) -> (B, H, T)
+ x = self.fc_hist(x)
+ # (B, H, T) -> (B, T, H)
+ x = _time_to_feature(x)
+
+ # feature mixing for historical features (B, T, H) -> (B, T, H_S)
+ x = self.feature_mixing_hist(x)
+ if self.future_cov_dim:
+ # feature mixing for future features (B, T, F) -> (B, T, H_S)
+ x_future = self.feature_mixing_future(x_future)
+ # (B, T, H_S) + (B, T, H_S) -> (B, T, 2*H_S)
+ x = torch.cat([x, x_future], dim=-1)
+
+ if self.static_cov_dim:
+ # (B, C, S) -> (B, 1, C * S)
+ x_static = x_static.reshape(x_static.shape[0], 1, -1)
+ # repeat to match horizon (B, 1, C * S) -> (B, T, C * S)
+ x_static = x_static.repeat(1, self.output_chunk_length, 1)
+
+ for mixing_layer in self.conditional_mixer:
+ # conditional mixer layers with static covariates (B, T, 2 * H_S), (B, T, C * S) -> (B, T, H_S)
+ x = mixing_layer(x, x_static=x_static)
+
+ # linear transformation to generate the forecast (B, T, H_S) -> (B, T, C * N_P)
+ x = self.fc_out(x)
+ # (B, T, C * N_P) -> (B, T, C, N_P)
+ x = x.view(-1, self.output_chunk_length, self.output_dim, self.nr_params)
+ return x
+
+
+class TSMixerModel(MixedCovariatesTorchModel):
+ def __init__(
+ self,
+ input_chunk_length: int,
+ output_chunk_length: int,
+ output_chunk_shift: int = 0,
+ hidden_size: int = 64,
+ ff_size: int = 64,
+ num_blocks: int = 2,
+ activation: str = "ReLU",
+ dropout: float = 0.1,
+ norm_type: Union[str, nn.Module] = "LayerNorm",
+ normalize_before: bool = False,
+ use_static_covariates: bool = True,
+ **kwargs,
+ ) -> None:
+ """Time-Series Mixer (TSMixer): An All-MLP Architecture for Time Series.
+
+ This is an implementation of the TSMixer architecture, as outlined in [1]_. A major part of the architecture
+ was adopted from `this PyTorch implementation `_. Additional
+ changes were applied to increase model performance and efficiency.
+
+ TSMixer forecasts time series data by integrating historical time series data, future known inputs, and static
+ contextual information. It uses a combination of conditional feature mixing and mixer layers to process and
+ combine these different types of data for effective forecasting.
+
+ This model supports past covariates (known for `input_chunk_length` points before prediction time), future
+ covariates (known for `output_chunk_length` points after prediction time), static covariates, as well as
+ probabilistic forecasting.
+
+ Parameters
+ ----------
+ input_chunk_length
+ Number of time steps in the past to take as a model input (per chunk). Applies to the target
+ series, and past and/or future covariates (if the model supports it).
+ Also called: Encoder length
+ output_chunk_length
+ Number of time steps predicted at once (per chunk) by the internal model. Also, the number of future values
+ from future covariates to use as a model input (if the model supports future covariates). It is not the same
+ as forecast horizon `n` used in `predict()`, which is the desired number of prediction points generated
+ using either a one-shot- or autoregressive forecast. Setting `n <= output_chunk_length` prevents
+ auto-regression. This is useful when the covariates don't extend far enough into the future, or to prohibit
+ the model from using future values of past and / or future covariates for prediction (depending on the
+ model's covariate support).
+ Also called: Decoder length
+ output_chunk_shift
+ Optionally, the number of steps to shift the start of the output chunk into the future (relative to the
+ input chunk end). This will create a gap between the input and output. If the model supports
+ `future_covariates`, the future values are extracted from the shifted output chunk. Predictions will start
+ `output_chunk_shift` steps after the end of the target `series`. If `output_chunk_shift` is set, the model
+ cannot generate autoregressive predictions (`n > output_chunk_length`).
+ hidden_size
+ The hidden state size / size of the second feed-forward layer in the feature mixing MLP.
+ ff_size
+ The size of the first feed-forward layer in the feature mixing MLP.
+ num_blocks
+ The number of mixer blocks in the model. The number includes the first block and all subsequent blocks.
+ activation
+ The name of the activation function to use in the mixer layers. Default: `"ReLU"`. Must be one of
+ `"ReLU", "RReLU", "PReLU", "ELU", "Softplus", "Tanh", "SELU", "LeakyReLU", "Sigmoid", "GELU"`.
+ dropout
+ Fraction of neurons affected by dropout. This is compatible with Monte Carlo dropout at inference time
+ for model uncertainty estimation (enabled with ``mc_dropout=True`` at prediction time).
+ norm_type
+ The type of `LayerNorm` variant to use. Default: `"LayerNorm"`. If a string, must be one of
+ `"LayerNormNoBias", "LayerNorm", "TimeBatchNorm2d"`. Otherwise, must be a custom `nn.Module`.
+ normalize_before
+ Whether to apply layer normalization before or after mixer layer.
+ use_static_covariates
+ Whether the model should use static covariate information in case the input `series` passed to ``fit()``
+ contain static covariates. If ``True``, and static covariates are available at fitting time, will enforce
+ that all target `series` have the same static covariate dimensionality in ``fit()`` and ``predict()``.
+ **kwargs
+ Optional arguments to initialize the pytorch_lightning.Module, pytorch_lightning.Trainer, and
+ Darts' :class:`TorchForecastingModel`.
+
+ loss_fn
+ PyTorch loss function used for training. By default, the TFT
+ model is probabilistic and uses a ``likelihood`` instead
+ (``QuantileRegression``). To make the model deterministic, you
+ can set the ``likelihood`` to None and give a ``loss_fn``
+ argument.
+ likelihood
+ The likelihood model to be used for probabilistic forecasts.
+ torch_metrics
+ A torch metric or a ``MetricCollection`` used for evaluation. A full list of available metrics can be found
+ at https://torchmetrics.readthedocs.io/en/latest/. Default: ``None``.
+ optimizer_cls
+ The PyTorch optimizer class to be used. Default: ``torch.optim.Adam``.
+ optimizer_kwargs
+ Optionally, some keyword arguments for the PyTorch optimizer (e.g., ``{'lr': 1e-3}``
+ for specifying a learning rate). Otherwise, the default values of the selected ``optimizer_cls``
+ will be used. Default: ``None``.
+ lr_scheduler_cls
+ Optionally, the PyTorch learning rate scheduler class to be used. Specifying ``None`` corresponds
+ to using a constant learning rate. Default: ``None``.
+ lr_scheduler_kwargs
+ Optionally, some keyword arguments for the PyTorch learning rate scheduler. Default: ``None``.
+ use_reversible_instance_norm
+ Whether to use reversible instance normalization `RINorm` against distribution shift as shown in [3]_.
+ It is only applied to the features of the target series and not the covariates.
+ batch_size
+ Number of time series (input and output sequences) used in each training pass. Default: ``32``.
+ n_epochs
+ Number of epochs over which to train the model. Default: ``100``.
+ model_name
+ Name of the model. Used for creating checkpoints and saving torch.Tensorboard data. If not specified,
+ defaults to the following string ``"YYYY-mm-dd_HH_MM_SS_torch_model_run_PID"``, where the initial part
+ of the name is formatted with the local date and time, while PID is the processed ID (preventing models
+ spawned at the same time by different processes to share the same model_name). E.g.,
+ ``"2021-06-14_09_53_32_torch_model_run_44607"``.
+ work_dir
+ Path of the working directory, where to save checkpoints and torch.Tensorboard summaries.
+ Default: current working directory.
+ log_torch.Tensorboard
+ If set, use torch.Tensorboard to log the different parameters. The logs will be located in:
+ ``"{work_dir}/darts_logs/{model_name}/logs/"``. Default: ``False``.
+ nr_epochs_val_period
+ Number of epochs to wait before evaluating the validation loss (if a validation
+ ``TimeSeries`` is passed to the :func:`fit()` method). Default: ``1``.
+ force_reset
+ If set to ``True``, any previously-existing model with the same name will be reset (all checkpoints will
+ be discarded). Default: ``False``.
+ save_checkpoints
+ Whether to automatically save the untrained model and checkpoints from training.
+ To load the model from checkpoint, call :func:`MyModelClass.load_from_checkpoint()`, where
+ :class:`MyModelClass` is the :class:`TorchForecastingModel` class that was used (such as :class:`TFTModel`,
+ :class:`NBEATSModel`, etc.). If set to ``False``, the model can still be manually saved using
+ :func:`save()` and loaded using :func:`load()`. Default: ``False``.
+ add_encoders
+ A large number of past and future covariates can be automatically generated with `add_encoders`.
+ This can be done by adding multiple pre-defined index encoders and/or custom user-made functions that
+ will be used as index encoders. Additionally, a transformer such as Darts' :class:`Scaler` can be added to
+ transform the generated covariates. This happens all under one hood and only needs to be specified at
+ model creation.
+ Read :meth:`SequentialEncoder ` to find out more about
+ ``add_encoders``. Default: ``None``. An example showing some of ``add_encoders`` features:
+
+ .. highlight:: python
+ .. code-block:: python
+
+ def encode_year(idx):
+ return (idx.year - 1950) / 50
+
+ add_encoders={
+ 'cyclic': {'future': ['month']},
+ 'datetime_attribute': {'future': ['hour', 'dayofweek']},
+ 'position': {'past': ['relative'], 'future': ['relative']},
+ 'custom': {'past': [encode_year]},
+ 'transformer': Scaler(),
+ 'tz': 'CET'
+ }
+ ..
+ random_state
+ Control the randomness of the weight's initialization. Check this
+ `link `_ for more details.
+ Default: ``None``.
+ pl_trainer_kwargs
+ By default :class:`TorchForecastingModel` creates a PyTorch Lightning Trainer with several useful presets
+ that performs the training, validation and prediction processes. These presets include automatic
+ checkpointing, torch.Tensorboard logging, setting the torch device and more.
+ With ``pl_trainer_kwargs`` you can add additional kwargs to instantiate the PyTorch Lightning trainer
+ object. Check the `PL Trainer documentation
+ `_ for more information about the
+ supported kwargs. Default: ``None``.
+ Running on GPU(s) is also possible using ``pl_trainer_kwargs`` by specifying keys ``"accelerator",
+ "devices", and "auto_select_gpus"``. Some examples for setting the devices inside the ``pl_trainer_kwargs``
+ dict:
+
+ - ``{"accelerator": "cpu"}`` for CPU,
+ - ``{"accelerator": "gpu", "devices": [i]}`` to use only GPU ``i`` (``i`` must be an integer),
+ - ``{"accelerator": "gpu", "devices": -1, "auto_select_gpus": True}`` to use all available GPUS.
+
+ For more info, see here:
+ https://pytorch-lightning.readthedocs.io/en/stable/common/trainer.html#trainer-flags , and
+ https://pytorch-lightning.readthedocs.io/en/stable/accelerators/gpu_basic.html#train-on-multiple-gpus
+
+ With parameter ``"callbacks"`` you can add custom or PyTorch-Lightning built-in callbacks to Darts'
+ :class:`TorchForecastingModel`. Below is an example for adding EarlyStopping to the training process.
+ The model will stop training early if the validation loss `val_loss` does not improve beyond
+ specifications. For more information on callbacks, visit:
+ `PyTorch Lightning Callbacks
+ `_
+
+ .. highlight:: python
+ .. code-block:: python
+
+ from pytorch_lightning.callbacks.early_stopping import EarlyStopping
+
+ # stop training when validation loss does not decrease more than 0.05 (`min_delta`) over
+ # a period of 5 epochs (`patience`)
+ my_stopper = EarlyStopping(
+ monitor="val_loss",
+ patience=5,
+ min_delta=0.05,
+ mode='min',
+ )
+
+ pl_trainer_kwargs={"callbacks": [my_stopper]}
+ ..
+
+ Note that you can also use a custom PyTorch Lightning Trainer for training and prediction with optional
+ parameter ``trainer`` in :func:`fit()` and :func:`predict()`.
+ show_warnings
+ whether to show warnings raised from PyTorch Lightning. Useful to detect potential issues of
+ your forecasting use case. Default: ``False``.
+
+ References
+ ----------
+ .. [1] https://arxiv.org/abs/2303.06053
+
+ Examples
+ --------
+ >>> from darts.datasets import WeatherDataset
+ >>> from darts.models import TSMixerModel
+ >>> series = WeatherDataset().load()
+ >>> # predicting temperatures
+ >>> target = series['T (degC)'][:100]
+ >>> # optionally, use past observed rainfall (pretending to be unknown beyond index 100)
+ >>> past_cov = series['rain (mm)'][:100]
+ >>> # optionally, use future atmospheric pressure (pretending this component is a forecast)
+ >>> future_cov = series['p (mbar)'][:106]
+ >>> model = TSMixerModel(
+ >>> input_chunk_length=6,
+ >>> output_chunk_length=6,
+ >>> use_reversible_instance_norm=True,
+ >>> n_epochs=20
+ >>> )
+ >>> model.fit(target, past_covariates=past_cov, future_covariates=future_cov)
+ >>> pred = model.predict(6)
+ >>> pred.values()
+ array([[3.92519848],
+ [4.05650312],
+ [4.21781987],
+ [4.29394973],
+ [4.4122863 ],
+ [4.42762751]])
+ """
+ model_kwargs = {key: val for key, val in self.model_params.items()}
+ super().__init__(**self._extract_torch_model_params(**model_kwargs))
+
+ # extract pytorch lightning module kwargs
+ self.pl_module_params = self._extract_pl_module_params(**model_kwargs)
+
+ # Model specific parameters
+ self.ff_size = ff_size
+ self.dropout = dropout
+ self.num_blocks = num_blocks
+ self.activation = activation
+ self.normalize_before = normalize_before
+ self.norm_type = norm_type
+ self.hidden_size = hidden_size
+ self._considers_static_covariates = use_static_covariates
+
+ def _create_model(self, train_sample: MixedCovariatesTrainTensorType) -> nn.Module:
+ """
+ Parameters
+ ----------
+ train_sample
+ contains the following torch.Tensors: `(past_target, past_covariates, historic_future_covariates,
+ future_covariates, static_covariates, future_target)`:
+
+ - past/historic torch.Tensors have shape (input_chunk_length, n_variables)
+ - future torch.Tensors have shape (output_chunk_length, n_variables)
+ - static covariates have shape (component, static variable)
+ """
+ (
+ past_target,
+ past_covariates,
+ historic_future_covariates,
+ future_covariates,
+ static_covariates,
+ future_target,
+ ) = train_sample
+
+ input_dim = past_target.shape[1]
+ output_dim = future_target.shape[1]
+
+ static_cov_dim = (
+ static_covariates.shape[0] * static_covariates.shape[1]
+ if static_covariates is not None
+ else 0
+ )
+ future_cov_dim = (
+ future_covariates.shape[1] if future_covariates is not None else 0
+ )
+ past_cov_dim = past_covariates.shape[1] if past_covariates is not None else 0
+ nr_params = 1 if self.likelihood is None else self.likelihood.num_parameters
+
+ return _TSMixerModule(
+ input_dim=input_dim,
+ output_dim=output_dim,
+ future_cov_dim=future_cov_dim,
+ past_cov_dim=past_cov_dim,
+ static_cov_dim=static_cov_dim,
+ nr_params=nr_params,
+ hidden_size=self.hidden_size,
+ ff_size=self.ff_size,
+ num_blocks=self.num_blocks,
+ activation=self.activation,
+ dropout=self.dropout,
+ norm_type=self.norm_type,
+ normalize_before=self.normalize_before,
+ **self.pl_module_params,
+ )
+
+ @property
+ def supports_multivariate(self) -> bool:
+ return True
+
+ @property
+ def supports_static_covariates(self) -> bool:
+ return True
+
+ @property
+ def supports_future_covariates(self) -> bool:
+ return True
+
+ @property
+ def supports_past_covariates(self) -> bool:
+ return True
diff --git a/darts/tests/models/forecasting/test_global_forecasting_models.py b/darts/tests/models/forecasting/test_global_forecasting_models.py
index b8b020f342..dd3e6faf8d 100644
--- a/darts/tests/models/forecasting/test_global_forecasting_models.py
+++ b/darts/tests/models/forecasting/test_global_forecasting_models.py
@@ -33,6 +33,7 @@
TFTModel,
TiDEModel,
TransformerModel,
+ TSMixerModel,
)
from darts.models.forecasting.torch_forecasting_model import (
DualCovariatesTorchModel,
@@ -155,6 +156,14 @@
},
40.0,
),
+ (
+ TSMixerModel,
+ {
+ "n_epochs": 10,
+ "pl_trainer_kwargs": tfm_kwargs["pl_trainer_kwargs"],
+ },
+ 60.0,
+ ),
(
GlobalNaiveAggregate,
{
@@ -527,7 +536,7 @@ def test_future_covariates(self):
@pytest.mark.parametrize(
"model_cls,ts",
product(
- [TFTModel, DLinearModel, NLinearModel, TiDEModel],
+ [TFTModel, DLinearModel, NLinearModel, TiDEModel, TSMixerModel],
[ts_w_static_cov, ts_shared_static_cov, ts_comps_static_cov],
),
)
diff --git a/darts/tests/models/forecasting/test_historical_forecasts.py b/darts/tests/models/forecasting/test_historical_forecasts.py
index 236933b714..e92eedffdc 100644
--- a/darts/tests/models/forecasting/test_historical_forecasts.py
+++ b/darts/tests/models/forecasting/test_historical_forecasts.py
@@ -38,6 +38,7 @@
TFTModel,
TiDEModel,
TransformerModel,
+ TSMixerModel,
)
from darts.utils.likelihood_models import GaussianLikelihood, QuantileRegression
@@ -235,6 +236,17 @@
(IN_LEN, OUT_LEN),
"MixedCovariates",
),
+ (
+ TSMixerModel,
+ {
+ "input_chunk_length": IN_LEN,
+ "output_chunk_length": OUT_LEN,
+ "n_epochs": NB_EPOCH,
+ **tfm_kwargs,
+ },
+ (IN_LEN, OUT_LEN),
+ "MixedCovariates",
+ ),
(
GlobalNaiveAggregate,
{
diff --git a/darts/tests/models/forecasting/test_probabilistic_models.py b/darts/tests/models/forecasting/test_probabilistic_models.py
index 7b728efcbb..169f9b6a1e 100644
--- a/darts/tests/models/forecasting/test_probabilistic_models.py
+++ b/darts/tests/models/forecasting/test_probabilistic_models.py
@@ -35,6 +35,7 @@
TFTModel,
TiDEModel,
TransformerModel,
+ TSMixerModel,
)
from darts.models.forecasting.torch_forecasting_model import TorchForecastingModel
from darts.utils.likelihood_models import (
@@ -194,6 +195,24 @@
0.06,
0.1,
),
+ (
+ TSMixerModel,
+ {
+ "input_chunk_length": 10,
+ "output_chunk_length": 5,
+ "n_epochs": 100,
+ "random_state": 0,
+ "num_blocks": 1,
+ "hidden_size": 32,
+ "dropout": 0.2,
+ "ff_size": 32,
+ "batch_size": 8,
+ "likelihood": GaussianLikelihood(),
+ **tfm_kwargs,
+ },
+ 0.06,
+ 0.1,
+ ),
]
diff --git a/darts/tests/models/forecasting/test_torch_forecasting_model.py b/darts/tests/models/forecasting/test_torch_forecasting_model.py
index 73ec9bb19b..0e04f821b8 100644
--- a/darts/tests/models/forecasting/test_torch_forecasting_model.py
+++ b/darts/tests/models/forecasting/test_torch_forecasting_model.py
@@ -41,6 +41,7 @@
TFTModel,
TiDEModel,
TransformerModel,
+ TSMixerModel,
)
from darts.models.components.layer_norm_variants import RINorm
from darts.utils.likelihood_models import (
@@ -66,6 +67,7 @@
(TFTModel, {"add_relative_index": 2, **kwargs}),
(TiDEModel, kwargs),
(TransformerModel, kwargs),
+ (TSMixerModel, kwargs),
(GlobalNaiveSeasonal, kwargs),
(GlobalNaiveAggregate, kwargs),
(GlobalNaiveDrift, kwargs),
@@ -1505,6 +1507,7 @@ def test_rin(self, model_config):
(NHiTSModel, {}),
(TransformerModel, {}),
(TCNModel, {}),
+ (TSMixerModel, {}),
(BlockRNNModel, {}),
(GlobalNaiveSeasonal, {}),
(GlobalNaiveAggregate, {}),
diff --git a/darts/tests/models/forecasting/test_tsmixer.py b/darts/tests/models/forecasting/test_tsmixer.py
new file mode 100644
index 0000000000..6ae3abe39e
--- /dev/null
+++ b/darts/tests/models/forecasting/test_tsmixer.py
@@ -0,0 +1,371 @@
+from darts.logging import get_logger
+
+logger = get_logger(__name__)
+
+try:
+ import numpy as np
+ import pandas as pd
+ import pytest
+ import torch
+ from torch import nn
+
+ from darts import concatenate
+ from darts.models.forecasting.tsmixer_model import TimeBatchNorm2d, TSMixerModel
+ from darts.tests.conftest import tfm_kwargs
+ from darts.utils import timeseries_generation as tg
+ from darts.utils.likelihood_models import GaussianLikelihood
+
+ TORCH_AVAILABLE = True
+
+except ImportError:
+ logger.warning("Torch not available. TSMixerModel tests will be skipped.")
+ TORCH_AVAILABLE = False
+
+
+@pytest.mark.skipif(
+ TORCH_AVAILABLE is False,
+ reason="Torch not available. TSMixerModel tests will be skipped.",
+)
+class TestTSMixerModel:
+ np.random.seed(42)
+ torch.manual_seed(42)
+
+ def test_creation(self):
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ likelihood=GaussianLikelihood(),
+ )
+
+ assert model.input_chunk_length == 1
+
+ def test_fit(self):
+ large_ts = tg.constant_timeseries(length=10, value=1.0)
+ small_ts = tg.constant_timeseries(length=10, value=0.1)
+
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ n_epochs=10,
+ random_state=42,
+ **tfm_kwargs,
+ )
+
+ model.fit(large_ts)
+ pred = model.predict(n=2).values()[0]
+
+ # Test whether model trained on one series is better
+ # than one trained on another
+ model2 = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ n_epochs=10,
+ random_state=42,
+ **tfm_kwargs,
+ )
+
+ model2.fit(small_ts)
+ pred2 = model2.predict(n=2).values()[0]
+ assert abs(pred2 - 0.1) < abs(pred - 0.1)
+
+ # test short predict
+ pred3 = model2.predict(n=1)
+ assert len(pred3) == 1
+
+ def test_likelihood_fit(self):
+ ts = tg.constant_timeseries(length=3)
+
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ n_epochs=1,
+ random_state=42,
+ likelihood=GaussianLikelihood(),
+ **tfm_kwargs,
+ )
+ model.fit(ts)
+ # sampled from distribution
+ pred = model.predict(n=1, num_samples=20)
+ assert pred.n_samples == 20
+
+ # direct distribution parameter prediction
+ pred = model.predict(n=1, num_samples=1, predict_likelihood_parameters=True)
+ assert pred.n_components == 2
+ assert pred.n_samples == 1
+
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ n_epochs=1,
+ random_state=42,
+ **tfm_kwargs,
+ )
+ model.fit(ts)
+ # mc dropout
+ pred = model.predict(n=1, mc_dropout=True, num_samples=10)
+ assert pred.n_samples == 10
+
+ def test_logtensorboard(self, tmpdir_module):
+ ts = tg.constant_timeseries(length=4)
+
+ # Test basic fit and predict
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ n_epochs=1,
+ log_tensorboard=True,
+ batch_size=2,
+ work_dir=tmpdir_module,
+ pl_trainer_kwargs={
+ "log_every_n_steps": 1,
+ **tfm_kwargs["pl_trainer_kwargs"],
+ },
+ )
+ model.fit(ts)
+ _ = model.predict(n=2)
+
+ def test_static_covariates_support(self):
+ target_multi = concatenate(
+ [tg.sine_timeseries(length=10, freq="h")] * 2, axis=1
+ )
+
+ target_multi = target_multi.with_static_covariates(
+ pd.DataFrame(
+ [[0.0, 1.0, 0, 2], [2.0, 3.0, 1, 3]],
+ columns=["st1", "st2", "cat1", "cat2"],
+ )
+ )
+
+ # should work with cyclic encoding for time index
+ model = TSMixerModel(
+ input_chunk_length=3,
+ output_chunk_length=4,
+ add_encoders={"cyclic": {"future": "hour", "past": "hour"}},
+ pl_trainer_kwargs={
+ "fast_dev_run": True,
+ **tfm_kwargs["pl_trainer_kwargs"],
+ },
+ )
+ model.fit(target_multi, verbose=False)
+
+ assert model.model.static_cov_dim == np.prod(
+ target_multi.static_covariates.values.shape
+ )
+
+ # raise an error when trained with static covariates of wrong dimensionality
+ target_multi = target_multi.with_static_covariates(
+ pd.concat([target_multi.static_covariates] * 2, axis=1)
+ )
+ with pytest.raises(ValueError):
+ model.predict(n=1, series=target_multi, verbose=False)
+
+ # raise an error when trained with static covariates and trying to predict without
+ with pytest.raises(ValueError):
+ model.predict(
+ n=1, series=target_multi.with_static_covariates(None), verbose=False
+ )
+
+ # with `use_static_covariates=False`, we can predict without static covs
+ model = TSMixerModel(
+ input_chunk_length=3,
+ output_chunk_length=4,
+ use_static_covariates=False,
+ n_epochs=1,
+ **tfm_kwargs,
+ )
+ model.fit(target_multi)
+ preds = model.predict(n=2, series=target_multi.with_static_covariates(None))
+ assert preds.static_covariates is None
+
+ model = TSMixerModel(
+ input_chunk_length=3,
+ output_chunk_length=4,
+ use_static_covariates=False,
+ n_epochs=1,
+ **tfm_kwargs,
+ )
+ model.fit(target_multi.with_static_covariates(None))
+ preds = model.predict(n=2, series=target_multi)
+ assert preds.static_covariates.equals(target_multi.static_covariates)
+
+ @pytest.mark.parametrize("enable_rin", [True, False])
+ def test_future_covariate_handling(self, enable_rin):
+ ts_time_index = tg.sine_timeseries(length=2, freq="h")
+
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ add_encoders={"cyclic": {"future": "hour"}},
+ use_reversible_instance_norm=enable_rin,
+ **tfm_kwargs,
+ )
+ model.fit(ts_time_index, verbose=False, epochs=1)
+
+ def test_past_covariate_handling(self):
+ ts_time_index = tg.sine_timeseries(length=2, freq="h")
+
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ add_encoders={"cyclic": {"past": "hour"}},
+ **tfm_kwargs,
+ )
+ model.fit(ts_time_index, verbose=False, epochs=1)
+
+ def test_future_and_past_covariate_handling(self):
+ ts_time_index = tg.sine_timeseries(length=2, freq="h")
+
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ add_encoders={"cyclic": {"future": "hour", "past": "hour"}},
+ **tfm_kwargs,
+ )
+ model.fit(ts_time_index, verbose=False, epochs=1)
+
+ def test_future_past_and_static_covariate_as_timeseries_handling(self):
+ ts_time_index = tg.sine_timeseries(length=2, freq="h")
+ ts_time_index = ts_time_index.with_static_covariates(
+ pd.DataFrame(
+ [
+ [
+ 0.0,
+ ]
+ ],
+ columns=["st1"],
+ )
+ )
+ for enable_rin in [True, False]:
+ # test with past_covariates timeseries
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ add_encoders={"cyclic": {"future": "hour"}},
+ use_reversible_instance_norm=enable_rin,
+ **tfm_kwargs,
+ )
+ model.fit(
+ ts_time_index,
+ past_covariates=ts_time_index,
+ verbose=False,
+ epochs=1,
+ )
+
+ # test with past_covariates and future_covariates timeseries
+ model = TSMixerModel(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ add_encoders={"cyclic": {"future": "hour", "past": "hour"}},
+ use_reversible_instance_norm=enable_rin,
+ **tfm_kwargs,
+ )
+ model.fit(
+ ts_time_index,
+ past_covariates=ts_time_index,
+ future_covariates=ts_time_index,
+ verbose=False,
+ epochs=1,
+ )
+
+ @pytest.mark.parametrize(
+ "norm_type, expect_exception",
+ [
+ ("LayerNorm", False),
+ ("LayerNormNoBias", False),
+ (nn.LayerNorm, False),
+ ("TimeBatchNorm2d", False),
+ ("invalid", True),
+ ],
+ )
+ def test_layer_norms_with_parametrization(self, norm_type, expect_exception):
+ series = tg.sine_timeseries(length=3)
+ base_model = TSMixerModel
+
+ if expect_exception:
+ with pytest.raises(ValueError):
+ model = base_model(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ norm_type=norm_type,
+ **tfm_kwargs,
+ )
+ model.fit(series, epochs=1)
+ else:
+ model = base_model(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ norm_type=norm_type,
+ **tfm_kwargs,
+ )
+ model.fit(series, epochs=1)
+
+ @pytest.mark.parametrize(
+ "activation, expect_error",
+ [
+ ("ReLU", False),
+ ("RReLU", False),
+ ("PReLU", False),
+ ("ELU", False),
+ ("Softplus", False),
+ ("Tanh", False),
+ ("SELU", False),
+ ("LeakyReLU", False),
+ ("Sigmoid", False),
+ ("invalid", True),
+ ],
+ )
+ def test_activation_functions(self, activation, expect_error):
+ series = tg.sine_timeseries(length=3)
+ base_model = TSMixerModel
+
+ if expect_error:
+ with pytest.raises(ValueError):
+ model = base_model(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ activation=activation,
+ **tfm_kwargs,
+ )
+ model.fit(series, epochs=1)
+ else:
+ model = base_model(
+ input_chunk_length=1,
+ output_chunk_length=1,
+ activation=activation,
+ **tfm_kwargs,
+ )
+ model.fit(series, epochs=1)
+
+ def test_time_batch_norm_3d(self):
+ torch.manual_seed(0)
+
+ layer = TimeBatchNorm2d()
+ # 4D does not work
+ with pytest.raises(ValueError):
+ layer.forward(torch.randn(3, 3, 3, 3))
+
+ # 2D does not work
+ with pytest.raises(ValueError):
+ layer.forward(torch.randn(3, 3))
+
+ # 3D works
+ norm = layer.forward(torch.randn(3, 3, 3)).detach()
+ assert norm.mean().numpy() == pytest.approx(0.0, abs=0.1)
+ assert norm.std().numpy() == pytest.approx(1.0, abs=0.1)
+
+ @pytest.mark.parametrize("batch_size", [1, 2, 5, 10])
+ def test_time_batch_norm_2d_different_batch_sizes(self, batch_size):
+ layer = TimeBatchNorm2d()
+ input_tensor = torch.randn(batch_size, 3, 3)
+ output = layer.forward(input_tensor)
+ assert output.shape == input_tensor.shape
+
+ def test_time_batch_norm_2d_gradients(self):
+ normalized_shape = (10, 32)
+ layer = TimeBatchNorm2d(normalized_shape)
+ input_tensor = torch.randn(5, 10, 32, requires_grad=True)
+
+ output = layer.forward(input_tensor)
+ output.mean().backward()
+
+ assert input_tensor.grad is not None
diff --git a/docs/source/examples.rst b/docs/source/examples.rst
index 72b2557920..9fd96c177a 100644
--- a/docs/source/examples.rst
+++ b/docs/source/examples.rst
@@ -177,6 +177,16 @@ TiDE model example notebook:
examples/18-TiDE-examples.ipynb
+TimeSeries Mixer (TSMixer) Model
+=======================================
+
+TSMixer model example notebook:
+
+.. toctree::
+ :maxdepth: 1
+
+ 21-TSMixer-examples.ipynb
+
Ensemble Models
=============================
diff --git a/docs/userguide/covariates.md b/docs/userguide/covariates.md
index d9ec6cc72e..cc4c564b87 100644
--- a/docs/userguide/covariates.md
+++ b/docs/userguide/covariates.md
@@ -152,6 +152,7 @@ GFMs are models that can be trained on multiple target (and covariate) time seri
| [DLinearModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.dlinear.html#darts.models.forecasting.dlinear.DLinearModel) | β
| β
| β
|
| [NLinearModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.nlinear.html#darts.models.forecasting.nlinear.NLinearModel) | β
| β
| β
|
| [TiDEModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.tide_model.html#darts.models.forecasting.tide_model.TiDEModel) | β
| β
| β
|
+| [TSMixerModel](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.tsmixer_model.html#darts.models.forecasting.tsmixer_model.TSMixerModel) | β
| β
| β
|
| Ensemble Models (f) | β
| β
| β
|
**Table 1: Darts' forecasting models and their covariate support**
diff --git a/docs/userguide/torch_forecasting_models.md b/docs/userguide/torch_forecasting_models.md
index 0c2ba84fde..662bc4bc66 100644
--- a/docs/userguide/torch_forecasting_models.md
+++ b/docs/userguide/torch_forecasting_models.md
@@ -116,6 +116,7 @@ Each Torch Forecasting Model inherits from one `{X}CovariatesModel` (covariate c
| `NLinearModel` | | | | | β
|
| `DLinearModel` | | | | | β
|
| `TiDEModel` | | | | | β
|
+| `TSMixerModel` | | | | | β
|
**Table 2: Darts' Torch Forecasting Model covariate support**
diff --git a/examples/21-TSMixer-examples.ipynb b/examples/21-TSMixer-examples.ipynb
new file mode 100644
index 0000000000..1d1735f909
--- /dev/null
+++ b/examples/21-TSMixer-examples.ipynb
@@ -0,0 +1,1025 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Time Series Mixer (TSMixer)\n",
+ "This notebook walks through how to use Darts' `TSMixerModel` and benchmarks it against `TiDEModel`.\n",
+ "\n",
+ "TSMixer (Time-series Mixer) is an all-MLP architecture for time series forecasting. \n",
+ "\n",
+ "It does so by integrating historical time series data, future known inputs, and static contextual information. The architecture uses a combination of conditional feature mixing and mixer layers to process and combine these different types of data for effective forecasting.\n",
+ "\n",
+ "Translated to Darts, this model supports all types of covariates (past, future, and/or static).\n",
+ "\n",
+ "See the original paper and model description [here](https://arxiv.org/abs/2303.06053).\n",
+ "\n",
+ "According to the authors, the model outperforms several state-of-the-art models on multivariate forecasting tasks.\n",
+ "\n",
+ "Let's see how it performs against `TideModel` on the ETTh1 and ETTh2 datasets."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# fix python path if working locally\n",
+ "from utils import fix_pythonpath_if_working_locally\n",
+ "\n",
+ "fix_pythonpath_if_working_locally()\n",
+ "%matplotlib inline"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "%load_ext autoreload\n",
+ "%autoreload 2\n",
+ "%matplotlib inline"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import warnings\n",
+ "\n",
+ "warnings.filterwarnings(\"ignore\")\n",
+ "import logging\n",
+ "\n",
+ "logging.disable(logging.CRITICAL)\n",
+ "\n",
+ "import torch\n",
+ "import matplotlib.pyplot as plt\n",
+ "import numpy as np\n",
+ "import pandas as pd\n",
+ "from pytorch_lightning.callbacks.early_stopping import EarlyStopping\n",
+ "\n",
+ "from darts import concatenate\n",
+ "from darts.dataprocessing.transformers.scaler import Scaler\n",
+ "from darts.datasets import ETTh1Dataset, ETTh2Dataset\n",
+ "from darts.metrics import mae, mse, mql\n",
+ "from darts.models import TiDEModel, TSMixerModel\n",
+ "from darts.utils.likelihood_models import QuantileRegression\n",
+ "from darts.utils.callbacks import TFMProgressBar"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Data Loading and preparation\n",
+ "We consider the ETTh1 and ETTh2 datasets which contain hourly multivariate data of an electricity transformer (load, oil temperature, ...).\n",
+ "You can find more information [here](https://unit8co.github.io/darts/generated_api/darts.datasets.html#darts.datasets.ETTh1Dataset).\n",
+ "\n",
+ "We will add static information to each transformer time series, that identifies whether it is the `ETTh1` or `ETTh2` transformer.\n",
+ "Both TSMixer and TiDE can levarage this information."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " component | \n",
+ " HUFL | \n",
+ " HULL | \n",
+ " MUFL | \n",
+ " MULL | \n",
+ " LUFL | \n",
+ " LULL | \n",
+ " OT | \n",
+ "
\n",
+ " \n",
+ " date | \n",
+ " | \n",
+ " | \n",
+ " | \n",
+ " | \n",
+ " | \n",
+ " | \n",
+ " | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " 2016-07-01 00:00:00 | \n",
+ " 5.827 | \n",
+ " 2.009 | \n",
+ " 1.599 | \n",
+ " 0.462 | \n",
+ " 4.203 | \n",
+ " 1.340 | \n",
+ " 30.531000 | \n",
+ "
\n",
+ " \n",
+ " 2016-07-01 01:00:00 | \n",
+ " 5.693 | \n",
+ " 2.076 | \n",
+ " 1.492 | \n",
+ " 0.426 | \n",
+ " 4.142 | \n",
+ " 1.371 | \n",
+ " 27.787001 | \n",
+ "
\n",
+ " \n",
+ " 2016-07-01 02:00:00 | \n",
+ " 5.157 | \n",
+ " 1.741 | \n",
+ " 1.279 | \n",
+ " 0.355 | \n",
+ " 3.777 | \n",
+ " 1.218 | \n",
+ " 27.787001 | \n",
+ "
\n",
+ " \n",
+ " 2016-07-01 03:00:00 | \n",
+ " 5.090 | \n",
+ " 1.942 | \n",
+ " 1.279 | \n",
+ " 0.391 | \n",
+ " 3.807 | \n",
+ " 1.279 | \n",
+ " 25.044001 | \n",
+ "
\n",
+ " \n",
+ " 2016-07-01 04:00:00 | \n",
+ " 5.358 | \n",
+ " 1.942 | \n",
+ " 1.492 | \n",
+ " 0.462 | \n",
+ " 3.868 | \n",
+ " 1.279 | \n",
+ " 21.948000 | \n",
+ "
\n",
+ " \n",
+ " ... | \n",
+ " ... | \n",
+ " ... | \n",
+ " ... | \n",
+ " ... | \n",
+ " ... | \n",
+ " ... | \n",
+ " ... | \n",
+ "
\n",
+ " \n",
+ " 2018-06-26 15:00:00 | \n",
+ " -1.674 | \n",
+ " 3.550 | \n",
+ " -5.615 | \n",
+ " 2.132 | \n",
+ " 3.472 | \n",
+ " 1.523 | \n",
+ " 10.904000 | \n",
+ "
\n",
+ " \n",
+ " 2018-06-26 16:00:00 | \n",
+ " -5.492 | \n",
+ " 4.287 | \n",
+ " -9.132 | \n",
+ " 2.274 | \n",
+ " 3.533 | \n",
+ " 1.675 | \n",
+ " 11.044000 | \n",
+ "
\n",
+ " \n",
+ " 2018-06-26 17:00:00 | \n",
+ " 2.813 | \n",
+ " 3.818 | \n",
+ " -0.817 | \n",
+ " 2.097 | \n",
+ " 3.716 | \n",
+ " 1.523 | \n",
+ " 10.271000 | \n",
+ "
\n",
+ " \n",
+ " 2018-06-26 18:00:00 | \n",
+ " 9.243 | \n",
+ " 3.818 | \n",
+ " 5.472 | \n",
+ " 2.097 | \n",
+ " 3.655 | \n",
+ " 1.432 | \n",
+ " 9.778000 | \n",
+ "
\n",
+ " \n",
+ " 2018-06-26 19:00:00 | \n",
+ " 10.114 | \n",
+ " 3.550 | \n",
+ " 6.183 | \n",
+ " 1.564 | \n",
+ " 3.716 | \n",
+ " 1.462 | \n",
+ " 9.567000 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
17420 rows Γ 7 columns
\n",
+ "
"
+ ],
+ "text/plain": [
+ "component HUFL HULL MUFL MULL LUFL LULL OT\n",
+ "date \n",
+ "2016-07-01 00:00:00 5.827 2.009 1.599 0.462 4.203 1.340 30.531000\n",
+ "2016-07-01 01:00:00 5.693 2.076 1.492 0.426 4.142 1.371 27.787001\n",
+ "2016-07-01 02:00:00 5.157 1.741 1.279 0.355 3.777 1.218 27.787001\n",
+ "2016-07-01 03:00:00 5.090 1.942 1.279 0.391 3.807 1.279 25.044001\n",
+ "2016-07-01 04:00:00 5.358 1.942 1.492 0.462 3.868 1.279 21.948000\n",
+ "... ... ... ... ... ... ... ...\n",
+ "2018-06-26 15:00:00 -1.674 3.550 -5.615 2.132 3.472 1.523 10.904000\n",
+ "2018-06-26 16:00:00 -5.492 4.287 -9.132 2.274 3.533 1.675 11.044000\n",
+ "2018-06-26 17:00:00 2.813 3.818 -0.817 2.097 3.716 1.523 10.271000\n",
+ "2018-06-26 18:00:00 9.243 3.818 5.472 2.097 3.655 1.432 9.778000\n",
+ "2018-06-26 19:00:00 10.114 3.550 6.183 1.564 3.716 1.462 9.567000\n",
+ "\n",
+ "[17420 rows x 7 columns]"
+ ]
+ },
+ "execution_count": 4,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "series = []\n",
+ "for idx, ds in enumerate([ETTh1Dataset, ETTh2Dataset]):\n",
+ " trafo = ds().load().astype(np.float32)\n",
+ " trafo = trafo.with_static_covariates(pd.DataFrame({\"transformer_id\": [idx]}))\n",
+ " series.append(trafo)\n",
+ "series[0].pd_dataframe()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Before training, we split the data into train, validation, and test sets. The model will learn from the train set, use the validation set to determine when to stop training, and finally be evaluated on the test set."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "train, val, test = [], [], []\n",
+ "for trafo in series:\n",
+ " train_, temp = trafo.split_after(0.6)\n",
+ " val_, test_ = temp.split_after(0.5)\n",
+ " train.append(train_)\n",
+ " val.append(val_)\n",
+ " test.append(test_)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Lets look at the splits for the first column \"HUFL\" for each transformer"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "show_col = \"HUFL\"\n",
+ "for idx, (train_, val_, test_) in enumerate(zip(train, val, test)):\n",
+ " train_[show_col].plot(label=f\"train_trafo_{idx}\")\n",
+ " val_[show_col].plot(label=f\"val_trafo_{idx}\")\n",
+ " test_[show_col].plot(label=f\"test_trafo_{idx}\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Now let's scale the data. To avoid leaking information from the validation and test sets, we scale the data based on the properties of the train set."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "scaler = Scaler() # default uses sklearn's MinMaxScaler\n",
+ "train = scaler.fit_transform(train)\n",
+ "val = scaler.transform(val)\n",
+ "test = scaler.transform(test)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Model Parameter Setup\n",
+ "Boilerplate code is no fun, especially in the context of training multiple models to compare performance. To avoid this, we use a common configuration that can be used with any Darts `TorchForecastingModel`.\n",
+ "\n",
+ "A few interesting things about these parameters:\n",
+ "\n",
+ "- **Gradient clipping:** Mitigates exploding gradients during backpropagation by setting an upper limit on the gradient for a batch.\n",
+ "\n",
+ "- **Learning rate:** The majority of the learning done by a model is in the earlier epochs. As training goes on it is often helpful to reduce the learning rate to fine-tune the model. That being said, it can also lead to significant overfitting.\n",
+ "\n",
+ "- **Early stopping:** To avoid overfitting, we can use early stopping. It monitors a metric on the validation set and stops training once the metric is not improving anymore based on a custom condition.\n",
+ "\n",
+ "- **Likelihood and Loss Functions:** You can either make the model probabilistic with a `likelihood`, or deterministic with a `loss_fn`. In this notebook we train probabilistic models using QuantileRegression.\n",
+ "\n",
+ "- **Reversible Instance Normalization:** Use [Reversible Instance Normalization](https://openreview.net/forum?id=cGDAkQo1C0p) which in most of the cases improves model performance.\n",
+ "\n",
+ "- **Encoders:** We can encode time axis/calendar information and use them as past or future covariates using `add_encoders`. Here, we'll add cyclic encodings of the hour, day of the week, and month as future covariates"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def create_params(\n",
+ " input_chunk_length: int,\n",
+ " output_chunk_length: int,\n",
+ " full_training=True,\n",
+ "):\n",
+ " # early stopping: this setting stops training once the the validation\n",
+ " # loss has not decreased by more than 1e-5 for 10 epochs\n",
+ " early_stopper = EarlyStopping(\n",
+ " monitor=\"val_loss\",\n",
+ " patience=10,\n",
+ " min_delta=1e-5,\n",
+ " mode=\"min\",\n",
+ " )\n",
+ "\n",
+ " # PyTorch Lightning Trainer arguments (you can add any custom callback)\n",
+ " if full_training:\n",
+ " limit_train_batches = None\n",
+ " limit_val_batches = None\n",
+ " max_epochs = 200\n",
+ " batch_size = 256\n",
+ " else:\n",
+ " limit_train_batches = 20\n",
+ " limit_val_batches = 10\n",
+ " max_epochs = 40\n",
+ " batch_size = 64\n",
+ "\n",
+ " # only show the training and prediction progress bars\n",
+ " progress_bar = TFMProgressBar(\n",
+ " enable_sanity_check_bar=False, enable_validation_bar=False\n",
+ " )\n",
+ " pl_trainer_kwargs = {\n",
+ " \"gradient_clip_val\": 1,\n",
+ " \"max_epochs\": max_epochs,\n",
+ " \"limit_train_batches\": limit_train_batches,\n",
+ " \"limit_val_batches\": limit_val_batches,\n",
+ " \"accelerator\": \"auto\",\n",
+ " \"callbacks\": [early_stopper, progress_bar],\n",
+ " }\n",
+ "\n",
+ " # optimizer setup, uses Adam by default\n",
+ " optimizer_cls = torch.optim.Adam\n",
+ " optimizer_kwargs = {\n",
+ " \"lr\": 1e-4,\n",
+ " }\n",
+ "\n",
+ " # learning rate scheduler\n",
+ " lr_scheduler_cls = torch.optim.lr_scheduler.ExponentialLR\n",
+ " lr_scheduler_kwargs = {\"gamma\": 0.999}\n",
+ "\n",
+ " # for probabilistic models, we use quantile regression, and set `loss_fn` to `None`\n",
+ " likelihood = QuantileRegression()\n",
+ " loss_fn = None\n",
+ "\n",
+ " return {\n",
+ " \"input_chunk_length\": input_chunk_length, # lookback window\n",
+ " \"output_chunk_length\": output_chunk_length, # forecast/lookahead window\n",
+ " \"use_reversible_instance_norm\": True,\n",
+ " \"optimizer_kwargs\": optimizer_kwargs,\n",
+ " \"pl_trainer_kwargs\": pl_trainer_kwargs,\n",
+ " \"lr_scheduler_cls\": lr_scheduler_cls,\n",
+ " \"lr_scheduler_kwargs\": lr_scheduler_kwargs,\n",
+ " \"likelihood\": likelihood, # use a `likelihood` for probabilistic forecasts\n",
+ " \"loss_fn\": loss_fn, # use a `loss_fn` for determinsitic model\n",
+ " \"save_checkpoints\": True, # checkpoint to retrieve the best performing model state,\n",
+ " \"force_reset\": True,\n",
+ " \"batch_size\": batch_size,\n",
+ " \"random_state\": 42,\n",
+ " \"add_encoders\": {\n",
+ " \"cyclic\": {\n",
+ " \"future\": [\"hour\", \"dayofweek\", \"month\"]\n",
+ " } # add cyclic time axis encodings as future covariates\n",
+ " },\n",
+ " }"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Model configuration\n",
+ "Let's use the last week of hourly data as lookback window (`input_chunk_length`) and train a probabilistic model to predict the next 24 hours directly (`output_chunk_length`). Additionally, we tell the model to use the static information. To keep the notebook simple, we'll set `full_training=False`. To get even better performance, set `full_training=True`.\n",
+ "\n",
+ "Apart from that, we use our helper function to set up all the common model arguments."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "input_chunk_length = 7 * 24\n",
+ "output_chunk_length = 24\n",
+ "use_static_covariates = True\n",
+ "full_training = False"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# create the models\n",
+ "model_tsm = TSMixerModel(\n",
+ " **create_params(\n",
+ " input_chunk_length,\n",
+ " output_chunk_length,\n",
+ " full_training=full_training,\n",
+ " ),\n",
+ " use_static_covariates=use_static_covariates,\n",
+ " model_name=\"tsm\",\n",
+ ")\n",
+ "model_tide = TiDEModel(\n",
+ " **create_params(\n",
+ " input_chunk_length,\n",
+ " output_chunk_length,\n",
+ " full_training=full_training,\n",
+ " ),\n",
+ " use_static_covariates=use_static_covariates,\n",
+ " model_name=\"tide\",\n",
+ ")\n",
+ "models = {\n",
+ " \"TSM\": model_tsm,\n",
+ " \"TiDE\": model_tide,\n",
+ "}"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Model Training"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Now let's train all of the models. When using early stopping it is important to save checkpoints. This allows us to continue past the best model configuration and then restore the optimal weights once training has been completed."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "1ab2f4e3c6a14b4687d70b402b9920ac",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ "Training: 0it [00:00, ?it/s]"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "c8efee5bcaef467499408860f691509d",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ "Training: 0it [00:00, ?it/s]"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "# train the models and load the model from its best state/checkpoint\n",
+ "for model_name, model in models.items():\n",
+ " model.fit(\n",
+ " series=train,\n",
+ " val_series=val,\n",
+ " )\n",
+ " # load from checkpoint returns a new model object, we store it in the models dict\n",
+ " models[model_name] = model.load_from_checkpoint(\n",
+ " model_name=model.model_name, best=True\n",
+ " )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Backtest the probabilistic models\n",
+ "\n",
+ "Let's configure the prediction. For this example, we will:\n",
+ "- generate **historical forecasts** on the test set using the **pre-trained models**. Each forecast covers a 24 hour horizon, and the time between two consecutive forecasts is also 24 hours. This will give us **276 multivariate forecasts per transformer** to evaluate the model!\n",
+ "- generate **500 stochastic samples** for each prediction point (since we have trained probabilistic models)\n",
+ "- evaluate/**backtest** the probabilistic historical forecasts for some quantiles **using the Mean Quantile Loss** (`mql()`).\n",
+ "\n",
+ "And we'll create some helper functions to generating the forecasts, computing the backtest, and to visualize the predictions."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# configure the probabilistic prediction\n",
+ "num_samples = 500\n",
+ "forecast_horizon = output_chunk_length\n",
+ "\n",
+ "# compute the Mean Quantile Loss over these quantiles\n",
+ "evaluate_quantiles = [0.05, 0.1, 0.2, 0.5, 0.8, 0.9, 0.95]\n",
+ "\n",
+ "\n",
+ "def historical_forecasts(model):\n",
+ " \"\"\"Generates probabilistic historical forecasts for each transformer\n",
+ " and returns the inverse transforms results.\n",
+ "\n",
+ " Each forecast covers 24h (forecast_horizon). The time between two forecasts\n",
+ " (stride) is also 24 hours.\n",
+ " \"\"\"\n",
+ " hfc = model.historical_forecasts(\n",
+ " series=test,\n",
+ " forecast_horizon=forecast_horizon,\n",
+ " stride=forecast_horizon,\n",
+ " last_points_only=False,\n",
+ " retrain=False,\n",
+ " num_samples=num_samples,\n",
+ " verbose=True,\n",
+ " )\n",
+ " return scaler.inverse_transform(hfc)\n",
+ "\n",
+ "\n",
+ "def backtest(model, hfc, name):\n",
+ " \"\"\"Evaluates probabilistic historical forecasts using the Mean Quantile\n",
+ " Loss (MQL) over a set of quantiles.\"\"\"\n",
+ " # add metric specific kwargs\n",
+ " metric_kwargs = [{\"q\": q} for q in evaluate_quantiles]\n",
+ " metrics = [mql for _ in range(len(evaluate_quantiles))]\n",
+ " bt = model.backtest(\n",
+ " series=series,\n",
+ " historical_forecasts=hfc,\n",
+ " last_points_only=False,\n",
+ " metric=metrics,\n",
+ " metric_kwargs=metric_kwargs,\n",
+ " verbose=True,\n",
+ " )\n",
+ " bt = pd.DataFrame(\n",
+ " bt,\n",
+ " columns=[f\"q_{q}\" for q in evaluate_quantiles],\n",
+ " index=[f\"{trafo}_{name}\" for trafo in [\"ETTh1\", \"ETTh2\"]],\n",
+ " )\n",
+ " return bt\n",
+ "\n",
+ "\n",
+ "def generate_plots(n_days, hfcs):\n",
+ " \"\"\"Plot the probabilistic forecasts for each model, transformer and transformer\n",
+ " feature against the ground truth.\"\"\"\n",
+ " # concatenate historical forecasts into contiguous time series\n",
+ " # (works because forecast_horizon=stride)\n",
+ " hfcs_plot = {}\n",
+ " for model_name, hfc_model in hfcs.items():\n",
+ " hfcs_plot[model_name] = [\n",
+ " concatenate(hfc_series[-n_days:], axis=0) for hfc_series in hfc_model\n",
+ " ]\n",
+ "\n",
+ " # remember start and end points for plotting the target series\n",
+ " hfc_ = hfcs_plot[model_name][0]\n",
+ " start, end = hfc_.start_time(), hfc_.end_time()\n",
+ "\n",
+ " # for each target column...\n",
+ " for col in series[0].columns:\n",
+ " fig, axes = plt.subplots(ncols=2, figsize=(12, 6))\n",
+ " # ... and for each transformer...\n",
+ " for trafo_idx, trafo in enumerate(series):\n",
+ " trafo[col][start:end].plot(label=\"ground truth\", ax=axes[trafo_idx])\n",
+ " # ... plot the historical forecasts for each model\n",
+ " for model_name, hfc in hfcs_plot.items():\n",
+ " hfc[trafo_idx][col].plot(\n",
+ " label=model_name + \"_q0.05-q0.95\", ax=axes[trafo_idx]\n",
+ " )\n",
+ " axes[trafo_idx].set_title(f\"ETTh{trafo_idx + 1}: {col}\")\n",
+ " plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Okay, now we're ready to evaluate the models"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Model: TSM\n",
+ "Generating historical forecasts..\n"
+ ]
+ },
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "809ff39dfd7b4192b102d9151b2c1417",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ "Predicting: 0it [00:00, ?it/s]"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Evaluating historical forecasts..\n",
+ "Model: TiDE\n",
+ "Generating historical forecasts..\n"
+ ]
+ },
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "ca2e5b2a7d634d7ea619998ce8a11dd7",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ "Predicting: 0it [00:00, ?it/s]"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Evaluating historical forecasts..\n"
+ ]
+ }
+ ],
+ "source": [
+ "bts = {}\n",
+ "hfcs = {}\n",
+ "for model_name, model in models.items():\n",
+ " print(f\"Model: {model_name}\")\n",
+ " print(\"Generating historical forecasts..\")\n",
+ " hfcs[model_name] = historical_forecasts(models[model_name])\n",
+ "\n",
+ " print(\"Evaluating historical forecasts..\")\n",
+ " bts[model_name] = backtest(models[model_name], hfcs[model_name], model_name)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Let's see how they performed.\n",
+ "\n",
+ "> **Note:** These results are likely to improve/change when setting `full_training=True`"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " q_0.05 | \n",
+ " q_0.1 | \n",
+ " q_0.2 | \n",
+ " q_0.5 | \n",
+ " q_0.8 | \n",
+ " q_0.9 | \n",
+ " q_0.95 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " ETTh1_TSM | \n",
+ " 0.501772 | \n",
+ " 0.769545 | \n",
+ " 1.136141 | \n",
+ " 1.568439 | \n",
+ " 1.098847 | \n",
+ " 0.721835 | \n",
+ " 0.442062 | \n",
+ "
\n",
+ " \n",
+ " ETTh1_TiDE | \n",
+ " 0.573716 | \n",
+ " 0.885452 | \n",
+ " 1.298672 | \n",
+ " 1.671870 | \n",
+ " 1.151501 | \n",
+ " 0.727515 | \n",
+ " 0.446724 | \n",
+ "
\n",
+ " \n",
+ " ETTh2_TSM | \n",
+ " 0.659187 | \n",
+ " 1.030655 | \n",
+ " 1.508628 | \n",
+ " 1.932923 | \n",
+ " 1.317960 | \n",
+ " 0.857147 | \n",
+ " 0.524620 | \n",
+ "
\n",
+ " \n",
+ " ETTh2_TiDE | \n",
+ " 0.627251 | \n",
+ " 0.982114 | \n",
+ " 1.450893 | \n",
+ " 1.897117 | \n",
+ " 1.323661 | \n",
+ " 0.862239 | \n",
+ " 0.528638 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
"
+ ],
+ "text/plain": [
+ " q_0.05 q_0.1 q_0.2 q_0.5 q_0.8 q_0.9 \\\n",
+ "ETTh1_TSM 0.501772 0.769545 1.136141 1.568439 1.098847 0.721835 \n",
+ "ETTh1_TiDE 0.573716 0.885452 1.298672 1.671870 1.151501 0.727515 \n",
+ "ETTh2_TSM 0.659187 1.030655 1.508628 1.932923 1.317960 0.857147 \n",
+ "ETTh2_TiDE 0.627251 0.982114 1.450893 1.897117 1.323661 0.862239 \n",
+ "\n",
+ " q_0.95 \n",
+ "ETTh1_TSM 0.442062 \n",
+ "ETTh1_TiDE 0.446724 \n",
+ "ETTh2_TSM 0.524620 \n",
+ "ETTh2_TiDE 0.528638 "
+ ]
+ },
+ "execution_count": 14,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "bt_df = pd.concat(bts.values(), axis=0).sort_index()\n",
+ "bt_df"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The backtest gives us the Mean Quantile Loss for the selected quantiles over all transformer features per transformer and model. The lower the value, the better. The `q_0.5` is identical to the Mean Absolute Error (MAE) between the median prediction and the ground truth.\n",
+ "\n",
+ "Both models seem to have performed comparably well. And how does it look on average over all quantiles?"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "ETTh1_TSM 0.891234\n",
+ "ETTh1_TiDE 0.965064\n",
+ "ETTh2_TSM 1.118732\n",
+ "ETTh2_TiDE 1.095988\n",
+ "dtype: float64"
+ ]
+ },
+ "execution_count": 15,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "bt_df.mean(axis=1)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Here the results are also very similar. It seems that TSMixer performed better for ETTh1, and TiDEModel for ETTh2.\n",
+ "\n",
+ "And last but not least, let's have look at the predictions for the last `n_days=3` days in the test set.\n",
+ "\n",
+ "> Note: The prediction intervals are expected to get narrower when `full_training=True`"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "