Devops/add back tests to package (#2875)

dennisbader · web-flow · commit 34a2886b6916 · 2025-08-18T16:19:41.000+02:00
* add back tests to darts package source for conda build

* update docs

* update changelog

* remove examples diff
diff --git a/.dockerignore b/.dockerignore
@@ -3,6 +3,5 @@
 **/__pycache__
 
 # Test files
-darts/tests/
 darts/examples/static
 **/darts_logs
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -11,8 +11,12 @@ but cannot always guarantee backwards compatibility. Changes that may **break co
 
 **Improved**
 
+- Improved the documentation regarding the new classification models from version 0.37.0. [#2875](https://github.com/unit8co/darts/pull/2875) by [Dennis Bader](https://github.com/dennisbader).
+
 **Fixed**
 
+- Added back the unit tests into the source code for our conda package. [#2875](https://github.com/unit8co/darts/pull/2875) by [Dennis Bader](https://github.com/dennisbader).
+
 **Dependencies**
 
 ### For developers of the library:
diff --git a/README.md b/README.md
diff --git a/darts/models/forecasting/__init__.py b/darts/models/forecasting/__init__.py
@@ -1,6 +1,9 @@
 """
 Forecasting Models
-------------------
+==================
+
+Regression Models
+-----------------
 
 Baseline Models (`LocalForecastingModel <https://unit8co.github.io/darts/userguide/covariates.html#local-forecasting-models-lfms>`_)
     - :class:`~darts.models.forecasting.baselines.NaiveMean`
@@ -29,13 +32,13 @@
     - :class:`~darts.models.forecasting.fft.FFT`
     - :class:`~darts.models.forecasting.kalman_forecaster.KalmanForecaster`
     - :class:`~darts.models.forecasting.sf_croston.Croston`
-Regression Models (`GlobalForecastingModel <https://unit8co.github.io/darts/userguide/covariates.html#global-forecasting-models-gfms>`_)
+SKLearn-Like Models (`GlobalForecastingModel <https://unit8co.github.io/darts/userguide/covariates.html#global-forecasting-models-gfms>`_)
     - :class:`~darts.models.forecasting.sklearn_model.SKLearnModel`
     - :class:`~darts.models.forecasting.linear_regression_model.LinearRegressionModel`
     - :class:`~darts.models.forecasting.random_forest.RandomForestModel`
+    - :class:`~darts.models.forecasting.catboost_model.CatBoostModel`
     - :class:`~darts.models.forecasting.lgbm.LightGBMModel`
     - :class:`~darts.models.forecasting.xgboost.XGBModel`
-    - :class:`~darts.models.forecasting.catboost_model.CatBoostModel`
 PyTorch (Lightning)-based Models (`GlobalForecastingModel <https://unit8co.github.io/darts/userguide/covariates.html#global-forecasting-models-gfms>`_)
     - :class:`~darts.models.forecasting.rnn_model.RNNModel`
     - :class:`~darts.models.forecasting.block_rnn_model.BlockRNNModel`
@@ -54,4 +57,13 @@
 Conformal Models  (`GlobalForecastingModel <https://unit8co.github.io/darts/userguide/covariates.html#global-forecasting-models-gfms>`_)
     - :class:`~darts.models.forecasting.conformal_models.ConformalNaiveModel`
     - :class:`~darts.models.forecasting.conformal_models.ConformalQRModel`
+
+Classification Models
+---------------------
+
+SKLearn-Like Models (`GlobalForecastingModel <https://unit8co.github.io/darts/userguide/covariates.html#global-forecasting-models-gfms>`_)
+    - :class:`~darts.models.forecasting.sklearn_model.SKLearnClassifierModel`
+    - :class:`~darts.models.forecasting.catboost_model.CatBoostClassifierModel`
+    - :class:`~darts.models.forecasting.lgbm.LightGBMClassifierModel`
+    - :class:`~darts.models.forecasting.xgboost.XGBClassifierModel`
 """
diff --git a/darts/models/forecasting/catboost_model.py b/darts/models/forecasting/catboost_model.py
@@ -1,10 +1,24 @@
 """
-CatBoost model
---------------
+CatBoost Models
+---------------
 
-CatBoost based regression model.
+This module offers wrappers around CatBoost's Gradient Boosted Trees algorithms.
 
-This implementation comes with the ability to produce probabilistic forecasts.
+* :class:`~darts.models.forecasting.catboost_model.CatBoostModel` - Wrapper around CatBoost's `CatBoostRegressor`
+* :class:`~darts.models.forecasting.catboost_model.CatBoostClassifierModel` - Wrapper around CatBoost's
+  `CatBoostClassifier`
+
+The wrappers come with all capabilities of Darts' `SKLearn*Model`.
+
+For detailed examples and tutorials, see:
+
+* `SKLearn-Like Regression Model Examples
+  <https://unit8co.github.io/darts/examples/20-SKLearnModel-examples.html>`_
+* `SKLearn-Like Classification Model Examples
+  <https://unit8co.github.io/darts/examples/24-SKLearnClassifierModel-examples.html>`_
+
+To enable CatBoost support in Darts, follow the detailed install instructions for CatBoost in the INSTALL:
+https://github.com/unit8co/darts/blob/master/INSTALL.md
 """
 
 from collections.abc import Sequence
diff --git a/darts/models/forecasting/conformal_models.py b/darts/models/forecasting/conformal_models.py
@@ -1,6 +1,6 @@
 """
 Conformal Models
----------------
+----------------
 
 A collection of conformal prediction models for pre-trained global forecasting models.
 """
diff --git a/darts/models/forecasting/lgbm.py b/darts/models/forecasting/lgbm.py
@@ -1,10 +1,18 @@
 """
-LightGBM Model
---------------
+LightGBM Models
+---------------
 
-This is a LightGBM implementation of Gradient Boosted Trees algorithm.
+This module offers wrappers around LightGBM's Gradient Boosted Trees algorithms.
 
-This implementation comes with the ability to produce probabilistic forecasts.
+* :class:`~darts.models.forecasting.lgbm.LightGBMModel` - Wrapper around LightGBM's `LGBMRegressor`
+* :class:`~darts.models.forecasting.lgbm.LightGBMClassifierModel` - Wrapper around LightGBM's `LGBMClassifier`
+
+The wrappers come with all capabilities of Darts' `SKLearn*Model`.
+
+For detailed examples and tutorials, see:
+
+* `SKLearn-Like Regression Model Examples <https://unit8co.github.io/darts/examples/20-SKLearnModel-examples.html>`_
+* `SKLearn-Like Classification Model Examples <https://unit8co.github.io/darts/examples/24-SKLearnClassifierModel-examples.html>`_
 
 To enable LightGBM support in Darts, follow the detailed install instructions for LightGBM in the INSTALL:
 https://github.com/unit8co/darts/blob/master/INSTALL.md
diff --git a/darts/models/forecasting/linear_regression_model.py b/darts/models/forecasting/linear_regression_model.py
@@ -1,5 +1,5 @@
 """
-Linear Regression model
+Linear Regression Model
 -----------------------
 
 A forecasting model using a linear regression of some of the target series' lags, as well as optionally some
diff --git a/darts/models/forecasting/nlinear.py b/darts/models/forecasting/nlinear.py
@@ -1,6 +1,6 @@
 """
 N-Linear
-------
+--------
 """
 
 import torch
diff --git a/darts/models/forecasting/regression_ensemble_model.py b/darts/models/forecasting/regression_ensemble_model.py
@@ -1,5 +1,5 @@
 """
-Regression ensemble model
+Regression Ensemble Model
 -------------------------
 
 An ensemble model which uses a regression model to compute the ensemble forecast.
diff --git a/darts/models/forecasting/sf_auto_mfles.py b/darts/models/forecasting/sf_auto_mfles.py
@@ -1,6 +1,6 @@
 """
 AutoMFLES
------------
+---------
 """
 
 from typing import Optional
diff --git a/darts/models/forecasting/sf_auto_theta.py b/darts/models/forecasting/sf_auto_theta.py
@@ -1,6 +1,6 @@
 """
 AutoTheta
------------
+---------
 """
 
 from typing import Optional
diff --git a/darts/models/forecasting/sf_croston.py b/darts/models/forecasting/sf_croston.py
@@ -1,5 +1,5 @@
 """
-Croston method
+Croston Method
 --------------
 """
 
diff --git a/darts/models/forecasting/sklearn_model.py b/darts/models/forecasting/sklearn_model.py
@@ -1,37 +1,58 @@
 """
-SKLearn Model
--------------
-A `SKLearnModel` forecasts future values of a target series based on
+SKLearn-Like Models
+-------------------
 
-* The target series (past lags only)
+Darts provides a comprehensive set of forecasting models that wrap around scikit-learn-like machine learning algorithms.
+These models can forecast future values of a target series based on lagged features of the following inputs:
 
-* An optional past_covariates series (past lags only)
+* **Target series** (past lags only) - the main time series to be forecasted
+* **Past covariates** (past lags only) - external variables known only up to the prediction time
+* **Future covariates** (past and future lags) - external variables known or forecasted into the future
+* **Static covariates** - time-invariant features associated with each target series
 
-* An optional future_covariates series (possibly past and future lags)
+The models learn in a supervised way by tabularizing time series data into (lagged) feature-target pairs, making them
+compatible with any scikit-learn-like model that has ``fit()`` and ``predict()`` methods.
 
-* Available static covariates
+Lags can be specified in multiple ways:
 
+* As an integer for the target and past covariates series: represents the number of past lags to consider
+* As a tuple (integer, integer) for the future covariates series: represents the number of (past, future) lags to
+  consider
+* As a list: explicitly enumerate specific lags (negative values for past, non-negative for future)
+* As a dictionary: specify different lags for different series components (columns)
 
-The regression models are learned in a supervised way, and they can wrap around any "scikit-learn like" regression model
-acting on tabular data having ``fit()`` and ``predict()`` methods.
+We offer both regression and classification models, each tailored for specific forecasting tasks.
 
-Darts also provides the following models:
+**Regression Models:** These models predict continuous numerical values, making them ideal for forecasting future
+trends and patterns in time series data. Utilize these models to gain insights into potential future outcomes based on
+historical data.
 
-- :class:`~darts.models.forecasting.linear_regression_model.LinearRegressionModel` : wrapping around scikit-learn's
+* :class:`~darts.models.forecasting.sklearn_model.SKLearnModel` - Generic wrapper around any scikit-learn regression
+  model (default: `LinearRegression`)
+* :class:`~darts.models.forecasting.linear_regression_model.LinearRegressionModel` - Wrapper around scikit-learn's
   `LinearRegression`
-- :class:`~darts.models.forecasting.random_forest.RandomForestModel` : wrapping around scikit-learn's
+* :class:`~darts.models.forecasting.random_forest.RandomForestModel` - Wrapper around scikit-learn's
   `RandomForestRegressor`
-- :class:`~darts.models.forecasting.xgboost.XGBModel` : wrapping around XGBoost's `XGBRegressor`
-- :class:`~darts.models.forecasting.lgbm.LightGBMModel` : wrapping around LightGBM's `LGBMRegressor`
-- :class:`~darts.models.forecasting.catboost_model.CatBoostModel` : wrapping around CatBoost's `CatBoostRegressor`
-
-Behind the scenes this model tabularizes the time series data to make it work with regression models.
-
-The lags can be specified either using an integer - in which case it represents the _number_ of (past or future) lags
-to take into consideration, or as a list - in which case the lags have to be enumerated (strictly negative values
-denoting past lags and positive values including 0 denoting future lags).
-When static covariates are present, they are appended to the lagged features. When multiple time series are passed,
-if their static covariates do not have the same size, the shorter ones are padded with 0 valued features.
+* :class:`~darts.models.forecasting.catboost_model.CatBoostModel` - Wrapper around CatBoost's `CatBoostRegressor`
+* :class:`~darts.models.forecasting.lgbm.LightGBMModel` - Wrapper around LightGBM's `LGBMRegressor`
+* :class:`~darts.models.forecasting.xgboost.XGBModel` - Wrapper around XGBoost's `XGBRegressor`
+
+**Classification Models:** These models predict categorical class labels, enabling effective time series labeling and
+future class prediction. These models are perfect for scenarios where identifying distinct categories or states over
+time is crucial.
+
+* :class:`~darts.models.forecasting.sklearn_model.SKLearnClassifierModel` - Generic wrapper around any scikit-learn
+  classification model (default: `LogisticRegression`)
+* :class:`~darts.models.forecasting.catboost_model.CatBoostClassifierModel` - Wrapper around CatBoost's
+  `CatBoostClassifier`
+* :class:`~darts.models.forecasting.lgbm.LightGBMClassifierModel` - Wrapper around LightGBM's `LGBMClassifier`
+* :class:`~darts.models.forecasting.xgboost.XGBClassifierModel` - Wrapper around XGBoost's `XGBClassifier`
+
+For detailed examples and tutorials, see:
+
+* `SKLearn-Like Regression Model Examples <https://unit8co.github.io/darts/examples/20-SKLearnModel-examples.html>`_
+* `SKLearn-Like Classification Model Examples
+  <https://unit8co.github.io/darts/examples/24-SKLearnClassifierModel-examples.html>`_
 """
 
 from abc import ABC, abstractmethod
@@ -2164,8 +2185,8 @@ def __init__(
 
         .. note::
             If target series `lags` are provided, they will be treated as numeric input features (and not categorical).
-            For categorical input feature support, consider models such as:
-            class:`~darts.models.forecasting.lgbm.LightGBMClassifierModel` or
+            For categorical input feature support, consider models such as
+            :class:`~darts.models.forecasting.lgbm.LightGBMClassifierModel` or
             :class:`~darts.models.forecasting.catboost_model.CatBoostClassifierModel`.
 
         Parameters
diff --git a/darts/models/forecasting/tft_model.py b/darts/models/forecasting/tft_model.py
@@ -1,6 +1,6 @@
 """
 Temporal Fusion Transformer (TFT)
--------
+---------------------------------
 """
 
 from collections.abc import Sequence
diff --git a/darts/models/forecasting/tide_model.py b/darts/models/forecasting/tide_model.py
@@ -1,6 +1,6 @@
 """
 Time-series Dense Encoder (TiDE)
-------
+--------------------------------
 """
 
 from typing import Optional
diff --git a/darts/models/forecasting/varima.py b/darts/models/forecasting/varima.py
@@ -1,6 +1,6 @@
 """
 VARIMA
------
+------
 
 Models for VARIMA (Vector Autoregressive moving average) [1]_.
 The implementations is wrapped around `statsmodels <https://github.com/statsmodels/statsmodels>`_.
diff --git a/darts/models/forecasting/xgboost.py b/darts/models/forecasting/xgboost.py
@@ -1,10 +1,21 @@
 """
-XGBoost Model
--------------
+XGBoost Models
+--------------
 
-Regression model based on XGBoost.
+This module offers wrappers around XGBoost's Gradient Boosted Trees algorithms.
 
-This implementation comes with the ability to produce probabilistic forecasts.
+* :class:`~darts.models.forecasting.xgboost.XGBModel` - Wrapper around XGBoost's `XGBRegressor`
+* :class:`~darts.models.forecasting.xgboost.XGBClassifierModel` - Wrapper around XGBoost's
+  `XGBClassifier`
+
+The wrappers come with all capabilities of Darts' `SKLearn*Model`.
+
+For detailed examples and tutorials, see:
+
+* `SKLearn-Like Regression Model Examples
+  <https://unit8co.github.io/darts/examples/20-SKLearnModel-examples.html>`_
+* `SKLearn-Like Classification Model Examples
+  <https://unit8co.github.io/darts/examples/24-SKLearnClassifierModel-examples.html>`_
 """
 
 from collections.abc import Sequence
diff --git a/docs/userguide/covariates.md b/docs/userguide/covariates.md
diff --git a/setup.py b/setup.py
diff --git a/setup_u8darts.py b/setup_u8darts.py

-Original file line number
+Diff line change
@@ @@ -1,6 +1,6 @@ @@
 """
 Conformal Models
 ----------------
 +----------------
 A collection of conformal prediction models for pre-trained global forecasting models.
 """