Fix/val_sample_weight error for models inherited from RegressionModel (#2626)

* Fix/val_sample_weight error for models inherited from RegressionModel

* modify change log

* modify the statement

* update changelog

* add unit tests

* Update CHANGELOG.md

---------

Co-authored-by: madtoinou <[email protected]>
Co-authored-by: dennisbader <[email protected]>

Author: 3 people

CHANGELOG.md

@@ -20,6 +20,7 @@ but cannot always guarantee backwards compatibility. Changes that may **break co
 - Improvements to `ForecastingModel`:
   - Added parameter `clean: bool` to `ForecastingModel.save()` to store a cleaned version of the model (removes training data from global models, and Lightning Trainer-related parameters from torch models). [#2649](https://github.com/unit8co/darts/pull/2649) by [Jonas Blanc](https://github.com/jonasblanc).
 - Added parameter `pl_trainer_kwargs` to `TorchForecastingModel.load()` to setup a new Lightning Trainer used to configure the model for downstream tasks (e.g. prediction). [#2649](https://github.com/unit8co/darts/pull/2649) by [Jonas Blanc](https://github.com/jonasblanc).
+- Fixed a bug in `LightGBMModel`, `XGBModel`, and `CatBoostModel` which raised an error when calling `fit()` with `val_sample_weight`. [#2626](https://github.com/unit8co/darts/pull/2626) by [Kylin Schmidt](https://github.com/kylinschmidt).
 - Improved the documentation of how `WindowedAnomalyScorer` extract the training data from the input series. [#2674](https://github.com/unit8co/darts/pull/2674) by [Dennis Bader](https://github.com/dennisbader).
 
 **Fixed**

darts/models/forecasting/regression_model.py

@@ -591,7 +591,7 @@ def _add_val_set_to_kwargs(
             val_weights = val_weights or None
         else:
             val_sets = [(val_samples, val_labels)]
-            val_weights = val_weight
+            val_weights = [val_weight]
 
         val_set_name, val_weight_name = self.val_set_params
         return dict(kwargs, **{val_set_name: val_sets, val_weight_name: val_weights})

darts/tests/models/forecasting/test_regression_models.py

@@ -1867,6 +1867,34 @@ def test_not_enough_covariates(self, config):
                 future_covariates=future_covariates[: -26 + req_future_offset],
             )
 
+    @pytest.mark.parametrize(
+        "config",
+        product(
+            [(XGBModel, xgb_test_params)]
+            + ([(LightGBMModel, lgbm_test_params)] if lgbm_available else [])
+            + ([(CatBoostModel, cb_test_params)] if cb_available else []),
+            [True, False],
+        ),
+    )
+    def test_val_set_weights_runnability_trees(self, config):
+        """Tests using weights in val set for single and multi series."""
+        (model_cls, model_kwargs), single_series = config
+        model = model_cls(lags=10, **model_kwargs)
+
+        series = tg.sine_timeseries(length=20)
+        weights = tg.linear_timeseries(length=20)
+        if not single_series:
+            series = [series] * 2
+            weights = [weights] * 2
+
+        model.fit(
+            series=series,
+            val_series=series,
+            sample_weight=weights,
+            val_sample_weight=weights,
+        )
+        _ = model.predict(1, series=series)
+
     @pytest.mark.parametrize(
         "config",
         product(