Fixes bug when using historical forecast with discrete lags < -1

CHANGELOG.md

@@ -28,6 +28,7 @@ but cannot always guarantee backwards compatibility. Changes that may **break co
 - 🔴 / 🟢 Fixed a bug which raised an error when loading torch models that were saved with Darts versions < 0.33.0. This is a breaking change and models saved with version 0.33.0 will not be loadable anymore. [#2692](https://github.com/unit8co/darts/pull/2692) by [Dennis Bader](https://github.com/dennisbader).
 - Fixed a bug in `StaticCovariatesTransformer` which raised an error when trying to inverse transform one-hot encoded categorical static covariates with identical values across time-series. Each categorical static covariates is now referred to by `{covariate_name}_{category_name}`, regardless of the number of categories. [#2710](https://github.com/unit8co/darts/pull/2710) by [Antoine Madrona](https://github.com/madtoinou)
 - Fixed a bug in `13-TFT-examples.ipynb` where two calls to `TimeSeries.from_series()` were not providing `series` but `pd.Index`. The method calls were changed to `TimeSeries.from_values()`. [#2719](https://github.com/unit8co/darts/pull/2719) by [Jules Authier](https://github.com/authierj)
+- Fixed a bug in `RegressionModel` where performing optimized historical forecasts with `max(lags) < -1` resulted in forecasts that extended too far into the future. [#2715](https://github.com/unit8co/darts/pull/2715) by [Jules Authier](https://github.com/authierj)
 
 **Dependencies**
 

darts/tests/utils/historical_forecasts/test_historical_forecasts.py

@@ -72,6 +72,8 @@
         {},
         (5, 3),
     ),
+    (LinearRegressionModel, {"lags": [-5]}, {}, (5, 1)),
+    (LinearRegressionModel, {"lags": [-5], "output_chunk_shift": 1}, {}, (5, 2)),
 ]
 if not isinstance(CatBoostModel, NotImportedModule):
     models_reg_no_cov_cls_kwargs.append((
@@ -665,7 +667,7 @@ def test_historical_forecasts_negative_rangeindex(self):
     def test_historical_forecasts(self, config):
         """Tests historical forecasts with retraining for expected forecast lengths and times"""
         forecast_horizon = 8
-        # if no fit and retrain=false, should fit at fist iteration
+        # if no fit and retrain=false, should fit at first iteration
         model_cls, kwargs, model_kwarg, bounds = config
         model = model_cls(**kwargs, **model_kwarg)
         # set train length to be the minimum required training length
@@ -1248,11 +1250,12 @@ def test_regression_auto_start_multiple_no_cov(self, config):
             [ts_univariate, ts_multivariate],
             models_reg_no_cov_cls_kwargs + models_reg_cov_cls_kwargs,
             [True, False],
+            [True, False],
             [1, 5],
         ),
     )
     def test_optimized_historical_forecasts_regression(self, config):
-        ts, model_config, multi_models, forecast_horizon = config
+        ts, model_config, multi_models, overlap_end, forecast_horizon = config
         # slightly longer to not affect the last predictable timestamp
         ts_covs = self.ts_covs
         start = 14
@@ -1301,6 +1304,7 @@ def test_optimized_historical_forecasts_regression(self, config):
                         last_points_only=last_points_only,
                         stride=stride,
                         forecast_horizon=forecast_horizon,
+                        overlap_end=overlap_end,
                         enable_optimization=False,
                     )
 
@@ -1317,6 +1321,7 @@ def test_optimized_historical_forecasts_regression(self, config):
                         last_points_only=last_points_only,
                         stride=stride,
                         forecast_horizon=forecast_horizon,
+                        overlap_end=overlap_end,
                     )
 
                     self.helper_compare_hf(hist_fct, opti_hist_fct)

darts/utils/historical_forecasts/utils.py

@@ -782,8 +782,7 @@ def _adjust_historical_forecasts_time_index(
     show_warnings: bool,
 ) -> TimeIndex:
     """
-    Shrink the beginning and end of the historical forecasts time index based on the values of `start`,
-    `forecast_horizon` and `overlap_end`.
+    Shrink the beginning and end of the historical forecasts time index based on the value of `start`.
     """
     # retrieve actual start
     # when applicable, shift the start of the forecastable index based on `start`
@@ -992,7 +991,7 @@ def _get_historical_forecast_boundaries(
         overlap_end,
     )
 
-    # adjust boundaries based on start, forecast_horizon and overlap_end
+    # adjust boundaries based on start
     historical_forecasts_time_index = _adjust_historical_forecasts_time_index(
         series=series,
         series_idx=series_idx,
@@ -1020,7 +1019,13 @@ def _get_historical_forecast_boundaries(
     hist_fct_tgt_start, hist_fct_tgt_end = historical_forecasts_time_index
     if min_target_lag is not None:
         hist_fct_tgt_start += min_target_lag * freq
-    hist_fct_tgt_end -= 1 * freq
+
+    # target lag has a gap between the max lag and the present
+    if hasattr(model, "lags") and model._get_lags("target"):
+        hist_fct_tgt_end += 1 * freq * model._get_lags("target")[-1]
+    else:
+        hist_fct_tgt_end -= 1 * freq
+
     # past lags are <= 0
     hist_fct_pc_start, hist_fct_pc_end = historical_forecasts_time_index
     if min_past_cov_lag is not None: