feat(tasks)!: remove the tasks subpackage; relocate ENTSO-E CLI to spotforecast2.entsoe_cli

_quarto.yml

@@ -130,10 +130,10 @@ website:
             contents:
               - text: "autocorrelation"
                 file: docs/reference/stats.autocorrelation.qmd
-          - section: "Tasks"
+          - section: "ENTSO-E CLI"
             contents:
-              - text: "task_entsoe"
-                file: docs/reference/tasks.task_entsoe.qmd
+              - text: "entsoe_cli"
+                file: docs/reference/entsoe_cli.qmd
 
       - section: "Processing Guides"
         contents:
@@ -251,10 +251,10 @@ quartodoc:
       contents:
         - stats.autocorrelation
 
-    - title: "Tasks"
-      desc: "ENTSO-E forecasting task and CLI."
+    - title: "ENTSO-E CLI"
+      desc: "ENTSO-E download / train / predict command-line pipeline."
       contents:
-        - tasks.task_entsoe
+        - entsoe_cli
 
     - title: "Warnings"
       desc: "Warning-style configuration for spotforecast2."

docs/multitask/entsoe.qmd

@@ -30,9 +30,11 @@ import pandas as pd
 warnings.filterwarnings("ignore")
 
 from spotforecast2_safe.configurator import ConfigEntsoe
+from spotforecast2_safe.multitask.factories import (
+    default_lgbm_forecaster_factory as entsoe_lgbm_factory,
+)
 
 from spotforecast2.multitask.multi import MultiTask
-from spotforecast2.tasks.task_entsoe import entsoe_lgbm_factory
 
 CACHE_HOME = tempfile.mkdtemp()
 ```

docs/tasks/entsoe.qmd

@@ -275,107 +275,94 @@ print(ts_clean.values)  # [1.0, 2.0, 3.0, 4.0, 5.0]
 
 ## Forecaster Models
 
+The pipeline builds its forecasters through *factory* functions that take a
+`ConfigEntsoe` and return a ready-to-fit
+`spotforecast2_safe.forecaster.recursive.ForecasterRecursive`.
+
 ### LightGBM Forecaster
 
-Create a LightGBM-based recursive forecaster:
+The stock LightGBM factory lives in the safe package:
 
-```python
-from spotforecast2.tasks.task_entsoe import ForecasterRecursiveLGBM, config
+```{python}
+from spotforecast2_safe.configurator import ConfigEntsoe
+from spotforecast2_safe.multitask.factories import default_lgbm_forecaster_factory
 
-model = ForecasterRecursiveLGBM(iteration=1)
+config = ConfigEntsoe()
+forecaster = default_lgbm_forecaster_factory(config)
 
-print(model.name)             # 'lgbm'
-print(model.random_state)     # 314159 (from config)
-print(len(model.preprocessor.periods))  # 5 (from config)
+print(type(forecaster).__name__)
+print("lags:", len(forecaster.lags))
 ```
 
 ### XGBoost Forecaster
 
-Create an XGBoost-based recursive forecaster:
-
-```python
-from spotforecast2.tasks.task_entsoe import ForecasterRecursiveXGB, config
-
-model = ForecasterRecursiveXGB(iteration=1, lags=24)
+The XGBoost variant ships with the CLI module (xgboost is not a
+safe-package dependency):
 
-print(model.name)  # 'xgb'
-```
-
-### Custom Configuration Forecaster
-
-Override default configuration values:
-
-```python
-from spotforecast2.tasks.task_entsoe import ForecasterRecursiveLGBM
-from spotforecast2_safe.data import Period
+```{python}
+from spotforecast2_safe.configurator import ConfigEntsoe
 
-custom_periods = [
-    Period(name='hourly', n_periods=24, column='hour', input_range=(1, 24)),
-]
+from spotforecast2.entsoe_cli import entsoe_xgb_factory
 
-model = ForecasterRecursiveLGBM(
-    iteration=1,
-    lags=48,
-    periods=custom_periods,
-    country_code='FR',
-    random_state=42
-)
+config = ConfigEntsoe()
+forecaster = entsoe_xgb_factory(config)
 
-print(len(model.preprocessor.periods))  # 1
-print(model.preprocessor.country_code)  # 'FR'
+print(type(forecaster.estimator).__name__)
 ```
 
----
+Both factories honour `config.random_state`, `config.lags_consider`, and
+`config.window_size`; supply your own factory through
+`config.forecaster_factory` to customise further (see the Multitask
+tutorial).
 
 ## Using the Python API (Notebooks & Quarto)
 
 ### Full Prediction Pipeline
 
-For users working in Jupyter Notebooks or Quarto, the entire ENTSO-E pipeline can be executed using the Python API. This approach is highly recommended for safety-critical research as it allows for precise control over time windows and hyperparameters.
+For users working in Jupyter Notebooks or Quarto, the entire ENTSO-E pipeline
+can be executed through `MultiTask` — the same path the CLI's `train` and
+`predict` subcommands take. This approach is recommended for research as it
+gives precise control over time windows and hyperparameters.
 
 ```python
-import pandas as pd
+import logging
 import os
+
+from spotforecast2_safe.configurator import ConfigEntsoe
+from spotforecast2_safe.data.entsoe_loader import entsoe_data_loader
+from spotforecast2_safe.data.fetch_data import get_cache_home
 from spotforecast2_safe.downloader.entsoe import download_new_data
-from spotforecast2_safe.manager.trainer import handle_training as handle_training_safe
-from spotforecast2_safe.manager.predictor import get_model_prediction as get_model_prediction_safe
-from spotforecast2.plots.plotter import make_plot
-from spotforecast2.tasks.task_entsoe import ForecasterRecursiveLGBM
-
-# 1. Setup Time Windows (Last 3 years until last month) and country:
-country_code = "ES"  
-now = pd.Timestamp.now(tz='UTC').floor('D')
-current_month_start = now.replace(day=1)
-last_month_start = (current_month_start - pd.Timedelta(days=1)).replace(day=1)
-
-# 2. Download Data (Optional, requires ENTSOE_API_KEY)
+from spotforecast2_safe.multitask.factories import default_lgbm_forecaster_factory
+
+from spotforecast2.multitask import MultiTask
+
+# 1. Download data (optional, requires ENTSOE_API_KEY)
 api_key = os.environ.get("ENTSOE_API_KEY")
 if api_key:
-    download_new_data(api_key=api_key, start="202301010000", country_code=country_code)
-
-# 3. Configure and Train
-# Explicit parameters override global configuration for reproducibility
-model_class = ForecasterRecursiveLGBM
-model_name = "lgbm_advanced"
-
-handle_training_safe(
-    model_class=model_class,
-    model_name=model_name,
-    train_size=pd.Timedelta(days=3 * 365),
-    end_dev=last_month_start.strftime("%Y-%m-%d %H:%M%z"),
-    country_code=country_code
-)
+    download_new_data(api_key=api_key, start="202301010000")
 
-# 4. Generate Predictions for the forecast horizon
-# The predictor will automatically load the model trained above
-predictions = get_model_prediction_safe(
-    model_name=model_name,
-    predict_size=24 * 31
+# 2. Wire the loader and factory into the config
+config = ConfigEntsoe()
+config.targets = ["Actual Load"]
+config.agg_weights = [1.0]
+config.bounds = [(-1e9, 1e9)]
+config.data_loader = entsoe_data_loader
+config.forecaster_factory = default_lgbm_forecaster_factory
+config.data_frame_name = "entsoe-lgbm"
+
+# 3. Run the five-step pipeline (task="defaults" trains; "predict" reuses
+#    the saved model)
+mt = MultiTask(
+    config,
+    task="defaults",
+    cache_home=get_cache_home(config.cache_home),
+    log_level=logging.ERROR,
 )
-
-# 5. Visualize Results
-if predictions:
-    make_plot(predictions)
+mt.prepare_data()
+mt.detect_outliers()
+mt.impute()
+mt.build_exogenous_features()
+mt.run(show=True)
 ```
 
 ---

pyproject.toml

@@ -34,16 +34,16 @@ dependencies = [
     # 22.0.0 made warm_start_lags the seed lag list itself (default
     # DEFAULT_WARM_START_LAGS, None disables) — consumed by SpotOptimStrategy.
     # 21.2.0 added max_time_spotoptim, forwarded as SpotOptim's max_time.
-    "spotforecast2-safe>=22.0.0,<23",
+    "spotforecast2-safe>=22.3.0,<23",
     # spotoptim 1.0 is sequential-only and lean: torch/tensorboard moved to its
     # ``[torch]`` extra. sf2 forwards tensorboard_* kwargs into SpotOptim, so we
     # pin the extra to keep the TensorBoard tuning dashboards working (they were
     # always available via spotoptim's old hard torch dependency).
     "spotoptim[torch]>=1.0.0,<2",
     "tqdm>=4.67.2",
-    # Directly imported by spotforecast2.tasks.task_entsoe (and the xgb
-    # forecaster model). Previously satisfied transitively; declared explicitly
-    # now that spotoptim 1.0 no longer pulls it in.
+    # Directly imported by spotforecast2.entsoe_cli (and the xgb forecaster
+    # model). Previously satisfied transitively; declared explicitly now that
+    # spotoptim 1.0 no longer pulls it in.
     "xgboost>=3.2.0",
 ]
 
@@ -74,7 +74,7 @@ dev = [
 ]
 
 [project.scripts]
-spotforecast2-entsoe = "spotforecast2.tasks.task_entsoe:main"
+spotforecast2-entsoe = "spotforecast2.entsoe_cli:main"
 
 [tool.uv]
 # Accept pre-releases ONLY for dependencies whose specifier carries an explicit