openstef 3.4.53__tar.gz → 3.4.55__tar.gz

{openstef-3.4.53 → openstef-3.4.55}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: openstef
-Version: 3.4.53
+Version: 3.4.55
 Summary: Open short term energy forecaster
 Home-page: https://github.com/OpenSTEF/openstef
 Author: Alliander N.V

{openstef-3.4.53 → openstef-3.4.55}/openstef/data_classes/prediction_job.py

@@ -9,7 +9,7 @@ from pydantic.v1 import BaseModel
 from openstef.data_classes.data_prep import DataPrepDataClass
 from openstef.data_classes.model_specifications import ModelSpecificationDataClass
 from openstef.data_classes.split_function import SplitFuncDataClass
-from openstef.enums import PipelineType, BiddingZone
+from openstef.enums import PipelineType, BiddingZone, AggregateFunction
 
 
 class PredictionJobDataClass(BaseModel):
@@ -83,6 +83,8 @@ class PredictionJobDataClass(BaseModel):
     data_balancing_ratio: Optional[float] = None
     """If data balancing is enabled, the data will be balanced with data from 1 year 
     ago in the future."""
+    rolling_aggregate_features: Optional[list[AggregateFunction]] = None
+    """If not None, rolling aggregate(s) of load will be used as features in the model."""
     depends_on: Optional[list[Union[int, str]]]
     """Link to another prediction job on which this prediction job might depend."""
     sid: Optional[str]

{openstef-3.4.53 → openstef-3.4.55}/openstef/enums.py

@@ -108,6 +108,13 @@ class BiddingZone(Enum):
     DE_AMP_LU = "DE_AMP_LU"
 
 
+class AggregateFunction(Enum):
+    MEAN = "mean"
+    MEDIAN = "median"
+    MAX = "max"
+    MIN = "min"
+
+
 class ModelType(Enum):
     XGB = "xgb"
     XGB_QUANTILE = "xgb_quantile"

{openstef-3.4.53 → openstef-3.4.55}/openstef/feature_engineering/apply_features.py

@@ -22,6 +22,7 @@ from openstef.feature_engineering.lag_features import generate_lag_feature_funct
 from openstef.feature_engineering.bidding_zone_to_country_mapping import (
     BIDDING_ZONE_TO_COUNTRY_CODE_MAPPING,
 )
+from openstef.feature_engineering.rolling_features import add_rolling_aggregate_features
 from openstef.feature_engineering.weather_features import (
     add_additional_solar_features,
     add_additional_wind_features,
@@ -130,5 +131,8 @@ def apply_features(
     # Adds daylight terrestrial feature
     data = add_daylight_terrestrial_feature(data)
 
+    if pj.get("rolling_aggregate_features") is not None:
+        data = add_rolling_aggregate_features(data, pj=pj)
+
     # Return dataframe including all requested features
     return data

{openstef-3.4.53 → openstef-3.4.55}/openstef/feature_engineering/feature_applicator.py

@@ -24,7 +24,9 @@ from openstef.feature_engineering.general import (
     remove_non_requested_feature_columns,
 )
 
-LATENCY_CONFIG = {"APX": 24}  # A specific latency is part of a specific feature.
+LATENCY_CONFIG = {
+    "day_ahead_electricity_price": 24
+}  # A specific latency is part of a specific feature.
 
 
 class AbstractFeatureApplicator(ABC):
@@ -94,7 +96,7 @@ class TrainFeatureApplicator(AbstractFeatureApplicator):
                 if not specified a default location is used
             latency_config: (Optional) Invalidate certain features that are not
                 available for a specific horizon due to data latency. Defaults to
-                ``{"APX": 24}``.
+                ``{"day_ahead_electricity_price": 24}``.
 
         Returns:
             Input DataFrame with an extra column for every added feature and sorted on the datetime index.

openstef-3.4.55/openstef/feature_engineering/rolling_features.py

@@ -0,0 +1,43 @@
+# SPDX-FileCopyrightText: 2017-2023 Contributors to the OpenSTEF project <korte.termijn.prognoses@alliander.com> # noqa E501>
+#
+# SPDX-License-Identifier: MPL-2.0
+from datetime import timedelta
+
+import pandas as pd
+
+from openstef.data_classes.prediction_job import PredictionJobDataClass
+
+
+def add_rolling_aggregate_features(
+    data: pd.DataFrame,
+    pj: PredictionJobDataClass,
+    rolling_window: timedelta = timedelta(hours=24),
+) -> pd.DataFrame:
+    """
+    Adds rolling aggregate features to the input dataframe.
+
+    These features are calculated with an aggregation over a rolling window of the data.
+    A list of requested features is used to determine whether to add the rolling features
+    or not.
+
+    Args:
+        data: Input dataframe to which the rolling features will be added.
+        pj: Prediction job data.
+        rolling_window: Rolling window size for the aggregation.
+
+    Returns:
+        DataFrame with added rolling features.
+    """
+    # Ensure the index is a DatetimeIndex
+    if not isinstance(data.index, pd.DatetimeIndex):
+        raise ValueError("The DataFrame index must be a DatetimeIndex.")
+
+    if "load" not in data.columns:
+        raise ValueError("The DataFrame must contain a 'load' column.")
+    rolling_window_load = data["load"].rolling(window=rolling_window)
+
+    for aggregate_func in pj["rolling_aggregate_features"]:
+        data[
+            f"rolling_{aggregate_func.value}_load_{rolling_window}"
+        ] = rolling_window_load.aggregate(aggregate_func.value)
+    return data

{openstef-3.4.53 → openstef-3.4.55}/openstef/model/regressors/linear_quantile.py

@@ -93,7 +93,7 @@ class LinearQuantileOpenstfRegressor(OpenstfRegressor, RegressorMixin):
             )
 
         if clipped_features is None:
-            clipped_features = ["APX"]
+            clipped_features = ["day_ahead_electricity_price"]
 
         self.quantiles = quantiles
         self.alpha = alpha

{openstef-3.4.53 → openstef-3.4.55}/openstef/tasks/create_components_forecast.py

@@ -6,7 +6,7 @@
 This code assumes trained models are available from the persistent storage.
 If these are not available run model_train.py to train all models.
 To provide the prognoses the following steps are carried out:
-  1. Get historic training data (TDCV, Load, Weather and APX price data)
+  1. Get historic training data (TDCV, Load, Weather and day_ahead_electricity_price price data)
   2. Apply features
   3. Load model
   4. Make component prediction

{openstef-3.4.53 → openstef-3.4.55}/openstef/tasks/create_forecast.py

@@ -6,7 +6,7 @@
 This code assumes trained models are available from the persistent storage. If these
 are not available run model_train.py to train all models.
 To provide the prognoses the folowing steps are carried out:
-  1. Get historic training data (TDCV, Load, Weather and APX price data)
+  1. Get historic training data (TDCV, Load, Weather and day_ahead_electricity_price price data)
   2. Apply features
   3. Load model
   4. Make prediction
@@ -24,7 +24,7 @@ from datetime import datetime, timedelta
 from pathlib import Path
 
 from openstef.data_classes.prediction_job import PredictionJobDataClass
-from openstef.enums import ModelType, PipelineType
+from openstef.enums import BiddingZone, ModelType, PipelineType
 from openstef.exceptions import InputDataOngoingZeroFlatlinerError
 from openstef.pipeline.create_forecast import create_forecast_pipeline
 from openstef.tasks.utils.predictionjobloop import PredictionJobLoop
@@ -82,8 +82,13 @@ def create_forecast_task(
         location=[pj["lat"], pj["lon"]],
         datetime_start=datetime_start,
         datetime_end=datetime_end,
+        market_price=pj.electricity_bidding_zone.value,
     )
 
+    # Add APX price to the input data for backward compatibility,remove this line when all models are retrained
+    if pj.electricity_bidding_zone == BiddingZone.NL:
+        input_data["APX"] = input_data["day_ahead_electricity_price"]
+
     try:
         # Make forecast with the forecast pipeline
         forecast = create_forecast_pipeline(

{openstef-3.4.53 → openstef-3.4.55}/openstef/tasks/train_model.py

@@ -4,7 +4,7 @@
 """This module contains the CRON job that is periodically executed to retrain the prognosis models.
 
 For this the folowing steps are caried out:
-  1. Get historic training data (TDCV, Load, Weather and APX price data)
+  1. Get historic training data (TDCV, Load, Weather and day_ahead_electricity_price price data)
   2. Apply features
   3. Train and Test the new model
   4. Check if new model performs better than the old model

{openstef-3.4.53 → openstef-3.4.55}/openstef/validation/validation.py

@@ -313,9 +313,9 @@ def calc_completeness_dataframe(
                     expected_numbers_timedelayed=value,
                 )
 
-        # Correct for APX being only expected to be available up to 24h
-        if "APX" in non_na_count.index:
-            non_na_count["APX"] += max([len(df) - 96, 0])
+        # Correct for day_ahead_electricity_price being only expected to be available up to 24h
+        if "day_ahead_electricity_price" in non_na_count.index:
+            non_na_count["day_ahead_electricity_price"] += max([len(df) - 96, 0])
 
         completeness_per_column_dataframe = non_na_count / (len(df))
 

{openstef-3.4.53 → openstef-3.4.55}/openstef.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: openstef
-Version: 3.4.53
+Version: 3.4.55
 Summary: Open short term energy forecaster
 Home-page: https://github.com/OpenSTEF/openstef
 Author: Alliander N.V

{openstef-3.4.53 → openstef-3.4.55}/openstef.egg-info/SOURCES.txt

@@ -40,6 +40,7 @@ openstef/feature_engineering/general.py
 openstef/feature_engineering/holiday_features.py
 openstef/feature_engineering/lag_features.py
 openstef/feature_engineering/missing_values_transformer.py
+openstef/feature_engineering/rolling_features.py
 openstef/feature_engineering/weather_features.py
 openstef/metrics/__init__.py
 openstef/metrics/figure.py

{openstef-3.4.53 → openstef-3.4.55}/setup.py

@@ -29,7 +29,7 @@ def read_long_description_from_readme():
 
 setup(
     name="openstef",
-    version="3.4.53",
+    version="3.4.55",
     packages=find_packages(include=["openstef", "openstef.*"]),
     description="Open short term energy forecaster",
     long_description=read_long_description_from_readme(),