remove all proloaf references

README.md

@@ -25,15 +25,20 @@ SPDX-License-Identifier: MPL-2.0
 OpenSTEF is a Python package designed for generating short-term forecasts in the energy sector. The repository includes all the essential components required for machine learning pipelines that facilitate the forecasting process. To utilize the package, users are required to furnish their own data storage and retrieval interface.
 
 # Table of contents
+- [OpenSTEF](#openstef)
+- [Table of contents](#table-of-contents)
 - [External information sources](#external-information-sources)
-- [Installation](install)
-- [Usage](usage)
-- [Reference Implementation](reference-implementation)
-- [Database connector for OpenSTEF](Openstef-dbc-Database-connector-for-openstef) 
-- [License](license)
-- [Licences third-party libraries](licenses-third-party-libraries)
-- [Contributing](contributing)
-- [Contact](contact)
+- [Installation](#installation)
+  - [Install the openstef package](#install-the-openstef-package)
+    - [Remark regarding installation within a **conda environment on Windows**:](#remark-regarding-installation-within-a-conda-environment-on-windows)
+- [Usage](#usage)
+  - [Reference Implementation](#reference-implementation)
+  - [Openstef-dbc - Database connector for openstef](#openstef-dbc---database-connector-for-openstef)
+  - [Example notebooks](#example-notebooks)
+  - [License](#license)
+  - [Licenses third-party libraries](#licenses-third-party-libraries)
+  - [Contributing](#contributing)
+  - [Contact](#contact)
 
 # External information sources
 - [Documentation website](https://openstef.github.io/openstef/index.html);
@@ -52,10 +57,6 @@ OpenSTEF is a Python package designed for generating short-term forecasts in the
 pip install openstef
 ```
 
-_**Optional**_: if you would like to use the proloaf model with OpenSTEF install the proloaf dependencies by running:
-```shell
-pip install openstef[proloaf]
-```
 ### Remark regarding installation within a **conda environment on Windows**:
 
 A version of the pywin32 package will be installed as a secondary dependency along with the installation of the openstef package. Since conda relies on an old version of pywin32, the new installation can break conda's functionality. The following command can solve this issue:

openstef/data_classes/prediction_job.py

@@ -25,7 +25,6 @@ class PredictionJobDataClass(BaseModel):
         - ``"xgb_quantile"``
         - ``"lgb"``
         - ``"linear"``
-        - ``"proloaf"`` (extra dependencies requiered, see README)
 
     If unsure what to pick, choose ``"xgb"``.
 

openstef/enums.py

@@ -10,7 +10,6 @@ class MLModelType(Enum):
     XGB_QUANTILE = "xgb_quantile"
     LGB = "lgb"
     LINEAR = "linear"
-    ProLoaf = "proloaf"
     ARIMA = "arima"
 
 

openstef/feature_engineering/apply_features.py

@@ -14,9 +14,6 @@
 import pandas as pd
 
 from openstef.data_classes.prediction_job import PredictionJobDataClass
-from openstef.feature_engineering.historic_features import (
-    add_historic_load_as_a_feature,
-)
 from openstef.feature_engineering.holiday_features import (
     generate_holiday_feature_functions,
 )
@@ -69,9 +66,6 @@ def apply_features(
                             np.random.uniform(0.7,1.7, 200)))
 
     """
-    # Add if needed the proloaf feature (historic_load)
-    data = add_historic_load_as_a_feature(data, pj)
-
     # Get lag feature functions
     feature_functions = generate_lag_feature_functions(feature_names, horizon)
 

openstef/feature_engineering/feature_applicator.py

@@ -149,11 +149,7 @@ def add_features(
 
         # NOTE this is required since apply_features could add additional features
         if self.feature_names is not None:
-            # Add horizon to requested features else it is removed, and if needed the proloaf feature (historic_load)
-            if pj.get("model") == "proloaf":
-                features = self.feature_names + ["historic_load"] + ["horizon"]
-            else:
-                features = self.feature_names + ["horizon"]
+            features = self.feature_names + ["horizon"]
             result = remove_non_requested_feature_columns(result, features)
 
         # Sort all features except for the (first) load and (last) horizon columns

openstef/model/model_creator.py

@@ -15,11 +15,6 @@
 from openstef.model.regressors.arima import ARIMAOpenstfRegressor
 
 logger = structlog.get_logger(__name__)
-try:
-    from openstef.model.regressors.proloaf import OpenstfProloafRegressor
-except ImportError:
-    logger.info("Proloaf not available, setting constructor to None")
-    OpenstfProloafRegressor = None
 
 valid_model_kwargs = {
     MLModelType.XGB: [
@@ -84,27 +79,6 @@
         "max_depth",
         "early_stopping_rounds",
     ],
-    MLModelType.ProLoaf: [
-        "relu_leak",
-        "encoder_features",
-        "decoder_features",
-        "core_layers",
-        "rel_linear_hidden_size",
-        "rel_core_hidden_size",
-        "dropout_fc",
-        "dropout_core",
-        "training_metric",
-        "metric_options",
-        "optimizer_name",
-        "early_stopping_patience",
-        "early_stopping_margin",
-        "learning_rate",
-        "max_epochs",
-        "device",
-        "batch_size",
-        "history_horizon",
-        "horizon_minutes",
-    ],
     MLModelType.LINEAR: [
         "missing_values",
         "imputation_strategy",
@@ -127,7 +101,6 @@ class ModelCreator:
         MLModelType.XGB: XGBOpenstfRegressor,
         MLModelType.LGB: LGBMOpenstfRegressor,
         MLModelType.XGB_QUANTILE: XGBQuantileOpenstfRegressor,
-        MLModelType.ProLoaf: OpenstfProloafRegressor,
         MLModelType.LINEAR: LinearOpenstfRegressor,
         MLModelType.ARIMA: ARIMAOpenstfRegressor,
     }

openstef/model/objective.py

@@ -2,7 +2,7 @@
 #
 # SPDX-License-Identifier: MPL-2.0
 import copy
-from datetime import datetime
+from datetime import datetime, timezone
 from typing import Any, Callable, Optional
 
 import optuna
@@ -59,7 +59,7 @@ def __init__(
         self.validation_data = None
         self.test_data = None
         self.model = model
-        self.start_time = datetime.utcnow()
+        self.start_time = datetime.now(timezone.utc)
         self.test_fraction = test_fraction
         self.validation_fraction = validation_fraction
         self.eval_metric = eval_metric
@@ -94,7 +94,7 @@ def __call__(
         split_args = self.split_args
         if split_args is None:
             split_args = {
-                "stratification_min_max": self.model_type != MLModelType.ProLoaf,
+                "stratification_min_max": True,
                 "back_test": True,
             }
         (
@@ -349,53 +349,6 @@ def get_pruning_callback(self, trial: optuna.trial.FrozenTrial):
         )
 
 
-class ProLoafRegressorObjective(RegressorObjective):
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-        self.model_type = MLModelType.ProLoaf
-
-    def get_params(self, trial: optuna.trial.FrozenTrial) -> dict:
-        """Get parameters for ProLoaf Regressor Objective with objective specific parameters.
-
-        Args: trial
-
-        Returns:
-            Dictionary with hyperparameter name as key and hyperparamer value as value.
-
-        """
-        # Filtered default parameters
-        model_params = super().get_params(trial)
-
-        # ProLoaf specific parameters
-        params = {
-            # TODO: look into optimizing this pipeline for proloaf
-            # "relu_leak": trial.suggest_float("relu_leak", 0.1, 1.0),
-            # "core_layers": trial.suggest_int("core_layers", 1, 3),
-            # "rel_linear_hidden_size": trial.suggest_float(
-            #    "rel_linear_hidden_size", 0.1, 1
-            # ),
-            # "rel_core_hidden_size": trial.suggest_float("rel_core_hidden_size", 0.1, 1),
-            # "dropout_fc": trial.suggest_float("dropout_fc", 0.1, 0.9),
-            # "dropout_core": trial.suggest_float("dropout_core", 0.1, 0.9),
-            # "early_stopping_patience": trial.suggest_int(
-            #    "early_stopping_patience", 5, 10
-            # ),
-            # "early_stopping_margin": trial.suggest_float(
-            #    "early_stopping_margin", 0.1, 0.9
-            # ),
-            "max_epochs": trial.suggest_int(
-                "max_epochs", 1, 1
-            ),  # TODO: change after having availability to gpu resource
-            "batch_size": trial.suggest_int("batch_size", 1, 24),
-        }
-        return {**model_params, **params}
-
-    def get_pruning_callback(self, trial: optuna.trial.FrozenTrial):
-        return optuna.integration.PyTorchLightningPruningCallback(
-            trial, monitor="val_loss"
-        )
-
-
 class LinearRegressorObjective(RegressorObjective):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)

openstef/model/objective_creator.py

@@ -8,7 +8,6 @@
 from openstef.model.objective import (
     LGBRegressorObjective,
     LinearRegressorObjective,
-    ProLoafRegressorObjective,
     RegressorObjective,
     XGBQuantileRegressorObjective,
     XGBRegressorObjective,
@@ -25,7 +24,6 @@ class ObjectiveCreator:
         MLModelType.XGB: XGBRegressorObjective,
         MLModelType.LGB: LGBRegressorObjective,
         MLModelType.XGB_QUANTILE: XGBQuantileRegressorObjective,
-        MLModelType.ProLoaf: ProLoafRegressorObjective,
         MLModelType.LINEAR: LinearRegressorObjective,
         MLModelType.ARIMA: ARIMARegressorObjective,
     }