[refactoring] Data validation unification (#46)

* replace assert with ValueError

* format code and docstrings

* rename variables and functions

* extract raising Value Error to a separate module

* throw DataValidationError for non user input validation

* correct docstrings in metrics.py

* Update error_utils.py

---------

Co-authored-by: LeonieFreisinger <[email protected]>

tests/test_evaluation.py

@@ -125,11 +125,11 @@ def test_evaluation_by_ID_for_forecast_step_invalid_input():
         benchmark.fcst_train[0]
     )  # ensure ID column in dataframe with single time series
     # calculate metrics by ID for selected forecast step
-    with pytest.raises(AssertionError):
+    with pytest.raises(ValueError):
         calculate_metrics_by_ID_for_forecast_step(
             fcst_df=fcst_test_peyton, df_historic=fcst_train_peyton, forecast_step_in_focus=1, freq="D"
         )
-    with pytest.raises(AssertionError):
+    with pytest.raises(ValueError):
         calculate_metrics_by_ID_for_forecast_step(
             fcst_df=fcst_test_peyton,
             df_historic=fcst_train_peyton,

tests/test_models.py

@@ -186,7 +186,7 @@ def test_seasonal_naive_model(dataset_input, model_classes_and_params_input):
 
 @pytest.mark.parametrize(*decorator_input)
 def test_seasonal_naive_model_invalid_input(dataset_input, model_classes_and_params_input):
-    log.info("Test invalid model input - Raise Assertion")
+    log.info("Test invalid model input - Raise ValueError")
     peyton_manning_df = pd.read_csv(PEYTON_FILE, nrows=NROWS)
     dataset_list = [
         Dataset(
@@ -209,7 +209,7 @@ def test_seasonal_naive_model_invalid_input(dataset_input, model_classes_and_par
         num_processes=1,
     )
 
-    with pytest.raises(AssertionError):
+    with pytest.raises(ValueError):
         _, _ = benchmark.run()
 
     log.info("#### Done with test_seasonal_naive_model_invalid_input")
@@ -345,6 +345,6 @@ def test_check_min_input_len():
         save_dir=SAVE_DIR,
         num_processes=1,
     )
-    with pytest.raises(AssertionError):
+    with pytest.raises(ValueError):
         results_train, results_test = benchmark.run()
     log.info("#### test_check_min_input_len")

tot/benchmark.py

@@ -11,6 +11,7 @@
 import pandas as pd
 
 from tot.datasets.dataset import Dataset
+from tot.error_utils import raise_if
 from tot.experiment import CrossValidationExperiment, Experiment, SimpleExperiment
 from tot.models.models import Model
 
@@ -98,8 +99,10 @@ def run(self, verbose=True):
                 log.info("exp {}/{}: {}".format(i + 1, len(self.experiments), exp.experiment_name))
         log.info("---- Staring Series of {} Experiments ----".format(len(self.experiments)))
         if self.num_processes > 1 and len(self.experiments) > 1:
-            if not all([exp.num_processes == 1 for exp in self.experiments]):
-                raise ValueError("can not set multiprocessing in experiments and Benchmark.")
+            raise_if(
+                not all([exp.num_processes == 1 for exp in self.experiments]),
+                "Cannot set multiprocessing in " "Experiments and Benchmark.",
+            )
             with Pool(self.num_processes) as pool:
                 args_list = [(exp, verbose, i + 1) for i, exp in enumerate(self.experiments)]
                 pool.map_async(