Merge pull request #2 from views-platform/test_ensemble_manager

Test ensemble manager

documentation/ADRs/012_artifact_naming_convention.md

@@ -18,7 +18,7 @@ The naming convention for a single model artifact will follow this structure:
 ```
 <run_type>_model_<timestamp>.pkl
 ```
-- run_type: The type of run (e.g., calibration, testing, forecasting).
+- run_type: The type of run (e.g., calibration, validation, forecasting).
 - timestamp: The timestamp when the model was trained. The format is`YYYYMMDD_HHMMSS`.
 
 Training is not an option for ensemble models right now as they are made by aggregating single models. So no decision has been made on the naming convention for ensemble models yet.

documentation/ADRs/013_output_naming_convention.md

@@ -20,15 +20,15 @@ The evaluation/output naming convention for using a single model will follow thi
 <evaluation/output>_<max_forecast_step>_<run_type>_<timestamp>.pkl
 ```
 - max_forecast_step: The maximum forecast step for the prediction.
-- run_type: The type of run (e.g., calibration, testing).
+- run_type: The type of run (e.g., calibration, validation).
 - timestamp: The timestamp when the model was trained. The format is`YYYYMMDD_HHMMSS`.
 
 The evaluation naming convention for using an ensemble model will follow this structure:
 ```
 <evaluation/output>_<max_forecast_step>_<run_type>_<model_name_1><timestamp_1>_<model_name2><timestamp_2>.pkl
 ```
 - max_forecast_step: The maximum forecast step for the prediction.
-- run_type: The type of run (e.g., calibration, testing).
+- run_type: The type of run (e.g., calibration, validation).
 - model_name: The name of the model used for the ensemble prediction.
 - timestamp: The timestamp when the model was trained. The format is`YYYYMMDD_HHMMSS`.
 

tests/test_cli_utils.py

@@ -25,14 +25,14 @@ def test_parse_args_run_type_calibration():
 
     assert args.run_type == "calibration"
 
-def test_parse_args_run_type_testing():
+def test_parse_args_run_type_validation():
     """
-    Test the run_type argument with value 'testing'.
+    Test the run_type argument with value 'validation'.
     """
-    sys.argv = ["main.py", "--run_type", "testing"]
+    sys.argv = ["main.py", "--run_type", "validation"]
     args = parse_args()
 
-    assert args.run_type == "testing"
+    assert args.run_type == "validation"
 
 def test_parse_args_run_type_forecasting():
     """
@@ -87,7 +87,7 @@ def test_parse_args_invalid_run_type_with_sweep():
     """
     Test invalid combination of --run_type and --sweep flags.
     """
-    sys.argv = ["main.py", "--run_type", "testing", "--sweep"]
+    sys.argv = ["main.py", "--run_type", "validation", "--sweep"]
     with pytest.raises(SystemExit):
         args = parse_args()
         validate_arguments(args)
@@ -104,7 +104,7 @@ def test_validate_arguments_sweep_with_non_calibration():
     """
     Test validate_arguments with --sweep and non-calibration run_type.
     """
-    sys.argv = ["main.py", "--run_type", "testing", "--sweep"]
+    sys.argv = ["main.py", "--run_type", "validation", "--sweep"]
     args = parse_args()
     with pytest.raises(SystemExit):
         validate_arguments(args)