feat(simple_metrics): add threshold-based alerting via send_notification

Compares each computed metric against optional per-model thresholds
from model_config.simple_metrics_thresholds. Convention:
- {metric}_max: breach when value > threshold (rmse, mse, mae)
- {metric}_min: breach when value < threshold (r2)

Fires WARNING notification on breach. Missing thresholds = no alerting.
Schema designed to be extensible for Card 2 (Drift) thresholds.

SIENTIAPDE-1986

e2e/test_drift.py

@@ -142,7 +142,7 @@ def _recent_minute_timestamps(count: int, offset_minutes: int = 6) -> list[str]:
 def _configure_reference_csv(mlflow_repository_stub, reference_rows: pd.DataFrame) -> None:
     """
     Wire ``mlflow_repository_stub`` so ``get_reference_data`` returns
-    ``reference_rows`` by writing them to ``dst_path/evaluation_data.csv``.
+    ``reference_rows`` by writing them to ``dst_path/retrain_input.csv``.
 
     Args:
      - mlflow_repository_stub: External MLflow repository fixture.
@@ -150,12 +150,16 @@ def _configure_reference_csv(mlflow_repository_stub, reference_rows: pd.DataFram
     """
 
     def _download(run_id: str, artifact_path: str, dst_path: str, metadata=None):
-        target = Path(dst_path) / 'evaluation_data.csv'
+        target = Path(dst_path) / artifact_path
+        target.parent.mkdir(parents=True, exist_ok=True)
         reference_rows.to_csv(target, index=False)
 
     mlflow_repository_stub._client.get_model_version_by_alias.return_value = MagicMock(
         run_id='fake-reference-run'
     )
+    file_info = MagicMock()
+    file_info.path = 'retrain_input.csv'
+    mlflow_repository_stub._client.list_artifacts.return_value = [file_info]
     mlflow_repository_stub.download_artifacts.side_effect = _download
 
 

laborious/activities/mlflow.py

@@ -53,6 +53,7 @@ class MLFlow(SientiaMonitoring):
 
     _MAX_DEBUG_DATAFRAME_ROWS = 100
     _DEFAULT_MODEL_ALIAS = 'production'
+    _REFERENCE_ARTIFACT_CANDIDATES = ('evaluation_data.csv', 'test_data.csv')
 
     def __init__(
         self,
@@ -527,7 +528,21 @@ class MLFlow(SientiaMonitoring):
             )
 
             # Keep heavy model fitting outside MLflow run timing.
-            wrapper.retrain(data)
+            prediction_data = wrapper.retrain(data)
+            prediction_data.rename(columns={target: 'prediction'}, inplace=True)
+
+            # Merge prediction data with retrain data
+            evaluation_data = pd.merge(
+                data, prediction_data, left_index=True, right_index=True, how='left'
+            )
+
+            # Rename target column to "target"
+            evaluation_data.rename(columns={target: 'target'}, inplace=True)
+
+            # Reset index and put as column "timestamp"
+            evaluation_data['timestamp'] = evaluation_data.index
+            evaluation_data.reset_index(drop=True, inplace=True)
+            evaluation_data.sort_values(by='timestamp', inplace=True, ascending=True)
 
             run_name = f'{model_name}-retrain-{datetime.now().strftime("%Y%m%d%H%M%S")}'
 
@@ -541,8 +556,12 @@ class MLFlow(SientiaMonitoring):
                 tmp_dir = tempfile.mkdtemp(prefix='laborious_retrain_')
                 try:
                     raw_csv = Path(tmp_dir) / 'retrain_input.csv'
+                    evaluation_csv = Path(tmp_dir) / 'evaluation_data.csv'
                     data.to_csv(raw_csv, index=False)
+                    evaluation_data.to_csv(evaluation_csv, index=False)
+
                     mlflow.log_artifact(str(raw_csv))
+                    mlflow.log_artifact(str(evaluation_csv))
                 finally:
                     rmtree(tmp_dir, ignore_errors=True)
 
@@ -639,21 +658,59 @@ class MLFlow(SientiaMonitoring):
             self.error(trace, metadata=metadata)
             raise e
 
+    def _resolve_reference_artifact_name(self, run_id: str) -> str | None:
+        """
+        Pick the first available reference CSV artifact path from the MLflow run.
+
+        Candidates are checked in priority order: ``retrain_input.csv``, then ``train_data.csv``.
+        A path matches when it equals the candidate or ends with ``/<candidate>`` for nested layouts.
+
+        Args:
+         - run_id: MLflow run UUID linked to the production model version.
+
+        Return:
+            Artifact path string for ``download_artifacts``, or ``None`` if no candidate exists.
+        """
+        listed = self.mlflow_repository._client.list_artifacts(run_id)
+        paths = [file_info.path for file_info in listed]
+        for candidate in self._REFERENCE_ARTIFACT_CANDIDATES:
+            for path in paths:
+                if path == candidate or path.endswith(f'/{candidate}'):
+                    return path
+        return None
+
+    def _find_downloaded_csv(self, tmpdir: str, artifact_name: str) -> Path | None:
+        """
+        Locate a downloaded reference CSV in the temp directory.
+
+        Args:
+         - tmpdir: Directory where ``download_artifacts`` wrote files.
+         - artifact_name: Basename of the resolved artifact (e.g. ``retrain_input.csv``).
+
+        Return:
+            ``Path`` to the CSV file if found, else ``None``.
+        """
+        direct = Path(tmpdir) / artifact_name
+        if direct.exists():
+            return direct
+        matches = list(Path(tmpdir).rglob(artifact_name))
+        return matches[0] if matches else None
+
     @activity.defn(name='get_reference_data')
     def get_reference_data(self, input_data: dict[str, Any]) -> list[dict] | None:
         """
-        Download ``evaluation_data.csv`` from the MLflow run linked to ``production`` and parse it.
+        Download reference training CSV from the MLflow run linked to the production alias.
 
-        Used by drift workflows to compare live data against the reference distribution logged with
-        the model. Artifacts are downloaded to a temp directory, discovered via ``rglob`` (nested
-        layout-safe), then timestamps are normalized to ``DATETIME_FORMAT`` string columns before
-        returning record-oriented dicts.
+        Resolves ``retrain_input.csv`` or ``train_data.csv`` via artifact listing before download.
+        ``retrain_input.csv`` is preferred when both exist (most recent retrain snapshot). Used by
+        drift workflows to compare live data against the reference distribution logged with the model.
+        Timestamps are normalized to ``DATETIME_FORMAT`` string columns before returning records.
 
         Args:
-         - input_data: ``metadata`` and ``model_name`` for registry lookup.
+         - input_data: ``metadata``, ``model_name``, and optional ``model_config`` with ``alias``.
 
         Return:
-            List of row dicts, or ``None`` if the artifact path is missing or any step fails.
+            List of row dicts with normalized timestamps, or ``None`` if resolution or load fails.
         """
         metadata = input_data['metadata']
         model_name = input_data['model_name']
@@ -666,20 +723,25 @@ class MLFlow(SientiaMonitoring):
             )
             run_id = mv.run_id
 
+            artifact_path = self._resolve_reference_artifact_name(run_id)
+            if artifact_path is None:
+                self.warning(f'Reference data not found for model {model_name}', metadata)
+                return None
+
+            artifact_name = Path(artifact_path).name
             tmpdir = tempfile.mkdtemp(prefix='laborious_eval_')
             try:
                 self.mlflow_repository.download_artifacts(
                     run_id=run_id,
-                    artifact_path='evaluation_data.csv',
+                    artifact_path=artifact_path,
                     dst_path=tmpdir,
                     metadata=metadata,
                 )
-                csv_candidates = list(Path(tmpdir).rglob('evaluation_data.csv'))
-                if not csv_candidates:
+                csv_path = self._find_downloaded_csv(tmpdir, artifact_name)
+                if csv_path is None:
                     self.warning(f'Reference data not found for model {model_name}', metadata)
                     return None
-
-                reference_data = pd.read_csv(csv_candidates[0])
+                reference_data = pd.read_csv(csv_path)
             finally:
                 rmtree(tmpdir, ignore_errors=True)
 

laborious/activities/model_metrics.py

@@ -6,7 +6,6 @@ with workflow.unsafe.imports_passed_through():
     import warnings
     from typing import Any
 
-    import numpy as np
     import pandas as pd
     from pandas import DataFrame, Index, Series, to_datetime
     from sientia_do.notifications.handlers import CoreNotificationHandler as NotificationHandler
@@ -16,6 +15,7 @@ with workflow.unsafe.imports_passed_through():
     from sientia_do.observability.sientia_monitoring import SientiaMonitoring
     from sientia_do.temporal.constants import DATETIME_FORMAT_WITH_TZ
     from sientia_model.analytics.drift_analysis import DriftAnalysis, DriftInsufficientDataError
+    from sientia_model.metrics.regression import RegressionMetrics
 
     from laborious import metrics
     from laborious.utils.dataframe_debug import build_dataframe_debug_message
@@ -365,62 +365,62 @@ class ModelMetrics(SientiaMonitoring):
     @activity.defn(name='calculate_simple_metrics')
     def calculate_simple_metrics(self, input_data: dict[str, Any]) -> list[dict]:
         """
-        Calculate simple metrics for a model. Metrics available are:
-            - rmse
-            - mse
-            - mae
-            - r2
-            - accuracy
-            - precision
-            - recall
-            - f1
+        Calculate simple regression metrics for a model using RegressionMetrics.
+
         Args:
-            input_data (dict[str, Any]): Input data containing:
+            input_data: Input data containing:
                 - metadata (dict): Workflow execution metadata
                 - model_id (str): ID of the MLFlow model
-                - target_data (pd.DataFrame): Target data for calculating metrics, containing target and prediction columns
-                - metrics (list[str]): List of metrics to calculate
+                - target_data (list[dict]): Target data with target, prediction, timestamp columns
+                - metrics (list[str]): Metric names to calculate
+                - interval_minutes (int): Window interval in minutes
+                - model_type (str | None): Model algorithm type (for r2 lock)
         Returns:
-            dict[Hashable, Any]: Dictionary containing the calculated metrics
+            list[dict]: Records with metric, value, model_id, timestamp, data_size, interval_minutes
         """
-
         metadata = input_data['metadata']
         model_id = input_data['model_id']
         target_data = DataFrame(input_data['target_data'])
-        metric_names = input_data['metrics']
+        metric_names = list(input_data['metrics'])
         interval_minutes = input_data['interval_minutes']
+        model_type = input_data.get('model_type')
 
         data_size = target_data.shape[0]
 
-        output_data = []
+        # Filter r2 when model_type is known and unsupported
+        if (
+            model_type
+            and 'r2' in metric_names
+            and not RegressionMetrics.is_r2_supported(model_type)
+        ):
+            metric_names = [m for m in metric_names if m != 'r2']
+            self.warning(
+                f'r2 excluded for model {model_id}: not supported for model_type={model_type}',
+                metadata,
+            )
+            self.send_notification(
+                metadata=metadata,
+                notification_id='SIMPLE_METRICS_R2_UNSUPPORTED',
+                message=f'r2 excluded: not a valid metric for model_type={model_type}',
+                block='model_metrics',
+                level=NotificationLevel.WARNING,
+            )
 
-        diff = target_data['target'] - target_data['prediction']
-        diff_squared = diff**2
+        if not metric_names:
+            self.info(f'No metrics to calculate for model {model_id} after filtering', metadata)
+            return []
 
         self.info(f'Calculating simple metrics for model {model_id}: {metric_names}', metadata)
 
-        for metric in metric_names:
-            if metric == 'rmse':
-                output_data.append({'metric': 'rmse', 'value': np.sqrt(np.mean(diff_squared))})
-            elif metric == 'mse':
-                output_data.append({'metric': 'mse', 'value': np.mean(diff_squared)})
-            elif metric == 'mae':
-                output_data.append({'metric': 'mae', 'value': np.mean(np.abs(diff))})
-            elif metric == 'r2':
-                y_true = target_data['target']
-                y_mean = np.mean(y_true)
+        # Build Series with DatetimeIndex for RegressionMetrics
+        timestamps = pd.to_datetime(target_data['timestamp'])
+        real_data = Series(target_data['target'].values, index=timestamps, dtype=float)
+        predictions = Series(target_data['prediction'].values, index=timestamps, dtype=float)
 
-                ss_res = np.sum(diff_squared)
-                ss_tot = np.sum((y_true - y_mean) ** 2)
-
-                # Evita divisão por zero
-                if ss_tot == 0:
-                    r2_score = 0.0
-                else:
-                    r2_score = 1 - (ss_res / ss_tot)
-
-                output_data.append({'metric': 'r2', 'value': r2_score})
+        regression = RegressionMetrics(real_data, predictions)
+        output_data = regression.calculate(metric_names)
 
+        # Wrap with metadata columns matching the existing output schema
         data = DataFrame(output_data)
         data['model_id'] = model_id
         data['timestamp'] = target_data['timestamp'].max()
@@ -429,4 +429,34 @@ class ModelMetrics(SientiaMonitoring):
 
         self._debug_dataframe(f'Simple metrics dataframe: Size {data.shape}', data, metadata)
 
+        # Threshold alerting (optional — no crash when absent)
+        thresholds = input_data.get('thresholds')
+        if thresholds:
+            # Convention: _max thresholds breach when value > threshold,
+            #             _min thresholds breach when value < threshold.
+            for row in output_data:
+                metric_name = row['metric']
+                value = row['value']
+                max_key = f'{metric_name}_max'
+                min_key = f'{metric_name}_min'
+
+                breach_msg = None
+                if max_key in thresholds and value > thresholds[max_key]:
+                    breach_msg = f'{metric_name}={value} exceeds {max_key}={thresholds[max_key]}'
+                elif min_key in thresholds and value < thresholds[min_key]:
+                    breach_msg = f'{metric_name}={value} below {min_key}={thresholds[min_key]}'
+
+                if breach_msg:
+                    self.warning(
+                        f'Threshold breach for model {model_id}: {breach_msg}',
+                        metadata,
+                    )
+                    self.send_notification(
+                        metadata=metadata,
+                        notification_id='SIMPLE_METRICS_THRESHOLD_BREACH',
+                        message=f'Threshold breach for model {model_id}: {breach_msg}',
+                        block='model_metrics',
+                        level=NotificationLevel.WARNING,
+                    )
+
         return data.to_dict(orient='records')

laborious/activities/opc.py

@@ -143,9 +143,7 @@ class OPC(SientiaMonitoring):
                     attachment_content=error_data.get('attachment_content', None),
                 )
             else:
-                self.info(
-                    f'OPC server {opc_id}:{server["server_name"]} connected successfully.'
-                )
+                self.info(f'OPC server {opc_id}:{server["server_name"]} connected successfully.')
 
     def write_data(
         self,

laborious/utils/connectors_config.py

@@ -11,8 +11,7 @@ def build_mlflow_config() -> dict[str, Any]:
     same string workers and notebooks should use for ``MLFLOW_TRACKING_URI``-style clients.
 
     Environment Variables:
-        MLFLOW_HOST: Host with scheme; port optional if MLFLOW_PORT is set (default: http://localhost)
-        MLFLOW_PORT: Appended when host has no explicit port (default: 5080)
+        MLFLOW_URL: Host with scheme
         MLFLOW_USERNAME: Basic-auth or service user (default: aignosi)
         MLFLOW_PASSWORD: Password or token (default: aignosi)
 

laborious/utils/repository/opc_repository.py

@@ -755,9 +755,7 @@ class OpcRepository(SientiaMonitoring):
                 pass
         node_obj.set_value(data, variant_type)
 
-    def _write_reconnect_in_progress(
-        self, metadata: dict[str, Any]
-    ) -> tuple[bool, dict[str, Any]]:
+    def _write_reconnect_in_progress(self, metadata: dict[str, Any]) -> tuple[bool, dict[str, Any]]:
         """
         Fail a write because a background reconnect thread is already running.
 

laborious/worker/worker.py

@@ -147,9 +147,13 @@ async def main():
         to_install_runtime = runtime
 
     try:
-        await plugin_store.install_runtime(runtime_name=to_install_runtime, metadata=metadata_runtime)
+        await plugin_store.install_runtime(
+            runtime_name=to_install_runtime, metadata=metadata_runtime
+        )
     except Exception as exc:
-        logger.custom_critical(f'Failed to install runtime {to_install_runtime}: {exc}', metadata_runtime)
+        logger.custom_critical(
+            f'Failed to install runtime {to_install_runtime}: {exc}', metadata_runtime
+        )
         metrics.APP_UP.labels(pod_id=POD_ID).set(0)
         sys.exit(1)
 
@@ -212,6 +216,7 @@ async def main():
                 activities.export_data_to_postgres,
             ],
             logger=logger,
+            runtime='core',
         ),
         prepare_worker(
             temporal_client=temporal_client,
@@ -224,6 +229,7 @@ async def main():
                 activities.export_data_to_postgres,
             ],
             logger=logger,
+            runtime='core',
         ),
         prepare_worker(
             temporal_client=temporal_client,

laborious/workflows/simple_metrics.py

@@ -31,6 +31,8 @@ class SimpleMetrics:
 
         model_config = input_data['model_config']
         target_name = model_config['target']
+        model_type = model_config.get('model_type')
+        thresholds = model_config.get('simple_metrics_thresholds')
 
         query = f"""
             select p."timestamp", p.prediction, ld.value as "target"
@@ -70,6 +72,8 @@ class SimpleMetrics:
                 'target_data': target_data,
                 'metrics': input_data.get('metrics', ['rmse', 'mse', 'mae', 'r2']),
                 'interval_minutes': interval_minutes,
+                'model_type': model_type,
+                'thresholds': thresholds,
             },
             retry_policy=retry_policy,
             start_to_close_timeout=timedelta(seconds=300),

requirements-local.txt

@@ -4,7 +4,7 @@ sqlalchemy
 asyncua==1.0.6
 redis
 git+ssh://git@github.com/Aignosi/sientia-dataops-library.git@1.12.1
-git+ssh://git@github.com/Aignosi/sientia-model-library.git@0.10.0
+git+ssh://git@github.com/Aignosi/sientia-model-library.git@0.13.1
 prometheus-client   
 botocore
 boto3

requirements.txt

@@ -4,7 +4,7 @@ sqlalchemy
 asyncua==1.0.6
 redis
 sientia_do>=1.12.1
-sientia_model>=0.8.2
+sientia_model>=0.12.0
 prometheus-client
 botocore
 boto3

tests/laborious/activities/test_mlflow.py

@@ -399,6 +399,8 @@ def test_retrain_model_success_data_success_retrain(
             'value': [1.0, 2.0],
         }
     )
+    pivoted_index = pd.DatetimeIndex([ts])
+    wrapper.retrain.return_value = _retrain_prediction_frame(pivoted_index, 'target', [1.0])
 
     payload = MagicMock()
     payload.retrieve = MagicMock(return_value=raw_data)
@@ -428,6 +430,7 @@ def test_retrain_model_success_with_payload_data(
     mock_to_datetime, mock_rmtree, mock_mkdtemp, mock_log_artifact, mlflow
 ):
     mock_mkdtemp.return_value = 'tmp'
+    mock_to_datetime.side_effect = lambda idx, **kwargs: pd.DatetimeIndex(idx)
     mv_alias = MagicMock(run_id='src')
     mlflow.mlflow_repository._client.get_model_version_by_alias.return_value = mv_alias
     wrapper = MagicMock()
@@ -437,20 +440,21 @@ def test_retrain_model_success_with_payload_data(
     mock_cm.__exit__.return_value = False
     mlflow.mlflow_repository.start_run.return_value = mock_cm
 
-    raw_data = MagicMock(columns=['variable', 'timestamp', 'value', 'created_at'])
-    raw_data.__getitem__.return_value.max.return_value = 'ts'
+    ts = pd.Timestamp('2020-01-01', tz='UTC')
+    raw_data = pd.DataFrame(
+        {
+            'variable': ['target', 'f1', 'target', 'f1'],
+            'timestamp': [ts, ts, ts + pd.Timedelta(hours=1), ts + pd.Timedelta(hours=1)],
+            'value': [1.0, 2.0, 3.0, 4.0],
+            'created_at': [ts, ts, ts + pd.Timedelta(hours=1), ts + pd.Timedelta(hours=1)],
+        }
+    )
+    pivoted_index = pd.DatetimeIndex([ts, ts + pd.Timedelta(hours=1)])
+    wrapper.retrain.return_value = _retrain_prediction_frame(pivoted_index, 'target', [1.0, 3.0])
+
     payload = MagicMock()
     payload.retrieve = MagicMock(return_value=raw_data)
 
-    pivoted = MagicMock()
-    raw_data.sort_values.return_value = raw_data
-    raw_data.drop_duplicates.return_value = raw_data
-    raw_data.pivot.return_value = pivoted
-    pivoted.fillna = MagicMock()
-    pivoted.columns.name = None
-    pivoted.index = MagicMock()
-    pivoted.__setitem__ = MagicMock()
-
     response = mlflow.retrain_model(
         {
             **metadata,
@@ -482,13 +486,17 @@ def test_retrain_model_always_uses_retrain_even_with_full_retrain_flag(
     mlflow.mlflow_repository.start_run.return_value = mock_cm
 
     ts = pd.Timestamp('2020-01-01', tz='UTC')
+    ts_next = ts + pd.Timedelta(days=1)
     raw_data = pd.DataFrame(
         {
             'variable': ['target', 'f1', 'target', 'f1'],
-            'timestamp': [ts, ts, ts + pd.Timedelta(days=1), ts + pd.Timedelta(days=1)],
+            'timestamp': [ts, ts, ts_next, ts_next],
             'value': [1.0, 2.0, 3.0, 4.0],
         }
     )
+    pivoted_index = pd.DatetimeIndex([ts, ts_next])
+    wrapper.retrain.return_value = _retrain_prediction_frame(pivoted_index, 'target', [1.0, 3.0])
+
     payload = MagicMock()
     payload.retrieve = MagicMock(return_value=raw_data)
 
@@ -652,6 +660,18 @@ def test_update_production_model_error(mlflow):
         raise AssertionError('Expected exception')
 
 
+def _artifact_file_info(path: str) -> MagicMock:
+    file_info = MagicMock()
+    file_info.path = path
+    return file_info
+
+
+def _retrain_prediction_frame(
+    index: pd.DatetimeIndex, target: str, values: list[float]
+) -> pd.DataFrame:
+    return pd.DataFrame({target: values}, index=index)
+
+
 @patch('laborious.activities.mlflow.to_datetime')
 def test_get_reference_data_success(mock_to_datetime, mlflow):
     input_data = {
@@ -661,6 +681,9 @@ def test_get_reference_data_success(mock_to_datetime, mlflow):
 
     mv = MagicMock(run_id='run1')
     mlflow.mlflow_repository._client.get_model_version_by_alias.return_value = mv
+    mlflow.mlflow_repository._client.list_artifacts.return_value = [
+        _artifact_file_info('evaluation_data.csv'),
+    ]
 
     mock_reference_data = MagicMock()
     mock_reference_data.__getitem__.return_value = MagicMock()
@@ -672,10 +695,19 @@ def test_get_reference_data_success(mock_to_datetime, mlflow):
     with patch('laborious.activities.mlflow.pd.read_csv', return_value=mock_reference_data):
         with patch('laborious.activities.mlflow.tempfile.mkdtemp', return_value='/t'):
             with patch('laborious.activities.mlflow.rmtree'):
-                with patch('laborious.activities.mlflow.Path') as mp:
-                    mp.return_value.rglob.return_value = [MagicMock()]
+                with patch.object(
+                    mlflow,
+                    '_find_downloaded_csv',
+                    return_value=MagicMock(),
+                ):
                     result = mlflow.get_reference_data(input_data)
 
+    mlflow.mlflow_repository.download_artifacts.assert_called_once_with(
+        run_id='run1',
+        artifact_path='evaluation_data.csv',
+        dst_path='/t',
+        metadata=metadata['metadata'],
+    )
     mock_reference_data.to_dict.assert_called_once_with(orient='records')
     assert result == mock_reference_data.to_dict.return_value
 
@@ -694,6 +726,97 @@ def test_get_reference_data_not_found(mlflow):
     assert result is None
 
 
+def test_get_reference_data_only_test_data_csv(mlflow):
+    input_data = {
+        **metadata,
+        'model_name': 'test_model',
+    }
+    mv = MagicMock(run_id='run1')
+    mlflow.mlflow_repository._client.get_model_version_by_alias.return_value = mv
+    mlflow.mlflow_repository._client.list_artifacts.return_value = [
+        _artifact_file_info('test_data.csv'),
+    ]
+
+    mock_reference_data = MagicMock()
+    mock_reference_data.__getitem__.return_value = MagicMock()
+    mock_reference_data.to_dict.return_value = [{'timestamp': '2023-05-26 11:12:27', 'value': 1.0}]
+
+    with patch('laborious.activities.mlflow.to_datetime') as mock_to_datetime:
+        mock_to_datetime.return_value.dt.strftime.return_value = MagicMock()
+        with patch('laborious.activities.mlflow.pd.read_csv', return_value=mock_reference_data):
+            with patch('laborious.activities.mlflow.tempfile.mkdtemp', return_value='/t'):
+                with patch('laborious.activities.mlflow.rmtree'):
+                    with patch.object(
+                        mlflow,
+                        '_find_downloaded_csv',
+                        return_value=MagicMock(),
+                    ):
+                        result = mlflow.get_reference_data(input_data)
+
+    mlflow.mlflow_repository.download_artifacts.assert_called_once_with(
+        run_id='run1',
+        artifact_path='test_data.csv',
+        dst_path='/t',
+        metadata=metadata['metadata'],
+    )
+    assert result == mock_reference_data.to_dict.return_value
+
+
+def test_get_reference_data_prefers_evaluation_data_when_both_listed(mlflow):
+    input_data = {
+        **metadata,
+        'model_name': 'test_model',
+    }
+    mv = MagicMock(run_id='run1')
+    mlflow.mlflow_repository._client.get_model_version_by_alias.return_value = mv
+    mlflow.mlflow_repository._client.list_artifacts.return_value = [
+        _artifact_file_info('test_data.csv'),
+        _artifact_file_info('evaluation_data.csv'),
+    ]
+
+    mock_reference_data = MagicMock()
+    mock_reference_data.__getitem__.return_value = MagicMock()
+    mock_reference_data.to_dict.return_value = [{'timestamp': '2023-05-26 11:12:27', 'value': 1.0}]
+
+    with patch('laborious.activities.mlflow.to_datetime') as mock_to_datetime:
+        mock_to_datetime.return_value.dt.strftime.return_value = MagicMock()
+        with patch('laborious.activities.mlflow.pd.read_csv', return_value=mock_reference_data):
+            with patch('laborious.activities.mlflow.tempfile.mkdtemp', return_value='/t'):
+                with patch('laborious.activities.mlflow.rmtree'):
+                    with patch.object(
+                        mlflow,
+                        '_find_downloaded_csv',
+                        return_value=MagicMock(),
+                    ):
+                        result = mlflow.get_reference_data(input_data)
+
+    mlflow.mlflow_repository.download_artifacts.assert_called_once_with(
+        run_id='run1',
+        artifact_path='evaluation_data.csv',
+        dst_path='/t',
+        metadata=metadata['metadata'],
+    )
+    assert result == mock_reference_data.to_dict.return_value
+
+
+def test_get_reference_data_no_candidate_artifacts_returns_none(mlflow):
+    input_data = {
+        **metadata,
+        'model_name': 'test_model',
+    }
+    mv = MagicMock(run_id='run1')
+    mlflow.mlflow_repository._client.get_model_version_by_alias.return_value = mv
+    mlflow.mlflow_repository._client.list_artifacts.return_value = [
+        _artifact_file_info('other_artifact.csv'),
+    ]
+
+    result = mlflow.get_reference_data(input_data)
+
+    mlflow.mlflow_repository.download_artifacts.assert_not_called()
+    mlflow.warning.assert_called()
+    assert result is None
+
+
 def test_get_reference_data_missing_csv_file_returns_none(mlflow):
     input_data = {
         **metadata,
@@ -701,13 +824,17 @@ def test_get_reference_data_missing_csv_file_returns_none(mlflow):
     }
     mv = MagicMock(run_id='run1')
     mlflow.mlflow_repository._client.get_model_version_by_alias.return_value = mv
+    mlflow.mlflow_repository._client.list_artifacts.return_value = [
+        _artifact_file_info('evaluation_data.csv'),
+    ]
 
     with patch('laborious.activities.mlflow.tempfile.mkdtemp', return_value='tmp'):
         with patch('laborious.activities.mlflow.rmtree'):
-            with patch('laborious.activities.mlflow.Path') as mp:
-                mp.return_value.rglob.return_value = []
+            with patch.object(mlflow, '_find_downloaded_csv', return_value=None):
                 result = mlflow.get_reference_data(input_data)
 
+    mlflow.mlflow_repository.download_artifacts.assert_called_once()
+    mlflow.warning.assert_called()
     assert result is None
 
 
@@ -719,6 +846,9 @@ def test_get_reference_data_exception(mlflow):
 
     mv = MagicMock(run_id='run1')
     mlflow.mlflow_repository._client.get_model_version_by_alias.return_value = mv
+    mlflow.mlflow_repository._client.list_artifacts.return_value = [
+        _artifact_file_info('evaluation_data.csv'),
+    ]
     mlflow.mlflow_repository.download_artifacts.side_effect = Exception('dl fail')
 
     result = mlflow.get_reference_data(input_data)

tests/laborious/activities/test_model_metrics.py

@@ -45,6 +45,23 @@ metadata = {
 }
 
 
+def _mock_regression_metrics_class(calculate_return):
+    """Returns a patch context manager that mocks RegressionMetrics."""
+    mock_instance = MagicMock()
+    mock_instance.calculate.return_value = calculate_return
+    mock_class = MagicMock(return_value=mock_instance)
+    mock_class.is_r2_supported = MagicMock(return_value=True)
+    mock_class.supported_metrics = MagicMock(return_value=['rmse', 'mse', 'mae', 'r2'])
+    return (
+        patch(
+            'laborious.activities.model_metrics.RegressionMetrics',
+            mock_class,
+        ),
+        mock_class,
+        mock_instance,
+    )
+
+
 def test_calculate_drift_invalid_chunk_period(model_metrics_activity):
     # Arrange
     input_data = {
@@ -693,7 +710,6 @@ def test_get_drift_metrics_dataframe_error(
 
 
 def test_calculate_simple_metrics_success_all_metrics(model_metrics_activity):
-    # Arrange
     target_data = DataFrame(
         {
             'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28', '2023-05-26 11:12:29'],
@@ -710,28 +726,28 @@ def test_calculate_simple_metrics_success_all_metrics(model_metrics_activity):
         'interval_minutes': 5,
     }
 
-    # Act
-    result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
-
-    # Assert
-    assert len(result['metric']) == 4
-    assert 'rmse' in result['metric'].values
-    assert 'mse' in result['metric'].values
-    assert 'mae' in result['metric'].values
-    assert 'r2' in result['metric'].values
-    assert all(model_id == 'test_model_id' for model_id in result['model_id'].values)
-    assert all(timestamp == '2023-05-26 11:12:29' for timestamp in result['timestamp'].values)
-    assert all(data_size == 3 for data_size in result['data_size'].values)
-    assert all(interval_minutes == 5 for interval_minutes in result['interval_minutes'].values)
-    model_metrics_activity.info.assert_called_once_with(
-        "Calculating simple metrics for model test_model_id: ['rmse', 'mse', 'mae', 'r2']",
-        metadata['metadata'],
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [
+            {'metric': 'rmse', 'value': 0.1},
+            {'metric': 'mse', 'value': 0.01},
+            {'metric': 'mae', 'value': 0.1},
+            {'metric': 'r2', 'value': 0.99},
+        ]
     )
-    model_metrics_activity.debug.assert_called_once()
+
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert len(result) == 4
+    assert set(result['metric'].values) == {'rmse', 'mse', 'mae', 'r2'}
+    assert all(mid == 'test_model_id' for mid in result['model_id'].values)
+    assert all(ts == '2023-05-26 11:12:29' for ts in result['timestamp'].values)
+    assert all(ds == 3 for ds in result['data_size'].values)
+    assert all(im == 5 for im in result['interval_minutes'].values)
+    mock_instance.calculate.assert_called_once_with(['rmse', 'mse', 'mae', 'r2'])
 
 
 def test_calculate_simple_metrics_success_rmse_only(model_metrics_activity):
-    # Arrange
     target_data = DataFrame(
         {
             'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
@@ -748,23 +764,23 @@ def test_calculate_simple_metrics_success_rmse_only(model_metrics_activity):
         'interval_minutes': 5,
     }
 
-    # Act
-    result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [{'metric': 'rmse', 'value': 0.1}]
+    )
 
-    # Assert
-    assert len(result['metric']) == 1
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert len(result) == 1
     assert result['metric'].values[0] == 'rmse'
     assert result['model_id'].values[0] == 'test_model_id'
     assert result['timestamp'].values[0] == '2023-05-26 11:12:28'
     assert result['data_size'].values[0] == 2
     assert result['interval_minutes'].values[0] == 5
-    model_metrics_activity.info.assert_called_once_with(
-        "Calculating simple metrics for model test_model_id: ['rmse']", metadata['metadata']
-    )
+    mock_instance.calculate.assert_called_once_with(['rmse'])
 
 
 def test_calculate_simple_metrics_success_mse_only(model_metrics_activity):
-    # Arrange
     target_data = DataFrame(
         {
             'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
@@ -781,23 +797,23 @@ def test_calculate_simple_metrics_success_mse_only(model_metrics_activity):
         'interval_minutes': 5,
     }
 
-    # Act
-    result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [{'metric': 'mse', 'value': 0.01}]
+    )
 
-    # Assert
-    assert len(result['metric']) == 1
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert len(result) == 1
     assert result['metric'].values[0] == 'mse'
     assert result['model_id'].values[0] == 'test_model_id'
     assert result['timestamp'].values[0] == '2023-05-26 11:12:28'
     assert result['data_size'].values[0] == 2
     assert result['interval_minutes'].values[0] == 5
-    model_metrics_activity.info.assert_called_once_with(
-        "Calculating simple metrics for model test_model_id: ['mse']", metadata['metadata']
-    )
+    mock_instance.calculate.assert_called_once_with(['mse'])
 
 
 def test_calculate_simple_metrics_success_mae_only(model_metrics_activity):
-    # Arrange
     target_data = DataFrame(
         {
             'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
@@ -814,23 +830,23 @@ def test_calculate_simple_metrics_success_mae_only(model_metrics_activity):
         'interval_minutes': 5,
     }
 
-    # Act
-    result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [{'metric': 'mae', 'value': 0.1}]
+    )
 
-    # Assert
-    assert len(result['metric']) == 1
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert len(result) == 1
     assert result['metric'].values[0] == 'mae'
     assert result['model_id'].values[0] == 'test_model_id'
     assert result['timestamp'].values[0] == '2023-05-26 11:12:28'
     assert result['data_size'].values[0] == 2
     assert result['interval_minutes'].values[0] == 5
-    model_metrics_activity.info.assert_called_once_with(
-        "Calculating simple metrics for model test_model_id: ['mae']", metadata['metadata']
-    )
+    mock_instance.calculate.assert_called_once_with(['mae'])
 
 
 def test_calculate_simple_metrics_success_r2_only(model_metrics_activity):
-    # Arrange
     target_data = DataFrame(
         {
             'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
@@ -847,24 +863,24 @@ def test_calculate_simple_metrics_success_r2_only(model_metrics_activity):
         'interval_minutes': 5,
     }
 
-    # Act
-    result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [{'metric': 'r2', 'value': 0.95}]
+    )
 
-    # Assert
-    assert len(result['metric']) == 1
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert len(result) == 1
     assert result['metric'].values[0] == 'r2'
     assert result['model_id'].values[0] == 'test_model_id'
     assert result['timestamp'].values[0] == '2023-05-26 11:12:28'
     assert result['data_size'].values[0] == 2
     assert result['interval_minutes'].values[0] == 5
-    model_metrics_activity.info.assert_called_once_with(
-        "Calculating simple metrics for model test_model_id: ['r2']", metadata['metadata']
-    )
+    mock_instance.calculate.assert_called_once_with(['r2'])
 
 
-def test_calculate_simple_metrics_r2_zero_ss_tot(model_metrics_activity):
-    # Arrange
-    # All target values are the same, so ss_tot will be 0
+def test_calculate_simple_metrics_r2_zero_variance_delegates_to_lib(model_metrics_activity):
+    """r2 zero-variance is now the lib's responsibility. Activity just passes through."""
     target_data = DataFrame(
         {
             'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
@@ -881,24 +897,18 @@ def test_calculate_simple_metrics_r2_zero_ss_tot(model_metrics_activity):
         'interval_minutes': 5,
     }
 
-    # Act
-    result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
-
-    # Assert
-    assert len(result['metric']) == 1
-    assert result['metric'].values[0] == 'r2'
-    assert result['value'].values[0] == 0.0  # Should return 0.0 when ss_tot == 0
-    assert result['model_id'].values[0] == 'test_model_id'
-    assert result['timestamp'].values[0] == '2023-05-26 11:12:28'
-    assert result['data_size'].values[0] == 2
-    assert result['interval_minutes'].values[0] == 5
-    model_metrics_activity.info.assert_called_once_with(
-        "Calculating simple metrics for model test_model_id: ['r2']", metadata['metadata']
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [{'metric': 'r2', 'value': 0.0}]
     )
 
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert result['value'].values[0] == 0.0
+    mock_instance.calculate.assert_called_once_with(['r2'])
+
 
 def test_calculate_simple_metrics_success_multiple_metrics_subset(model_metrics_activity):
-    # Arrange
     target_data = DataFrame(
         {
             'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28', '2023-05-26 11:12:29'],
@@ -915,23 +925,26 @@ def test_calculate_simple_metrics_success_multiple_metrics_subset(model_metrics_
         'interval_minutes': 5,
     }
 
-    # Act
-    result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
-
-    # Assert
-    assert len(result['metric']) == 2
-    assert 'rmse' in result['metric'].values
-    assert 'mae' in result['metric'].values
-    assert all(model_id == 'test_model_id' for model_id in result['model_id'].values)
-    assert all(timestamp == '2023-05-26 11:12:29' for timestamp in result['timestamp'].values)
-    assert all(data_size == 3 for data_size in result['data_size'].values)
-    assert all(interval_minutes == 5 for interval_minutes in result['interval_minutes'].values)
-    model_metrics_activity.info.assert_called_once_with(
-        "Calculating simple metrics for model test_model_id: ['rmse', 'mae']", metadata['metadata']
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [
+            {'metric': 'rmse', 'value': 0.1},
+            {'metric': 'mae', 'value': 0.1},
+        ]
     )
 
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
 
-def test_calculate_simple_metrics_unknown_metric_ignored(model_metrics_activity):
+    assert len(result) == 2
+    assert set(result['metric'].values) == {'rmse', 'mae'}
+    assert all(mid == 'test_model_id' for mid in result['model_id'].values)
+    assert all(ts == '2023-05-26 11:12:29' for ts in result['timestamp'].values)
+    assert all(ds == 3 for ds in result['data_size'].values)
+    assert all(im == 5 for im in result['interval_minutes'].values)
+    mock_instance.calculate.assert_called_once_with(['rmse', 'mae'])
+
+
+def test_calculate_simple_metrics_unknown_metric_raises(model_metrics_activity):
     target_data = DataFrame(
         {
             'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
@@ -948,7 +961,211 @@ def test_calculate_simple_metrics_unknown_metric_ignored(model_metrics_activity)
         'interval_minutes': 5,
     }
 
-    result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+    mock_instance = MagicMock()
+    mock_instance.calculate.side_effect = ValueError('Unknown metric: unknown_metric')
+    mock_class = MagicMock(return_value=mock_instance)
+    mock_class.is_r2_supported = MagicMock(return_value=True)
 
-    assert len(result['metric']) == 1
+    with patch('laborious.activities.model_metrics.RegressionMetrics', mock_class):
+        with raises(ValueError, match='Unknown metric'):
+            model_metrics_activity.calculate_simple_metrics(input_data)
+
+
+def test_calculate_simple_metrics_r2_skipped_for_nonlinear_model(model_metrics_activity):
+    """When model_type is non-linear, r2 is excluded and a warning notification fires."""
+    target_data = DataFrame(
+        {
+            'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
+            'target': [1.0, 2.0],
+            'prediction': [1.1, 2.1],
+        }
+    )
+
+    input_data = {
+        **metadata,
+        'model_id': 'test_model_id',
+        'target_data': target_data.to_dict(),
+        'metrics': ['rmse', 'r2'],
+        'interval_minutes': 5,
+        'model_type': 'XGBoost',
+    }
+
+    mock_instance = MagicMock()
+    mock_instance.calculate.return_value = [{'metric': 'rmse', 'value': 0.1}]
+    mock_class = MagicMock(return_value=mock_instance)
+    mock_class.is_r2_supported = MagicMock(return_value=False)
+
+    with patch('laborious.activities.model_metrics.RegressionMetrics', mock_class):
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert len(result) == 1
     assert result['metric'].values[0] == 'rmse'
+    mock_class.is_r2_supported.assert_called_once_with('XGBoost')
+    mock_instance.calculate.assert_called_once_with(['rmse'])
+    model_metrics_activity.warning.assert_called_once()
+    model_metrics_activity.send_notification.assert_called_once()
+    call_kwargs = model_metrics_activity.send_notification.call_args.kwargs
+    assert call_kwargs['notification_id'] == 'SIMPLE_METRICS_R2_UNSUPPORTED'
+    assert call_kwargs['level'] == NotificationLevel.WARNING
+
+
+def test_calculate_simple_metrics_no_model_type_includes_r2(model_metrics_activity):
+    """When model_type is None (legacy input), r2 is included without check."""
+    target_data = DataFrame(
+        {
+            'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
+            'target': [1.0, 2.0],
+            'prediction': [1.1, 2.1],
+        }
+    )
+
+    input_data = {
+        **metadata,
+        'model_id': 'test_model_id',
+        'target_data': target_data.to_dict(),
+        'metrics': ['r2'],
+        'interval_minutes': 5,
+        # no model_type key
+    }
+
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [{'metric': 'r2', 'value': 0.95}]
+    )
+
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert result['metric'].values[0] == 'r2'
+    mock_class.is_r2_supported.assert_not_called()
+
+
+def test_calculate_simple_metrics_threshold_breach_rmse(model_metrics_activity):
+    """When rmse exceeds rmse_max, a WARNING notification fires."""
+    target_data = DataFrame(
+        {
+            'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
+            'target': [1.0, 2.0],
+            'prediction': [1.1, 2.1],
+        }
+    )
+
+    input_data = {
+        **metadata,
+        'model_id': 'test_model_id',
+        'target_data': target_data.to_dict(),
+        'metrics': ['rmse'],
+        'interval_minutes': 5,
+        'thresholds': {'rmse_max': 0.05},
+    }
+
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [
+            {'metric': 'rmse', 'value': 0.1},
+        ]
+    )
+
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert len(result) == 1
+    model_metrics_activity.send_notification.assert_called_once()
+    call_kwargs = model_metrics_activity.send_notification.call_args.kwargs
+    assert call_kwargs['notification_id'] == 'SIMPLE_METRICS_THRESHOLD_BREACH'
+    assert call_kwargs['level'] == NotificationLevel.WARNING
+    assert 'rmse' in call_kwargs['message']
+
+
+def test_calculate_simple_metrics_threshold_no_breach(model_metrics_activity):
+    """When rmse is below rmse_max, no notification fires."""
+    target_data = DataFrame(
+        {
+            'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
+            'target': [1.0, 2.0],
+            'prediction': [1.1, 2.1],
+        }
+    )
+
+    input_data = {
+        **metadata,
+        'model_id': 'test_model_id',
+        'target_data': target_data.to_dict(),
+        'metrics': ['rmse'],
+        'interval_minutes': 5,
+        'thresholds': {'rmse_max': 1.0},
+    }
+
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [
+            {'metric': 'rmse', 'value': 0.1},
+        ]
+    )
+
+    with patcher:
+        model_metrics_activity.calculate_simple_metrics(input_data)
+
+    model_metrics_activity.send_notification.assert_not_called()
+
+
+def test_calculate_simple_metrics_threshold_r2_below_min(model_metrics_activity):
+    """When r2 drops below r2_min, a WARNING notification fires."""
+    target_data = DataFrame(
+        {
+            'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
+            'target': [1.0, 2.0],
+            'prediction': [1.1, 2.1],
+        }
+    )
+
+    input_data = {
+        **metadata,
+        'model_id': 'test_model_id',
+        'target_data': target_data.to_dict(),
+        'metrics': ['r2'],
+        'interval_minutes': 5,
+        'thresholds': {'r2_min': 0.95},
+    }
+
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [
+            {'metric': 'r2', 'value': 0.8},
+        ]
+    )
+
+    with patcher:
+        model_metrics_activity.calculate_simple_metrics(input_data)
+
+    model_metrics_activity.send_notification.assert_called_once()
+    call_kwargs = model_metrics_activity.send_notification.call_args.kwargs
+    assert 'r2' in call_kwargs['message']
+
+
+def test_calculate_simple_metrics_no_thresholds_no_alert(model_metrics_activity):
+    """When thresholds is None (not configured), no alerting, no crash."""
+    target_data = DataFrame(
+        {
+            'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28'],
+            'target': [1.0, 2.0],
+            'prediction': [1.1, 2.1],
+        }
+    )
+
+    input_data = {
+        **metadata,
+        'model_id': 'test_model_id',
+        'target_data': target_data.to_dict(),
+        'metrics': ['rmse'],
+        'interval_minutes': 5,
+        # no thresholds key
+    }
+
+    patcher, mock_class, mock_instance = _mock_regression_metrics_class(
+        [
+            {'metric': 'rmse', 'value': 999.0},
+        ]
+    )
+
+    with patcher:
+        result = DataFrame(model_metrics_activity.calculate_simple_metrics(input_data))
+
+    assert len(result) == 1
+    model_metrics_activity.send_notification.assert_not_called()

tests/laborious/workflows/test_simple_metrics.py

@@ -103,6 +103,8 @@ async def test_run(workflow_mock: AsyncMock, simple_metrics: SimpleMetrics):
             'target_data': target_data,
             'metrics': input_data['metrics'],
             'interval_minutes': input_data['interval_minutes'],
+            'model_type': None,
+            'thresholds': None,
         },
         retry_policy=ANY,
         start_to_close_timeout=ANY,
@@ -209,6 +211,8 @@ async def test_run_default_metrics(workflow_mock: AsyncMock, simple_metrics: Sim
             'target_data': target_data,
             'metrics': ['rmse', 'mse', 'mae', 'r2'],  # Default value
             'interval_minutes': input_data['interval_minutes'],
+            'model_type': None,
+            'thresholds': None,
         },
         retry_policy=ANY,
         start_to_close_timeout=ANY,

values.yaml

@@ -233,11 +233,15 @@ global:
 runtimes:
   basic:
     replicas: 1
-  single:
+  legacy:
     replicas: 1
     env:
       - name: "GITHUB_BRANCH"
         value: "main"
+      - name: "MLFLOW_HOST"
+        value: "http://sientia-tracker-mlflow-tracking.sientia-tracker.svc.cluster.local"
+      - name: "MLFLOW_PORT"
+        value: "80"
 
 # -----------------------------------------------------------------------------
 # Chart-level configuration (applies to all runtimes)
@@ -305,8 +309,8 @@ ssh:
 
 # kubectl create secret docker-registry docker-hub-secret --namespace sientia --docker-server=http://aignosi.azurecr.io --docker-username=aignosi --docker-password=<pwd>
 #
-# helm upgrade --install sientia-laborious-worker sientia/sientia-module -n sientia --create-namespace -f ./values.yaml --version 0.6.5
-#
+# helm upgrade --install sientia-laborious-worker /home/grezewave/Documents/projects/sientia/sientia-core-applications/sientia-module -n sientia --create-namespace -f ./values.yaml
+
 # kubectl create secret generic git-ssh-key-sientia-laborious-worker \
 #   --namespace sientia \
 #   --from-file=ssh-privatekey=git_key \