gitlabds 2.1.0__tar.gz → 2.1.2__tar.gz

{gitlabds-2.1.0 → gitlabds-2.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gitlabds
-Version: 2.1.0
+Version: 2.1.2
 Summary: Gitlab Data Science and Modeling Tools
 Home-page: https://gitlab.com/gitlab-data/gitlabds
 Author: Kevin Dietz
@@ -764,7 +764,7 @@ results = gitlabds.prescriptions(
 #### Description
 Generate baseline feature distributions, importance scores, and drift thresholds in a single comprehensive artifact for model monitoring.
 
-`gitlabds.generate_baseline_features(training_data, feature_importance_df, importance_method="shapley_values", n_bins=10, psi_warning=0.1, psi_critical=0.2, ks_warning=0.1, ks_critical=0.2, js_warning=0.05, js_critical=0.1, output_path="baseline_features.json"):`
+`gitlabds.generate_baseline_features(training_data, feature_importance_df, importance_method="shapley_values", n_bins=10, psi_warning=0.1, psi_critical=0.2, ks_warning=0.2, ks_critical=0.3, js_warning=0.1, js_critical=0.2, output_path="baseline_features.json"):`
 
 #### Parameters:
 - **_training_data_** : Training feature data DataFrame
@@ -818,8 +818,10 @@ Generate baseline calibration data with train + test curves, prediction statisti
 - **_test_actuals_** : Test set actual binary outcomes (0/1)
 - **_model_configuration_** : Dictionary of model configuration parameters
 - **_n_bins_** : Number of bins for calibration curve (default: 10)
+- **_prediction_drift_warning_** :  Warning level for prediction score drift (default: 0.10), 
+- **_prediction_drift_critical_** : Critical level for prediction score drift (default: 0.20),  
 - **_output_path_** : Path to save the JSON file
-
+        
 #### Returns
 - None (saves baseline calibration to JSON file)
 
@@ -835,6 +837,8 @@ gitlabds.generate_baseline_calibration(
     test_predictions=y_test_pred,
     test_actuals=y_test,
     model_configuration=model_config,
+    prediction_drift_warning=0.15,
+    prediction_drift_critical=0.25
     output_path="model_baseline_calibration.json"
 )
 ```

{gitlabds-2.1.0 → gitlabds-2.1.2}/README.md

@@ -744,7 +744,7 @@ results = gitlabds.prescriptions(
 #### Description
 Generate baseline feature distributions, importance scores, and drift thresholds in a single comprehensive artifact for model monitoring.
 
-`gitlabds.generate_baseline_features(training_data, feature_importance_df, importance_method="shapley_values", n_bins=10, psi_warning=0.1, psi_critical=0.2, ks_warning=0.1, ks_critical=0.2, js_warning=0.05, js_critical=0.1, output_path="baseline_features.json"):`
+`gitlabds.generate_baseline_features(training_data, feature_importance_df, importance_method="shapley_values", n_bins=10, psi_warning=0.1, psi_critical=0.2, ks_warning=0.2, ks_critical=0.3, js_warning=0.1, js_critical=0.2, output_path="baseline_features.json"):`
 
 #### Parameters:
 - **_training_data_** : Training feature data DataFrame
@@ -798,8 +798,10 @@ Generate baseline calibration data with train + test curves, prediction statisti
 - **_test_actuals_** : Test set actual binary outcomes (0/1)
 - **_model_configuration_** : Dictionary of model configuration parameters
 - **_n_bins_** : Number of bins for calibration curve (default: 10)
+- **_prediction_drift_warning_** :  Warning level for prediction score drift (default: 0.10), 
+- **_prediction_drift_critical_** : Critical level for prediction score drift (default: 0.20),  
 - **_output_path_** : Path to save the JSON file
-
+        
 #### Returns
 - None (saves baseline calibration to JSON file)
 
@@ -815,6 +817,8 @@ gitlabds.generate_baseline_calibration(
     test_predictions=y_test_pred,
     test_actuals=y_test,
     model_configuration=model_config,
+    prediction_drift_warning=0.15,
+    prediction_drift_critical=0.25
     output_path="model_baseline_calibration.json"
 )
 ```

{gitlabds-2.1.0 → gitlabds-2.1.2}/gitlabds/baselines.py

@@ -11,10 +11,10 @@ def generate_baseline_features(
     # Drift thresholds (integrated)
     psi_warning: float = 0.1,
     psi_critical: float = 0.2,
-    ks_warning: float = 0.1,
-    ks_critical: float = 0.2,
-    js_warning: float = 0.05,
-    js_critical: float = 0.1,
+    ks_warning: float = 0.2,
+    ks_critical: float = 0.3,
+    js_warning: float = 0.1,
+    js_critical: float = 0.2,
     output_path: str = "baseline_features.json"
 ) -> None:
     """
@@ -218,6 +218,8 @@ def generate_baseline_calibration(
     test_actuals: np.ndarray,
     model_configuration: Dict[str, Any],
     n_bins: int = 10,
+    prediction_drift_warning: float = 0.10, 
+    prediction_drift_critical: float = 0.20,  
     output_path: str = "baseline_calibration.json"
 ) -> None:
     """
@@ -238,6 +240,10 @@ def generate_baseline_calibration(
         Dictionary of model configuration parameters (e.g., {'f1_threshold': 0.15})
     n_bins : int
         Number of bins for calibration curve (default: 10)
+    prediction_drift_warning: float (default: 0.10)
+        Warning level for prediction score drift
+    prediction_drift_critical: float (default: 0.20)
+        Critical level for prediction score drift
     output_path : str
         Path to save the JSON file
         
@@ -323,6 +329,13 @@ def generate_baseline_calibration(
         },
         "prediction_statistics": prediction_statistics,
         "model_configuration": model_configuration,
+        "drift_thresholds": {
+            "prediction_drift": {
+                "warning": prediction_drift_warning,
+                "critical": prediction_drift_critical,
+                "description": "Mean prediction shift percentage from baseline"
+            }
+        },
         "metadata": {
             "n_bins": n_bins,
             "train_sample_size": len(train_predictions),

{gitlabds-2.1.0 → gitlabds-2.1.2}/gitlabds/monitoring_metrics.py

@@ -55,6 +55,7 @@ def calculate_monitoring_metrics(
     baseline_features = baseline_features_data.get('baseline_features', {})
     feature_importance_data = baseline_features_data.get('feature_importance', {})
     drift_thresholds = baseline_features_data.get('drift_thresholds', {})
+    prediction_drift_thresholds = baseline_calibration_data.get('drift_thresholds', {}).get('prediction_drift', {})
     
     baseline_calibration = baseline_calibration_data.get('calibration_curves', {})
     model_configuration = baseline_calibration_data.get('model_configuration', {})
@@ -128,8 +129,8 @@ def calculate_monitoring_metrics(
             'model_name': model_name,
             'sub_model': sub_model,
             'model_version': model_version,
-            'feature_name': feature_name,
             'scoring_date': scoring_date.strftime('%Y-%m-%d %H:%M:%S'),
+            'feature_name': feature_name,
             'psi_value': psi_value,
             'feature_importance': feat_importance,
             'status': status
@@ -258,18 +259,22 @@ def calculate_monitoring_metrics(
             if baseline_std is not None:
                 prediction_drift['std_shift'] = float(current_std - baseline_std)
                 prediction_drift['std_shift_pct'] = float((current_std - baseline_std) / baseline_std * 100) if baseline_std > 0 else None
-            
-            # Determine prediction drift status
+
+            # Determine prediction drift status using configurable thresholds
             mean_shift_pct = abs(prediction_drift.get('mean_shift_pct', 0))
-            if mean_shift_pct >= 20:
-                prediction_drift['status'] = 'critical'
-            elif mean_shift_pct >= 10:
-                prediction_drift['status'] = 'warning'
+            critical_threshold = prediction_drift_thresholds.get('critical', 0.25)
+            warning_threshold = prediction_drift_thresholds.get('warning', 0.15)
+            
+            if mean_shift_pct >= critical_threshold:
+                prediction_drift_status = 'critical'
+            elif mean_shift_pct >= warning_threshold:
+                prediction_drift_status = 'warning'
             else:
-                prediction_drift['status'] = 'healthy'
-    
-    # Extract prediction drift status for separate column
-    prediction_drift_status = prediction_drift.get('status', 'unknown')
+                prediction_drift_status = 'healthy'
+
+        else:
+            prediction_drift_status = 'unknown'
+
     
     # =================================================================
     # 4. CREATE OUTPUT DATAFRAMES

{gitlabds-2.1.0 → gitlabds-2.1.2}/gitlabds.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gitlabds
-Version: 2.1.0
+Version: 2.1.2
 Summary: Gitlab Data Science and Modeling Tools
 Home-page: https://gitlab.com/gitlab-data/gitlabds
 Author: Kevin Dietz
@@ -764,7 +764,7 @@ results = gitlabds.prescriptions(
 #### Description
 Generate baseline feature distributions, importance scores, and drift thresholds in a single comprehensive artifact for model monitoring.
 
-`gitlabds.generate_baseline_features(training_data, feature_importance_df, importance_method="shapley_values", n_bins=10, psi_warning=0.1, psi_critical=0.2, ks_warning=0.1, ks_critical=0.2, js_warning=0.05, js_critical=0.1, output_path="baseline_features.json"):`
+`gitlabds.generate_baseline_features(training_data, feature_importance_df, importance_method="shapley_values", n_bins=10, psi_warning=0.1, psi_critical=0.2, ks_warning=0.2, ks_critical=0.3, js_warning=0.1, js_critical=0.2, output_path="baseline_features.json"):`
 
 #### Parameters:
 - **_training_data_** : Training feature data DataFrame
@@ -818,8 +818,10 @@ Generate baseline calibration data with train + test curves, prediction statisti
 - **_test_actuals_** : Test set actual binary outcomes (0/1)
 - **_model_configuration_** : Dictionary of model configuration parameters
 - **_n_bins_** : Number of bins for calibration curve (default: 10)
+- **_prediction_drift_warning_** :  Warning level for prediction score drift (default: 0.10), 
+- **_prediction_drift_critical_** : Critical level for prediction score drift (default: 0.20),  
 - **_output_path_** : Path to save the JSON file
-
+        
 #### Returns
 - None (saves baseline calibration to JSON file)
 
@@ -835,6 +837,8 @@ gitlabds.generate_baseline_calibration(
     test_predictions=y_test_pred,
     test_actuals=y_test,
     model_configuration=model_config,
+    prediction_drift_warning=0.15,
+    prediction_drift_critical=0.25
     output_path="model_baseline_calibration.json"
 )
 ```