featrixsphere 0.2.4984__tar.gz → 0.2.4991__tar.gz

{featrixsphere-0.2.4984 → featrixsphere-0.2.4991}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: featrixsphere
-Version: 0.2.4984
+Version: 0.2.4991
 Summary: Transform any CSV into a production-ready ML model in minutes, not months.
 Home-page: https://github.com/Featrix/sphere
 Author: Featrix

featrixsphere-0.2.4991/VERSION

@@ -0,0 +1 @@
+0.2.4991

{featrixsphere-0.2.4984 → featrixsphere-0.2.4991}/featrixsphere/__init__.py

@@ -38,7 +38,7 @@ Example:
     ...                                labels=['Experiment A', 'Experiment B'])
 """
 
-__version__ = "0.2.4984"
+__version__ = "0.2.4991"
 __author__ = "Featrix"
 __email__ = "support@featrix.com"
 __license__ = "MIT"

{featrixsphere-0.2.4984 → featrixsphere-0.2.4991}/featrixsphere/client.py

@@ -3985,108 +3985,6 @@ class FeatrixSphereClient:
         
         return predictor_id
     
-    def _resolve_predictor_id(self, session_id: str, predictor_id: str = None, target_column: str = None, debug: bool = False) -> Dict[str, Any]:
-        """
-        Resolve predictor_id or target_column to predictor information.
-        
-        Args:
-            session_id: Session ID to check
-            predictor_id: Specific predictor ID to resolve
-            target_column: Target column name (fallback if predictor_id not provided)
-            debug: Whether to print debug information
-            
-        Returns:
-            Dictionary with predictor info including target_column, path, predictor_id
-            
-        Raises:
-            ValueError: If predictor not found or ambiguous
-        """
-        available_predictors = self._get_available_predictors(session_id, debug=debug)
-        
-        if not available_predictors:
-            # Don't fail here - let the server try to find/auto-discover the predictor
-            # The server's /predict endpoint has smart fallback logic to find checkpoint files
-            # even if the session file wasn't properly updated (e.g., training crashed)
-            if debug:
-                print(f"⚠️  No predictors found via models endpoint, letting server handle discovery")
-            return {
-                'target_column': target_column,
-                'predictor_id': predictor_id,
-                'path': None,
-                'type': None,
-                'server_discovery': True  # Flag that server should auto-discover
-            }
-        
-        # If predictor_id is provided, find it directly (since it's now the key)
-        if predictor_id:
-            if predictor_id in available_predictors:
-                predictor_info = available_predictors[predictor_id]
-                return {
-                    'target_column': predictor_info.get('target_column'),
-                    'predictor_id': predictor_id,
-                    'path': predictor_info.get('path'),
-                    'type': predictor_info.get('type')
-                }
-            
-            # Predictor ID not found
-            all_predictor_ids = list(available_predictors.keys())
-            
-            raise ValueError(
-                f"Predictor ID '{predictor_id}' not found for session {session_id}. "
-                f"Available predictor IDs: {all_predictor_ids}"
-            )
-        
-        # Fallback to target_column validation (search through values)
-        if target_column is None:
-            # Auto-detect: only valid if there's exactly one predictor
-            if len(available_predictors) == 1:
-                predictor_id = list(available_predictors.keys())[0]
-                predictor_info = available_predictors[predictor_id]
-                return {
-                    'target_column': predictor_info.get('target_column'),
-                    'predictor_id': predictor_id,
-                    'path': predictor_info.get('path'),
-                    'type': predictor_info.get('type')
-                }
-            else:
-                # Show unique target columns for clarity
-                target_columns = list(set(pred.get('target_column') for pred in available_predictors.values()))
-                raise ValueError(
-                    f"Multiple predictors found for session {session_id} with target columns: {target_columns}. "
-                    f"Please specify predictor_id parameter for precise selection."
-                )
-        else:
-            # Find predictors by target column (there might be multiple)
-            matching_predictors = {
-                pred_id: pred_info for pred_id, pred_info in available_predictors.items()
-                if pred_info.get('target_column') == target_column
-            }
-            
-            if not matching_predictors:
-                target_columns = list(set(pred.get('target_column') for pred in available_predictors.values()))
-                raise ValueError(
-                    f"No trained predictor found for target column '{target_column}' in session {session_id}. "
-                    f"Available target columns: {target_columns}"
-                )
-            
-            if len(matching_predictors) == 1:
-                # Only one predictor for this target column
-                predictor_id = list(matching_predictors.keys())[0]
-                predictor_info = matching_predictors[predictor_id]
-                return {
-                    'target_column': target_column,
-                    'predictor_id': predictor_id,
-                    'path': predictor_info.get('path'),
-                    'type': predictor_info.get('type')
-                }
-            else:
-                # Multiple predictors for the same target column
-                predictor_ids = list(matching_predictors.keys())
-                raise ValueError(
-                    f"Multiple predictors found for target column '{target_column}' in session {session_id}: {predictor_ids}. "
-                    f"Please specify predictor_id parameter for precise selection."
-                )
-    
     def list_predictors(self, session_id: str, verbose: bool = True, debug: bool = False) -> Dict[str, Dict[str, Any]]:
         """
         List all available predictors in a session and their target columns.

{featrixsphere-0.2.4984 → featrixsphere-0.2.4991}/featrixsphere.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: featrixsphere
-Version: 0.2.4984
+Version: 0.2.4991
 Summary: Transform any CSV into a production-ready ML model in minutes, not months.
 Home-page: https://github.com/Featrix/sphere
 Author: Featrix