ml-dash 0.6.13__py3-none-any.whl → 0.6.14__py3-none-any.whl

ml_dash/__init__.py

@@ -43,11 +43,11 @@ from .params import ParametersBuilder
 from .run import RUN
 from .storage import LocalStorage
 
-__version__ = "0.6.13"
+__version__ = "0.6.14"
 
 # Required version - MUST match exactly (blocks all older versions)
 # Update this with EVERY release to force users to upgrade
-REQUIRED_VERSION = "0.6.13"
+REQUIRED_VERSION = "0.6.14"
 
 
 def _check_version_compatibility():

ml_dash/buffer.py

@@ -458,12 +458,10 @@ class BackgroundBufferManager:
                     logs=batch,
                 )
             except Exception as e:
-                warnings.warn(
-                    f"Failed to flush {len(batch)} logs to remote server: {e}. "
-                    f"Training will continue.",
-                    RuntimeWarning,
-                    stacklevel=3,
-                )
+                raise RuntimeError(
+                    f"Failed to flush {len(batch)} logs to remote server: {e}\n"
+                    f"Data loss occurred. Check your network connection and server status."
+                ) from e
 
         if self._experiment.run._storage:
             # Local storage writes one at a time (no batch API)
@@ -479,11 +477,10 @@ class BackgroundBufferManager:
                         timestamp=log_entry["timestamp"],
                     )
                 except Exception as e:
-                    warnings.warn(
-                        f"Failed to write log to local storage: {e}",
-                        RuntimeWarning,
-                        stacklevel=3,
-                    )
+                    raise RuntimeError(
+                        f"Failed to write log to local storage: {e}\n"
+                        f"Check disk space and file permissions."
+                    ) from e
 
         self._last_log_flush = time.time()
 
@@ -535,12 +532,10 @@ class BackgroundBufferManager:
                 )
             except Exception as e:
                 metric_display = f"'{metric_name}'" if metric_name else "unnamed metric"
-                warnings.warn(
-                    f"Failed to flush {len(batch)} points to {metric_display} on remote server: {e}. "
-                    f"Training will continue.",
-                    RuntimeWarning,
-                    stacklevel=3,
-                )
+                raise RuntimeError(
+                    f"Failed to flush {len(batch)} points to {metric_display} on remote server: {e}\n"
+                    f"Data loss occurred. Check your network connection and server status."
+                ) from e
 
         if self._experiment.run._storage:
             try:
@@ -556,11 +551,10 @@ class BackgroundBufferManager:
                 )
             except Exception as e:
                 metric_display = f"'{metric_name}'" if metric_name else "unnamed metric"
-                warnings.warn(
-                    f"Failed to flush {len(batch)} points to {metric_display} in local storage: {e}",
-                    RuntimeWarning,
-                    stacklevel=3,
-                )
+                raise RuntimeError(
+                    f"Failed to flush {len(batch)} points to {metric_display} in local storage: {e}\n"
+                    f"Check disk space and file permissions."
+                ) from e
 
         self._last_metric_flush[metric_name] = time.time()
 
@@ -597,12 +591,10 @@ class BackgroundBufferManager:
                     entries=batch,
                 )
             except Exception as e:
-                warnings.warn(
-                    f"Failed to flush {len(batch)} entries to track '{topic}' on remote server: {e}. "
-                    f"Training will continue.",
-                    RuntimeWarning,
-                    stacklevel=3,
-                )
+                raise RuntimeError(
+                    f"Failed to flush {len(batch)} entries to track '{topic}' on remote server: {e}\n"
+                    f"Data loss occurred. Check your network connection and server status."
+                ) from e
 
         # Write to local storage
         if self._experiment.run._storage:
@@ -615,11 +607,10 @@ class BackgroundBufferManager:
                     entries=batch,
                 )
             except Exception as e:
-                warnings.warn(
-                    f"Failed to flush {len(batch)} entries to track '{topic}' in local storage: {e}",
-                    RuntimeWarning,
-                    stacklevel=3,
-                )
+                raise RuntimeError(
+                    f"Failed to flush {len(batch)} entries to track '{topic}' in local storage: {e}\n"
+                    f"Check disk space and file permissions."
+                ) from e
 
         self._last_track_flush[topic] = time.time()
 
@@ -663,12 +654,10 @@ class BackgroundBufferManager:
                     if total_files > 1:
                         print(f"[ML-Dash]   [{completed}/{total_files}] Uploaded {file_entry['filename']}", flush=True)
                 except Exception as e:
-                    completed += 1
-                    warnings.warn(
-                        f"Failed to upload file {file_entry['filename']}: {e}",
-                        RuntimeWarning,
-                        stacklevel=3,
-                    )
+                    raise RuntimeError(
+                        f"Failed to upload file {file_entry['filename']}: {e}\n"
+                        f"File upload failed. Check network connection and file permissions."
+                    ) from e
 
     def _upload_single_file(self, file_entry: Dict[str, Any]) -> None:
         """

ml_dash/cli_commands/upload.py

@@ -306,7 +306,8 @@ def discover_experiments(
       with open(exp_json, "r") as f:
         metadata = json.load(f)
         prefix = metadata.get("prefix")
-    except:
+    except (FileNotFoundError, json.JSONDecodeError, KeyError) as e:
+      # Metadata file missing or invalid - will use path-based prefix
       pass
 
     # Extract project and experiment names from PREFIX (not path)

ml_dash/client.py

@@ -226,8 +226,13 @@ class RemoteClient:
             result = self.graphql_query(query)
             username = result.get("me", {}).get("username")
             return username
-        except Exception:
-            return None
+        except Exception as e:
+            # Re-raise authentication errors
+            from .auth.exceptions import AuthenticationError
+            if isinstance(e, AuthenticationError):
+                raise
+            # For other errors, raise a clear exception
+            raise RuntimeError(f"Failed to fetch namespace from server: {e}") from e
 
     def get_current_user(self) -> Optional[Dict[str, Any]]:
         """
@@ -264,8 +269,13 @@ class RemoteClient:
             """
             result = self.graphql_query(query)
             return result.get("me")
-        except Exception:
-            return None
+        except Exception as e:
+            # Re-raise authentication errors
+            from .auth.exceptions import AuthenticationError
+            if isinstance(e, AuthenticationError):
+                raise
+            # For other errors, raise a clear exception
+            raise RuntimeError(f"Failed to fetch current user from server: {e}") from e
 
     def _ensure_authenticated(self):
         """Check if authenticated, raise error if not."""

ml_dash/experiment.py

@@ -399,8 +399,11 @@ class Experiment:
           print(f"View results at: {experiment_url}")
 
       except Exception as e:
-        # Log error but don't fail the close operation
-        print(f"Warning: Failed to update experiment status: {e}")
+        # Raise on status update failure
+        raise RuntimeError(
+          f"Failed to update experiment status to COMPLETED: {e}\n"
+          f"Experiment may not be marked as completed on the server."
+        ) from e
 
     self._is_open = False
 
@@ -554,15 +557,10 @@ class Experiment:
             logs=[log_entry],  # Single log in array
           )
         except Exception as e:
-          # Log warning but don't crash training
-          import warnings
-
-          warnings.warn(
-            f"Failed to write log to remote server: {e}. Training will continue.",
-            RuntimeWarning,
-            stacklevel=4,
-          )
-          # Fall through to local storage if available
+          raise RuntimeError(
+            f"Failed to write log to remote server: {e}\n"
+            f"Data loss occurred. Check your network connection and server status."
+          ) from e
 
       if self.run._storage:
         # Local mode: write to file immediately
@@ -577,11 +575,10 @@ class Experiment:
             timestamp=log_entry["timestamp"],
           )
         except Exception as e:
-          import warnings
-
-          warnings.warn(
-            f"Failed to write log to local storage: {e}", RuntimeWarning, stacklevel=4
-          )
+          raise RuntimeError(
+            f"Failed to write log to local storage: {e}\n"
+            f"Check disk space and file permissions."
+          ) from e
 
   def _print_log(
     self, message: str, level: str, metadata: Optional[Dict[str, Any]]
@@ -1072,17 +1069,11 @@ class Experiment:
             metadata=metadata,
           )
         except Exception as e:
-          # Log warning but don't crash training
-          import warnings
-
           metric_display = f"'{name}'" if name else "unnamed metric"
-          warnings.warn(
-            f"Failed to log {metric_display} to remote server: {e}. "
-            f"Training will continue.",
-            RuntimeWarning,
-            stacklevel=3,
-          )
-          # Fall through to local storage if available
+          raise RuntimeError(
+            f"Failed to log {metric_display} to remote server: {e}\n"
+            f"Data loss occurred. Check your network connection and server status."
+          ) from e
 
       if self.run._storage:
         # Local mode: append to local storage
@@ -1098,14 +1089,11 @@ class Experiment:
             metadata=metadata,
           )
         except Exception as e:
-          import warnings
-
           metric_display = f"'{name}'" if name else "unnamed metric"
-          warnings.warn(
-            f"Failed to log {metric_display} to local storage: {e}",
-            RuntimeWarning,
-            stacklevel=3,
-          )
+          raise RuntimeError(
+            f"Failed to log {metric_display} to local storage: {e}\n"
+            f"Check disk space and file permissions."
+          ) from e
 
       return result
 
@@ -1141,15 +1129,10 @@ class Experiment:
             entries=[{"timestamp": timestamp, **data}],
           )
         except Exception as e:
-          # Log warning but don't crash training
-          import warnings
-
-          warnings.warn(
-            f"Failed to log track '{topic}' to remote server: {e}. "
-            f"Training will continue.",
-            RuntimeWarning,
-            stacklevel=3,
-          )
+          raise RuntimeError(
+            f"Failed to log track '{topic}' to remote server: {e}\n"
+            f"Data loss occurred. Check your network connection and server status."
+          ) from e
 
       if self.run._storage:
         # Local mode: append to local storage
@@ -1162,13 +1145,10 @@ class Experiment:
             entries=[{"timestamp": timestamp, **data}],
           )
         except Exception as e:
-          import warnings
-
-          warnings.warn(
-            f"Failed to log track '{topic}' to local storage: {e}",
-            RuntimeWarning,
-            stacklevel=3,
-          )
+          raise RuntimeError(
+            f"Failed to log track '{topic}' to local storage: {e}\n"
+            f"Check disk space and file permissions."
+          ) from e
 
   def _append_batch_to_metric(
     self,
@@ -1205,17 +1185,11 @@ class Experiment:
           metadata=metadata,
         )
       except Exception as e:
-        # Log warning but don't crash training
-        import warnings
-
         metric_display = f"'{name}'" if name else "unnamed metric"
-        warnings.warn(
-          f"Failed to log batch to {metric_display} on remote server: {e}. "
-          f"Training will continue.",
-          RuntimeWarning,
-          stacklevel=3,
-        )
-        # Fall through to local storage if available
+        raise RuntimeError(
+          f"Failed to log batch to {metric_display} on remote server: {e}\n"
+          f"Data loss occurred. Check your network connection and server status."
+        ) from e
 
     if self.run._storage:
       # Local mode: append batch to local storage
@@ -1231,14 +1205,11 @@ class Experiment:
           metadata=metadata,
         )
       except Exception as e:
-        import warnings
-
         metric_display = f"'{name}'" if name else "unnamed metric"
-        warnings.warn(
-          f"Failed to log batch to {metric_display} in local storage: {e}",
-          RuntimeWarning,
-          stacklevel=3,
-        )
+        raise RuntimeError(
+          f"Failed to log batch to {metric_display} in local storage: {e}\n"
+          f"Check disk space and file permissions."
+        ) from e
 
     return result
 

ml_dash/metric.py

@@ -69,9 +69,11 @@ class BufferManager:
                 value = float('nan')
             try:
                 self._buffers[prefix][key].append(float(value))
-            except (TypeError, ValueError):
-                # Skip non-numeric values silently
-                continue
+            except (TypeError, ValueError) as e:
+                raise ValueError(
+                    f"Cannot buffer non-numeric value for '{key}': {value!r} (type: {type(value).__name__})\n"
+                    f"Metrics must be numeric (int, float). Use exp.log() for non-numeric values."
+                ) from e
 
     def _compute_stats(self, values: List[float], aggs: tuple) -> Dict[str, float]:
         """
@@ -248,9 +250,11 @@ class SummaryCache:
                 value = float('nan')
             try:
                 self._buffer[key].append(float(value))
-            except (TypeError, ValueError):
-                # Skip non-numeric values silently
-                continue
+            except (TypeError, ValueError) as e:
+                raise ValueError(
+                    f"Cannot store non-numeric value for '{key}': {value!r} (type: {type(value).__name__})\n"
+                    f"SummaryCache only accepts numeric values. Use exp.log() for non-numeric data."
+                ) from e
 
     def set(self, **kwargs) -> None:
         """

{ml_dash-0.6.13.dist-info → ml_dash-0.6.14.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: ml-dash
-Version: 0.6.13
+Version: 0.6.14
 Summary: ML experiment tracking and data storage
 Keywords: machine-learning,experiment-tracking,mlops,data-storage
 Author: Ge Yang, Tom Tao

{ml_dash-0.6.13.dist-info → ml_dash-0.6.14.dist-info}/RECORD

@@ -1,4 +1,4 @@
-ml_dash/__init__.py,sha256=S-Q1EK0mKvV0WAOB-QjQF7NjLvYntB3_LIAbzKA45LA,3060
+ml_dash/__init__.py,sha256=0N78Rs6EH_pFB-OoRXx42ZXm1ghbvp2xd54it41CPXM,3060
 ml_dash/auth/__init__.py,sha256=3lwM-Y8UBHPU1gFW2JNpmXlPVTnkGudWLKNFFKulQfo,1200
 ml_dash/auth/constants.py,sha256=ku4QzQUMNjvyJwjy7AUdywMAZd59jXSxNHZxDiagUWU,280
 ml_dash/auth/device_flow.py,sha256=DQOdPNlZCuU1umZOA_A6WXdRM3zWphnyo9IntToBl_A,7921
@@ -6,7 +6,7 @@ ml_dash/auth/device_secret.py,sha256=qUsz6M9S1GEIukvmz57eJEp57srSx74O4MU9mZEeDlE
 ml_dash/auth/exceptions.py,sha256=IeBwUzoaTyFtPwd4quFOIel49inIzuabe_ChEeEXEWI,725
 ml_dash/auth/token_storage.py,sha256=9YQXGrn41UVyc1wUvZYbTYLzxSt5NGOyNFNjeX28bjA,7976
 ml_dash/auto_start.py,sha256=mYNjLGI2kyylIfOX5wGOR74gb9UlXg1n5OUQu7aw5SE,2412
-ml_dash/buffer.py,sha256=R3QaD55s1an3tJvzyXfrEuBxeJsN4o4QNdE__BoIr9w,26437
+ml_dash/buffer.py,sha256=vZwBRnnuiuvLHuWISTi2QYWadrJ6maVVa6oZSshByeM,26400
 ml_dash/cli.py,sha256=Vd0taM5MQrhvxqL2KQhklZ00wxZLdWl6Qw1IPkjPNuw,2897
 ml_dash/cli_commands/__init__.py,sha256=bjAmV7MsW-bhtW_4SnLJ0Cfkt9h82vMDC8ebW1Ke8KE,38
 ml_dash/cli_commands/api.py,sha256=NekZEJGWNpIfB6YrsrOw7kw7rZKjVudwgJWPZIy6ANQ,4535
@@ -17,13 +17,13 @@ ml_dash/cli_commands/login.py,sha256=zX-urtUrfzg2qOGtKNYQgj6UloN9kzj4zEO6h_xwuNs
 ml_dash/cli_commands/logout.py,sha256=lTUUNyRXqvo61qNkCd4KBrPUujDAHnNqsHkU6bHie0U,1332
 ml_dash/cli_commands/profile.py,sha256=PoRO1XA4bnOINptj4AO0SyNDBADeryPJBfgC74327e4,5997
 ml_dash/cli_commands/remove.py,sha256=AtDlUWkNeGcnZWN0Wbg6XoyYhFHkCFMPdxsGA33v38c,5325
-ml_dash/cli_commands/upload.py,sha256=o5D1be3V_wrzoVI3QHrKzyw2fG6ocD33MSZjzRsS60w,49659
-ml_dash/client.py,sha256=sQokhvofq56JxpmUBVUsfjjn5qKqyUXM9F3uxsUbUbA,66608
+ml_dash/cli_commands/upload.py,sha256=oZVU8m9Ey8N151KmUiQD_qQLEBBN4oqu5NcqUNOUJWY,49786
+ml_dash/client.py,sha256=ZWO1uzV-GLz-3R2ML90XPCV200e_ch7L-4iG07jkC84,67213
 ml_dash/config.py,sha256=oz2xvoBh2X_xUXWr92cPD5nFxXMT5LxVNypv5B5O0fA,3116
-ml_dash/experiment.py,sha256=JqNYsoAOsvqIRMVl6gHFodlZ79b0X2rUFGz6NRixKic,43726
+ml_dash/experiment.py,sha256=eSzrh49jP5dPx6rlEZZHLDMn2CxoA9uE_PSG3CpQQfA,43366
 ml_dash/files.py,sha256=tGJCTxPfd9vmfvIEqstZjzLvqmHzMZffPXHz0jU9bYU,54441
 ml_dash/log.py,sha256=E-DLg0vejVLLEyShJ_r0LneDMI0XU7XTH5iKWYJe9jI,5298
-ml_dash/metric.py,sha256=ghD1jnuv6dbjV1Jlo7q0mx9UEzpdto2Y1-oDWrSfg04,25809
+ml_dash/metric.py,sha256=hwdZIkHbt_wbJQPiO1LK3UJwusbAC7J-TR0gKeAnjHc,26206
 ml_dash/params.py,sha256=pPFvknJAJX5uhckzjO1r-HNnKbQFFKDISFmOXNET5eY,9046
 ml_dash/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 ml_dash/remote_auto_start.py,sha256=5fvQDHv1CWEKFb6WAa5_uyEInwV_SvotXjOO_6i6ZKE,1687
@@ -31,7 +31,7 @@ ml_dash/run.py,sha256=yAKZ9HtU4cidtbWMAY1IiDPVWwluVlicD5hsmVT89U0,11361
 ml_dash/snowflake.py,sha256=14rEpRU5YltsmmmZW0EMUy_hdv5S5ME9gWVtmdmwfiU,4917
 ml_dash/storage.py,sha256=x1W-dK6wQY36-LVOJ4kA8Dn07ObNQuIErQWJ3b0PoGY,44910
 ml_dash/track.py,sha256=Dfg1ZnmKZ_FlE5ZfG8Qld_wN4RIMs3nrOxrxwf3thiY,8164
-ml_dash-0.6.13.dist-info/WHEEL,sha256=z-mOpxbJHqy3cq6SvUThBZdaLGFZzdZPtgWLcP2NKjQ,79
-ml_dash-0.6.13.dist-info/entry_points.txt,sha256=dYs2EHX1uRNO7AQGNnVaJJpgiy0Z9q7tiy4fHSyaf3Q,46
-ml_dash-0.6.13.dist-info/METADATA,sha256=bKRP7yW8ULvNc5kqdeO0USgqCOWhXuV3EuhViNKOZIU,9536
-ml_dash-0.6.13.dist-info/RECORD,,
+ml_dash-0.6.14.dist-info/WHEEL,sha256=z-mOpxbJHqy3cq6SvUThBZdaLGFZzdZPtgWLcP2NKjQ,79
+ml_dash-0.6.14.dist-info/entry_points.txt,sha256=dYs2EHX1uRNO7AQGNnVaJJpgiy0Z9q7tiy4fHSyaf3Q,46
+ml_dash-0.6.14.dist-info/METADATA,sha256=flWNnOAY6fm6xBPb4aqoaHebFkHjaEcWRBaU51ckUjQ,9536
+ml_dash-0.6.14.dist-info/RECORD,,