nshtrainer 0.37.0__py3-none-any.whl → 0.39.0__py3-none-any.whl

nshtrainer/callbacks/checkpoint/_base.py

@@ -41,7 +41,7 @@ class BaseCheckpointCallbackConfig(CallbackConfigBase, ABC):
         self,
         root_config: "BaseConfig",
         dirpath: Path,
-    ) -> "CheckpointBase": ...
+    ) -> "CheckpointBase | None": ...
 
     @override
     def create_callbacks(self, root_config):
@@ -50,7 +50,8 @@ class BaseCheckpointCallbackConfig(CallbackConfigBase, ABC):
             or root_config.directory.resolve_subdirectory(root_config.id, "checkpoint")
         )
 
-        yield self.create_checkpoint(root_config, dirpath)
+        if (callback := self.create_checkpoint(root_config, dirpath)) is not None:
+            yield callback
 
 
 TConfig = TypeVar("TConfig", bound=BaseCheckpointCallbackConfig, infer_variance=True)

nshtrainer/callbacks/checkpoint/best_checkpoint.py

@@ -20,15 +20,26 @@ class BestCheckpointCallbackConfig(BaseCheckpointCallbackConfig):
     metric: MetricConfig | None = None
     """Metric to monitor, or `None` to use the default metric."""
 
+    throw_on_no_metric: bool = True
+    """
+    Whether to throw an error if no metric is provided and no primary metric is found in the root config.
+    """
+
     @override
     def create_checkpoint(self, root_config, dirpath):
         # Resolve metric
         if (metric := self.metric) is None and (
             metric := root_config.primary_metric
         ) is None:
-            raise ValueError(
-                "No metric provided and no primary metric found in the root config"
+            error_msg = (
+                "No metric provided and no primary metric found in the root config. "
+                "Cannot create BestCheckpointCallback."
             )
+            if self.throw_on_no_metric:
+                raise ValueError(error_msg)
+            else:
+                log.warning(error_msg)
+                return None
 
         return BestCheckpoint(self, dirpath, metric)
 

nshtrainer/callbacks/early_stopping.py

@@ -51,7 +51,7 @@ class EarlyStoppingConfig(CallbackConfigBase):
             metric := root_config.primary_metric
         ) is None:
             raise ValueError(
-                "Either `metric` or `root_config.primary_metric` must be set."
+                "Either `metric` or `root_config.primary_metric` must be set to use EarlyStopping."
             )
 
         yield EarlyStopping(self, metric)

nshtrainer/config.py

@@ -65,13 +65,13 @@ from nshtrainer.callbacks.wandb_upload_code import (
     WandbUploadCodeConfig as WandbUploadCodeConfig,
 )
 from nshtrainer.callbacks.wandb_watch import WandbWatchConfig as WandbWatchConfig
-from nshtrainer.config import LRSchedulerConfig as LRSchedulerConfig
 from nshtrainer.loggers._base import BaseLoggerConfig as BaseLoggerConfig
 from nshtrainer.loggers.csv import CSVLoggerConfig as CSVLoggerConfig
 from nshtrainer.loggers.tensorboard import (
     TensorboardLoggerConfig as TensorboardLoggerConfig,
 )
 from nshtrainer.loggers.wandb import WandbLoggerConfig as WandbLoggerConfig
+from nshtrainer.lr_scheduler import LRSchedulerConfig as LRSchedulerConfig
 from nshtrainer.lr_scheduler._base import LRSchedulerConfigBase as LRSchedulerConfigBase
 from nshtrainer.lr_scheduler.linear_warmup_cosine import (
     DurationConfig as DurationConfig,

nshtrainer/trainer/_config.py

@@ -263,7 +263,7 @@ class CheckpointSavingConfig(CallbackConfigBase):
     """Enable checkpoint saving."""
 
     checkpoint_callbacks: Sequence[CheckpointCallbackConfig] = [
-        BestCheckpointCallbackConfig(),
+        BestCheckpointCallbackConfig(throw_on_no_metric=False),
         LastCheckpointCallbackConfig(),
         OnExceptionCheckpointCallbackConfig(),
     ]

{nshtrainer-0.37.0.dist-info → nshtrainer-0.39.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: nshtrainer
-Version: 0.37.0
+Version: 0.39.0
 Summary: 
 Author: Nima Shoghi
 Author-email: nimashoghi@gmail.com

{nshtrainer-0.37.0.dist-info → nshtrainer-0.39.0.dist-info}/RECORD

@@ -11,13 +11,13 @@ nshtrainer/callbacks/_throughput_monitor_callback.py,sha256=aJo_11rc4lo0IYOd-kHm
 nshtrainer/callbacks/actsave.py,sha256=qbnaKts4_dvjPeAaPtv7Ds12_vEWzaHUfg_--49NB9I,4041
 nshtrainer/callbacks/base.py,sha256=NpjeKmonJ1Kaz5_39XSn3LlDwvbGjk6WV8BpHSNCvI4,3508
 nshtrainer/callbacks/checkpoint/__init__.py,sha256=g-3zIthupERKqWZQw-A_busQPaPRkto6iHBV-M7nK1Y,527
-nshtrainer/callbacks/checkpoint/_base.py,sha256=vvlwuD-20NozYVIolGGShmUdkkNYeuwN6xCoFnK4GiU,6157
-nshtrainer/callbacks/checkpoint/best_checkpoint.py,sha256=8BHgLAd3Tuzf5sup0guEAKF1jJiAwYsjdKBFYZw98ac,2171
+nshtrainer/callbacks/checkpoint/_base.py,sha256=X6844ceZ-EMz9UCVLJnA1d-ej2AkkRH90v2hy2zPDkc,6215
+nshtrainer/callbacks/checkpoint/best_checkpoint.py,sha256=MOC5jLlQnnc61yuiVc2-O4NeYclnkXroOPDV9-2zh5w,2553
 nshtrainer/callbacks/checkpoint/last_checkpoint.py,sha256=CWWv0cSwQ1VAX26N7hAyMxbNCk26Keh39oQguBEK5To,1102
 nshtrainer/callbacks/checkpoint/on_exception_checkpoint.py,sha256=ctT88EGT22_t_6tr5r7Sfo43cuve6XeroBnBYRMPOus,3372
 nshtrainer/callbacks/debug_flag.py,sha256=T7rkY9hYQ_-PsPo2XiQ4eVZ9bBsTd2knpZWctCbjxXc,2011
 nshtrainer/callbacks/directory_setup.py,sha256=c0uY0oTqLcQ3egInHO7G6BeQQgk_xvOLoHH8FR-9U0U,2629
-nshtrainer/callbacks/early_stopping.py,sha256=VWuJz0oN87b6SwBeVc32YNpeJr1wts8K45k8JJJmG9I,4617
+nshtrainer/callbacks/early_stopping.py,sha256=upCNtjkXmwGPwLJmDZacXdBzySG2i_sGIK9QjnaE6tU,4638
 nshtrainer/callbacks/ema.py,sha256=8-WHmKFP3VfnzMviJaIFmVD9xHPqIPmq9NRF5xdu3c8,12131
 nshtrainer/callbacks/finite_checks.py,sha256=gJC_RUr3ais3FJI0uB6wUZnDdE3WRwCix3ppA3PwQXA,2077
 nshtrainer/callbacks/gradient_skipping.py,sha256=EBNkANDnD3BTszWjnG-jwY8FEj-iRqhE3e1x5LQF6M8,3393
@@ -31,7 +31,7 @@ nshtrainer/callbacks/throughput_monitor.py,sha256=H_ocXErZxUO3dxFk8Tx_VQdpI9E_Zt
 nshtrainer/callbacks/timer.py,sha256=quS79oYClDUvQxJkNWmDMe0hwRUkkREgTgqzVrnom50,4607
 nshtrainer/callbacks/wandb_upload_code.py,sha256=OWG4UkL2SfW6oj6AGRXeBJsZmgsqeHLW2Fj8Jm4ga3I,2298
 nshtrainer/callbacks/wandb_watch.py,sha256=Y6SEXfIx3kDDQbI5zpP53BVq0FBLJbLd3RJsiHZk1-Y,2921
-nshtrainer/config.py,sha256=6jApGtO9DVFoXKr9_Z7-MFG5R4WXjbpzZ6jkNI3yD-Y,8306
+nshtrainer/config.py,sha256=pZyRZOkBRR7eBFRiHpHjQFNEFjaX9tYZIAqZtvKi6cA,8312
 nshtrainer/data/__init__.py,sha256=7mk1tr7SWUZ7ySbsf0y0ZPszk7u4QznPhQ-7wnpH9ec,149
 nshtrainer/data/balanced_batch_sampler.py,sha256=ybMJF-CguaZ17fLEweZ5suaGOiHOMEm3Bn8rQfGTzGQ,5445
 nshtrainer/data/transform.py,sha256=6SNs3_TpNpfhcwTwvPKyEJ3opM1OT7LmMEYQNHKgRl8,2227
@@ -82,7 +82,7 @@ nshtrainer/profiler/simple.py,sha256=MbMfsJvligd0mtGiltxJ0T8MQVDP9T9BzQZFwswl66Y
 nshtrainer/runner.py,sha256=USAjrExHkN5oVNVunsoPnLxfQrEHSaa54S3RipOe544,3605
 nshtrainer/scripts/find_packages.py,sha256=ixYivZobumyyGsf2B9oYMLyLTRcBzY_vUv-u3bNW-hs,1424
 nshtrainer/trainer/__init__.py,sha256=P2rmr8oBVTHk-HJHYPcUwWqDEArMbPR4_rPpATbWK3E,40
-nshtrainer/trainer/_config.py,sha256=ZIodM5Ek1lpkWFhQ_VfmKR7q1mZFFwtjfx8FH72H8WM,29174
+nshtrainer/trainer/_config.py,sha256=YqpGb4RodkUg87TVE5WBSc4CQkUF0z3qDRdil1HRxoM,29198
 nshtrainer/trainer/_runtime_callback.py,sha256=sd2cUdRJG-UCdQr9ruZvEYpNGNF1t2W2fuxwwVlQD9E,4164
 nshtrainer/trainer/checkpoint_connector.py,sha256=r0ir4xYSdf_jebM0x09qaO6nJsvsiRQDyM0fs80ppOQ,2347
 nshtrainer/trainer/signal_connector.py,sha256=2EzkVktlasl8PgWAKNLDZRUMY__gRlDy1HdinAU-tfU,10740
@@ -99,6 +99,6 @@ nshtrainer/util/seed.py,sha256=Or2wMPsnQxfnZ2xfBiyMcHFIUt3tGTNeMMyOEanCkqs,280
 nshtrainer/util/slurm.py,sha256=rofIU26z3SdL79SF45tNez6juou1cyDLz07oXEZb9Hg,1566
 nshtrainer/util/typed.py,sha256=NGuDkDzFlc1fAoaXjOFZVbmj0mRFjsQi1E_hPa7Bn5U,128
 nshtrainer/util/typing_utils.py,sha256=8ptjSSLZxlmy4FY6lzzkoGoF5fGNClo8-B_c0XHQaNU,385
-nshtrainer-0.37.0.dist-info/METADATA,sha256=Rd7HeNaz5lBQa0AspjWyRgFLctqpSfna2R9VnMEUURU,916
-nshtrainer-0.37.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-nshtrainer-0.37.0.dist-info/RECORD,,
+nshtrainer-0.39.0.dist-info/METADATA,sha256=bIWwvsGuePEZeT3Q8dYS8IK5Y6ZM4yqe9v41Ybs9OGM,916
+nshtrainer-0.39.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+nshtrainer-0.39.0.dist-info/RECORD,,