nshtrainer 1.0.0b46__py3-none-any.whl → 1.0.0b48__py3-none-any.whl

nshtrainer/_callback.py

@@ -4,46 +4,46 @@ from pathlib import Path
 from typing import TYPE_CHECKING, Any
 
 import torch
+from lightning.pytorch import LightningModule
 from lightning.pytorch.callbacks import Callback as _LightningCallback
 from lightning.pytorch.utilities.types import STEP_OUTPUT
 from torch.optim import Optimizer
 
 if TYPE_CHECKING:
-    from .model import LightningModuleBase
     from .trainer import Trainer
 
 
 class NTCallbackBase(_LightningCallback):
     def setup(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase, stage: str
+        self, trainer: Trainer, pl_module: LightningModule, stage: str
     ) -> None:
         """Called when fit, validate, test, predict, or tune begins."""
 
     def teardown(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase, stage: str
+        self, trainer: Trainer, pl_module: LightningModule, stage: str
     ) -> None:
         """Called when fit, validate, test, predict, or tune ends."""
 
-    def on_fit_start(self, trainer: Trainer, pl_module: LightningModuleBase) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
+    def on_fit_start(self, trainer: Trainer, pl_module: LightningModule) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
         """Called when fit begins."""
 
-    def on_fit_end(self, trainer: Trainer, pl_module: LightningModuleBase) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
+    def on_fit_end(self, trainer: Trainer, pl_module: LightningModule) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
         """Called when fit ends."""
 
     def on_sanity_check_start(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the validation sanity check starts."""
 
     def on_sanity_check_end(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the validation sanity check ends."""
 
     def on_train_batch_start(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         batch: Any,
         batch_idx: int,
     ) -> None:
@@ -52,7 +52,7 @@ class NTCallbackBase(_LightningCallback):
     def on_train_batch_end(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         outputs: STEP_OUTPUT,
         batch: Any,
         batch_idx: int,
@@ -66,12 +66,12 @@ class NTCallbackBase(_LightningCallback):
         """
 
     def on_train_epoch_start(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the train epoch begins."""
 
     def on_train_epoch_end(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the train epoch ends.
 
@@ -102,39 +102,39 @@ class NTCallbackBase(_LightningCallback):
         """
 
     def on_validation_epoch_start(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the val epoch begins."""
 
     def on_validation_epoch_end(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the val epoch ends."""
 
     def on_test_epoch_start(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the test epoch begins."""
 
     def on_test_epoch_end(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the test epoch ends."""
 
     def on_predict_epoch_start(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the predict epoch begins."""
 
     def on_predict_epoch_end(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the predict epoch ends."""
 
     def on_validation_batch_start(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         batch: Any,
         batch_idx: int,
         dataloader_idx: int = 0,
@@ -144,7 +144,7 @@ class NTCallbackBase(_LightningCallback):
     def on_validation_batch_end(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         outputs: STEP_OUTPUT,
         batch: Any,
         batch_idx: int,
@@ -155,7 +155,7 @@ class NTCallbackBase(_LightningCallback):
     def on_test_batch_start(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         batch: Any,
         batch_idx: int,
         dataloader_idx: int = 0,
@@ -165,7 +165,7 @@ class NTCallbackBase(_LightningCallback):
     def on_test_batch_end(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         outputs: STEP_OUTPUT,
         batch: Any,
         batch_idx: int,
@@ -176,7 +176,7 @@ class NTCallbackBase(_LightningCallback):
     def on_predict_batch_start(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         batch: Any,
         batch_idx: int,
         dataloader_idx: int = 0,
@@ -186,7 +186,7 @@ class NTCallbackBase(_LightningCallback):
     def on_predict_batch_end(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         outputs: Any,
         batch: Any,
         batch_idx: int,
@@ -194,40 +194,40 @@ class NTCallbackBase(_LightningCallback):
     ) -> None:
         """Called when the predict batch ends."""
 
-    def on_train_start(self, trainer: Trainer, pl_module: LightningModuleBase) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
+    def on_train_start(self, trainer: Trainer, pl_module: LightningModule) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
         """Called when the train begins."""
 
-    def on_train_end(self, trainer: Trainer, pl_module: LightningModuleBase) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
+    def on_train_end(self, trainer: Trainer, pl_module: LightningModule) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
         """Called when the train ends."""
 
     def on_validation_start(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the validation loop begins."""
 
     def on_validation_end(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the validation loop ends."""
 
-    def on_test_start(self, trainer: Trainer, pl_module: LightningModuleBase) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
+    def on_test_start(self, trainer: Trainer, pl_module: LightningModule) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
         """Called when the test begins."""
 
-    def on_test_end(self, trainer: Trainer, pl_module: LightningModuleBase) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
+    def on_test_end(self, trainer: Trainer, pl_module: LightningModule) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
         """Called when the test ends."""
 
     def on_predict_start(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called when the predict begins."""
 
-    def on_predict_end(self, trainer: Trainer, pl_module: LightningModuleBase) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
+    def on_predict_end(self, trainer: Trainer, pl_module: LightningModule) -> None:  # pyright: ignore[reportIncompatibleMethodOverride]
         """Called when predict ends."""
 
     def on_exception(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         exception: BaseException,
     ) -> None:
         """Called when any trainer execution is interrupted by an exception."""
@@ -253,7 +253,7 @@ class NTCallbackBase(_LightningCallback):
     def on_save_checkpoint(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         checkpoint: dict[str, Any],
     ) -> None:
         r"""Called when saving a checkpoint to give you a chance to store anything else you might want to save.
@@ -268,7 +268,7 @@ class NTCallbackBase(_LightningCallback):
     def on_load_checkpoint(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         checkpoint: dict[str, Any],
     ) -> None:
         r"""Called when loading a model checkpoint, use to reload state.
@@ -281,19 +281,19 @@ class NTCallbackBase(_LightningCallback):
         """
 
     def on_before_backward(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase, loss: torch.Tensor
+        self, trainer: Trainer, pl_module: LightningModule, loss: torch.Tensor
     ) -> None:
         """Called before ``loss.backward()``."""
 
     def on_after_backward(  # pyright: ignore[reportIncompatibleMethodOverride]
-        self, trainer: Trainer, pl_module: LightningModuleBase
+        self, trainer: Trainer, pl_module: LightningModule
     ) -> None:
         """Called after ``loss.backward()`` and before optimizers are stepped."""
 
     def on_before_optimizer_step(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         optimizer: Optimizer,
     ) -> None:
         """Called before ``optimizer.step()``."""
@@ -301,7 +301,7 @@ class NTCallbackBase(_LightningCallback):
     def on_before_zero_grad(  # pyright: ignore[reportIncompatibleMethodOverride]
         self,
         trainer: Trainer,
-        pl_module: LightningModuleBase,
+        pl_module: LightningModule,
         optimizer: Optimizer,
     ) -> None:
         """Called before ``optimizer.zero_grad()``."""
@@ -310,15 +310,15 @@ class NTCallbackBase(_LightningCallback):
         self,
         ckpt_path: Path,
         metadata_path: Path | None,
-        trainer: "Trainer",
-        pl_module: "LightningModuleBase",
+        trainer: Trainer,
+        pl_module: LightningModule,
     ) -> None:
         """Called after a checkpoint is saved."""
         pass
 
 
 def _call_on_checkpoint_saved(
-    trainer: "Trainer",
+    trainer: Trainer,
     ckpt_path: str | Path,
     metadata_path: str | Path | None,
 ):
@@ -333,5 +333,5 @@ def _call_on_checkpoint_saved(
             ckpt_path,
             metadata_path,
             trainer,
-            trainer._base_module,
+            trainer.lightning_module,
         )

nshtrainer/_checkpoint/metadata.py

@@ -16,6 +16,18 @@ from ..util.path import compute_file_checksum, try_symlink_or_copy
 if TYPE_CHECKING:
     from ..trainer.trainer import Trainer
 
+try:
+    from pydantic import BaseModel
+
+    _HAS_PYDANTIC = True
+except ImportError:
+    if not TYPE_CHECKING:
+        BaseModel = object
+    else:
+        from pydantic import BaseModel
+    _HAS_PYDANTIC = False
+
+
 log = logging.getLogger(__name__)
 
 
@@ -27,10 +39,10 @@ def _full_hparams_dict(trainer: Trainer):
     hparams["trainer"] = trainer.hparams.model_dump(mode="json")
 
     if trainer.lightning_module is not None:
-        from ..model import LightningModuleBase
-
-        if isinstance(trainer.lightning_module, LightningModuleBase):
-            hparams["model"] = trainer.lightning_module.hparams.model_dump(mode="json")
+        model_hparams = trainer.lightning_module.hparams
+        if _HAS_PYDANTIC and isinstance(model_hparams, BaseModel):
+            model_hparams = model_hparams.model_dump(mode="json")
+        hparams["model"] = dict(model_hparams)
 
     return hparams
 

nshtrainer/callbacks/checkpoint/best_checkpoint.py

@@ -51,7 +51,7 @@ class BestCheckpointCallbackConfig(BaseCheckpointCallbackConfig):
 class BestCheckpointCallback(CheckpointBase[BestCheckpointCallbackConfig]):
     @property
     def _metric_name_normalized(self):
-        return self.metric.name.replace("/", "_").replace(" ", "_").replace(".", "_")
+        return self.metric.monitor.replace("/", "_").replace(" ", "_").replace(".", "_")
 
     @override
     def __init__(
@@ -69,12 +69,12 @@ class BestCheckpointCallback(CheckpointBase[BestCheckpointCallbackConfig]):
 
     @override
     def default_filename(self):
-        return f"epoch{{epoch}}-step{{step}}-{self._metric_name_normalized}{{{self.metric.validation_monitor}}}"
+        return f"epoch{{epoch}}-step{{step}}-{self._metric_name_normalized}{{{self.metric.monitor}}}"
 
     @override
     def topk_sort_key(self, metadata: CheckpointMetadata):
         return metadata.metrics.get(
-            self.metric.validation_monitor,
+            self.metric.monitor,
             float("-inf" if self.metric.mode == "max" else "inf"),
         )
 

nshtrainer/callbacks/early_stopping.py

@@ -68,7 +68,7 @@ class EarlyStoppingCallback(_EarlyStopping):
         del config, metric
 
         super().__init__(
-            monitor=self.metric.validation_monitor,
+            monitor=self.metric.monitor,
             mode=self.metric.mode,
             patience=self.config.patience,
             min_delta=self.config.min_delta,

nshtrainer/callbacks/metric_validation.py

@@ -55,14 +55,14 @@ class MetricValidationCallback(Callback):
         self.metrics = metrics
 
     def _check_metrics(self, trainer: Trainer):
-        metric_names = ", ".join(metric.validation_monitor for metric in self.metrics)
+        metric_names = ", ".join(metric.monitor for metric in self.metrics)
         log.info(f"Validating metrics: {metric_names}...")
         logged_metrics = set(trainer.logged_metrics.keys())
 
         invalid_metrics: list[str] = []
         for metric in self.metrics:
-            if metric.validation_monitor not in logged_metrics:
-                invalid_metrics.append(metric.validation_monitor)
+            if metric.monitor not in logged_metrics:
+                invalid_metrics.append(metric.monitor)
 
         if invalid_metrics:
             msg = (

nshtrainer/callbacks/rlp_sanity_checks.py

@@ -171,7 +171,7 @@ class CustomRLPImplementation(Protocol):
     __reduce_lr_on_plateau__: bool
 
 
-class _RLPSanityCheckModuleMixin(LightningModule):
+class RLPSanityCheckModuleMixin(LightningModule):
     def reduce_lr_on_plateau_config(
         self,
         lr_scheduler: LRSchedulerTypeUnion | LRSchedulerConfigType,

nshtrainer/data/datamodule.py

@@ -11,13 +11,13 @@ from lightning.pytorch import LightningDataModule
 from typing_extensions import Never, TypeVar, deprecated, override
 
 from ..model.mixins.callback import CallbackRegistrarModuleMixin
-from ..model.mixins.debug import _DebugModuleMixin
+from ..model.mixins.debug import DebugModuleMixin
 
 THparams = TypeVar("THparams", bound=C.Config, infer_variance=True)
 
 
 class LightningDataModuleBase(
-    _DebugModuleMixin,
+    DebugModuleMixin,
     CallbackRegistrarModuleMixin,
     LightningDataModule,
     ABC,

nshtrainer/loggers/__init__.py

@@ -2,7 +2,6 @@ from __future__ import annotations
 
 from typing import Annotated
 
-import nshconfig as C
 from typing_extensions import TypeAliasType
 
 from .actsave import ActSaveLoggerConfig as ActSaveLoggerConfig

nshtrainer/lr_scheduler/base.py

@@ -3,9 +3,10 @@ from __future__ import annotations
 import math
 from abc import ABC, abstractmethod
 from collections.abc import Mapping
-from typing import TYPE_CHECKING, Literal
+from typing import Literal
 
 import nshconfig as C
+from lightning.pytorch import LightningModule
 from lightning.pytorch.utilities.types import (
     LRSchedulerConfigType,
     LRSchedulerTypeUnion,
@@ -13,9 +14,6 @@ from lightning.pytorch.utilities.types import (
 from torch.optim import Optimizer
 from typing_extensions import Never, NotRequired, TypedDict
 
-if TYPE_CHECKING:
-    from ..model.base import LightningModuleBase
-
 
 class LRSchedulerMetadata(TypedDict):
     interval: Literal["epoch", "step"]
@@ -44,13 +42,13 @@ class LRSchedulerConfigBase(C.Config, ABC):
 
     @abstractmethod
     def create_scheduler_impl(
-        self, optimizer: Optimizer, lightning_module: LightningModuleBase
+        self, optimizer: Optimizer, lightning_module: LightningModule
     ) -> LRSchedulerTypeUnion | LRSchedulerConfigType: ...
 
     def create_scheduler(
         self,
         optimizer: Optimizer,
-        lightning_module: LightningModuleBase,
+        lightning_module: LightningModule,
         lr: Never
         | None = None,  # Backward compatibility, should be removed in the future
     ) -> LRSchedulerConfigType:
@@ -87,7 +85,7 @@ class LRSchedulerConfigBase(C.Config, ABC):
 
         return scheduler
 
-    def compute_num_steps_per_epoch(self, lightning_module: LightningModuleBase) -> int:
+    def compute_num_steps_per_epoch(self, lightning_module: LightningModule) -> int:
         trainer = lightning_module.trainer
         # Use the Lightning trainer to convert the epoch-based values to step-based values
         _ = trainer.estimated_stepping_batches

nshtrainer/lr_scheduler/reduce_lr_on_plateau.py

@@ -49,13 +49,13 @@ class ReduceLROnPlateauConfig(LRSchedulerConfigBase):
         if (metric := self.metric) is None:
             from ..trainer import Trainer
 
-            assert isinstance(
-                trainer := lightning_module.trainer, Trainer
-            ), "The trainer must be a `nshtrainer.Trainer` instance."
+            assert isinstance(trainer := lightning_module.trainer, Trainer), (
+                "The trainer must be a `nshtrainer.Trainer` instance."
+            )
 
-            assert (
-                metric := trainer.hparams.primary_metric
-            ) is not None, "Primary metric must be provided if metric is not specified."
+            assert (metric := trainer.hparams.primary_metric) is not None, (
+                "Primary metric must be provided if metric is not specified."
+            )
 
         lr_scheduler = ReduceLROnPlateau(
             optimizer,
@@ -70,7 +70,7 @@ class ReduceLROnPlateauConfig(LRSchedulerConfigBase):
         )
         return {
             "scheduler": lr_scheduler,
-            "monitor": metric.validation_monitor,
+            "monitor": metric.monitor,
         }
 
     @override

nshtrainer/metrics/_config.py

@@ -7,8 +7,8 @@ import nshconfig as C
 
 
 class MetricConfig(C.Config):
-    name: str
-    """The name of the primary metric."""
+    monitor: str
+    """The name of the metric to monitor."""
 
     mode: Literal["min", "max"]
     """
@@ -17,23 +17,6 @@ class MetricConfig(C.Config):
     - "max" for metrics that should be maximized (e.g., accuracy)
     """
 
-    @property
-    def validation_monitor(self) -> str:
-        return f"val/{self.name}"
-
-    def __post_init__(self):
-        for split in ("train", "val", "test", "predict"):
-            if self.name.startswith(f"{split}/"):
-                raise ValueError(
-                    f"Primary metric name should not start with '{split}/'. "
-                    f"Just use '{self.name[len(split) + 1:]}' instead. "
-                    "The split name is automatically added depending on the context."
-                )
-
-    @classmethod
-    def loss(cls, mode: Literal["min", "max"] = "min"):
-        return cls(name="loss", mode=mode)
-
     @property
     def best(self):
         return builtins.min if self.mode == "min" else builtins.max

nshtrainer/model/base.py

@@ -15,9 +15,9 @@ from lightning.pytorch.utilities.model_helpers import is_overridden
 from lightning.pytorch.utilities.rank_zero import rank_zero_warn
 from typing_extensions import Never, TypeVar, deprecated, override
 
-from ..callbacks.rlp_sanity_checks import _RLPSanityCheckModuleMixin
+from ..callbacks.rlp_sanity_checks import RLPSanityCheckModuleMixin
 from .mixins.callback import CallbackModuleMixin
-from .mixins.debug import _DebugModuleMixin
+from .mixins.debug import DebugModuleMixin
 from .mixins.logger import LoggerLightningModuleMixin
 
 log = logging.getLogger(__name__)
@@ -54,81 +54,14 @@ VALID_REDUCE_OPS = (
 
 
 class LightningModuleBase(
-    _DebugModuleMixin,
-    _RLPSanityCheckModuleMixin,
+    DebugModuleMixin,
+    RLPSanityCheckModuleMixin,
     LoggerLightningModuleMixin,
     CallbackModuleMixin,
     LightningModule,
     ABC,
     Generic[THparams],
 ):
-    # region Debug
-    @property
-    def debug(self) -> bool:
-        if torch.jit.is_scripting():
-            return False
-
-        if (trainer := self._trainer) is None:
-            return False
-
-        from ..trainer import Trainer
-
-        if not isinstance(trainer, Trainer):
-            return False
-
-        return trainer.debug
-
-    @debug.setter
-    def debug(self, value: bool):
-        if torch.jit.is_scripting():
-            return
-
-        if (trainer := self._trainer) is None:
-            return
-
-        from ..trainer import Trainer
-
-        if not isinstance(trainer, Trainer):
-            return
-
-        trainer.debug = value
-
-    @torch.jit.unused
-    def breakpoint(self, rank_zero_only: bool = True):
-        if (
-            not rank_zero_only
-            or not torch.distributed.is_initialized()
-            or torch.distributed.get_rank() == 0
-        ):
-            breakpoint()
-
-        if rank_zero_only and torch.distributed.is_initialized():
-            _ = torch.distributed.barrier()
-
-    @torch.jit.unused
-    def ensure_finite(
-        self,
-        tensor: torch.Tensor,
-        name: str | None = None,
-        throw: bool = False,
-    ):
-        name_parts: list[str] = ["Tensor"]
-        if name is not None:
-            name_parts.append(name)
-        name = " ".join(name_parts)
-
-        not_finite = ~torch.isfinite(tensor)
-        if not_finite.any():
-            msg = f"{name} has {not_finite.sum().item()}/{not_finite.numel()} non-finite values."
-            if throw:
-                raise RuntimeError(msg)
-            else:
-                log.warning(msg)
-            return False
-        return True
-
-    # endregion
-
     # region Profiler
     @property
     def profiler(self) -> Profiler:

nshtrainer/model/mixins/debug.py

@@ -28,7 +28,7 @@ def _trainer(module: Any):
     return trainer
 
 
-class _DebugModuleMixin:
+class DebugModuleMixin:
     @property
     def nshtrainer_or_none(self):
         return _trainer(self)

nshtrainer/model/mixins/logger.py

@@ -54,6 +54,12 @@ class _LogContextKwargs:
         d = dataclasses.asdict(self)
         for field in self.__ignore_fields__:
             d.pop(field, None)
+
+        # Pop all None values
+        for k in list(d.keys()):
+            if d[k] is None:
+                d.pop(k)
+
         return d
 
 
@@ -134,18 +140,18 @@ class LoggerLightningModuleMixin(mixin_base_type(LightningModule)):
         self,
         name: str,
         value: _METRIC,
-        prog_bar: bool = False,
+        prog_bar: bool | None = None,
         logger: bool | None = None,
         on_step: bool | None = None,
         on_epoch: bool | None = None,
-        reduce_fx: str | Callable = "mean",
-        enable_graph: bool = False,
-        sync_dist: bool = False,
+        reduce_fx: str | Callable | None = None,
+        enable_graph: bool | None = None,
+        sync_dist: bool | None = None,
         sync_dist_group: Any | None = None,
-        add_dataloader_idx: bool = True,
+        add_dataloader_idx: bool | None = None,
         batch_size: int | None = None,
         metric_attribute: str | None = None,
-        rank_zero_only: bool = False,
+        rank_zero_only: bool | None = None,
     ) -> None:
         # If logging is disabled, then do nothing.
         if not self.logging_enabled:

nshtrainer/trainer/trainer.py

@@ -418,20 +418,6 @@ class Trainer(LightningTrainer):
 
         return tracker.time_elapsed(stage)
 
-    @property
-    def _base_module(self):
-        if self.lightning_module is None:
-            raise ValueError("LightningModule is not set.")
-
-        from ..model.base import LightningModuleBase
-
-        if not isinstance(self.lightning_module, LightningModuleBase):
-            raise ValueError(
-                f"LightningModule is not an instance of {LightningModuleBase}."
-            )
-
-        return self.lightning_module
-
     @override
     def _run(
         self, model: LightningModule, ckpt_path: str | Path | None = None

{nshtrainer-1.0.0b46.dist-info → nshtrainer-1.0.0b48.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: nshtrainer
-Version: 1.0.0b46
+Version: 1.0.0b48
 Summary: 
 Author: Nima Shoghi
 Author-email: nimashoghi@gmail.com

{nshtrainer-1.0.0b46.dist-info → nshtrainer-1.0.0b48.dist-info}/RECORD

@@ -1,7 +1,7 @@
 nshtrainer/.nshconfig.generated.json,sha256=yZd6cn1RhvNNJUgiUTRYut8ofZYvbulnpPG-rZIRhi4,106
 nshtrainer/__init__.py,sha256=g_moPnfQxSxFZX5NB9ILQQOJrt4RTRuiFt9N0STIpxM,874
-nshtrainer/_callback.py,sha256=tXQCDzS6CvMTuTY5lQSH5qZs1pXUi-gt9bQdpXMVdEs,12715
-nshtrainer/_checkpoint/metadata.py,sha256=XoKqY3eR95CYuc_Kk9ck-p4iM2Q1OXU3vSXNrzohHz0,5332
+nshtrainer/_callback.py,sha256=ZDppiJ4d65tRXTEWYPZLH_F1xFizdz1pkWJe_sQ5uII,12564
+nshtrainer/_checkpoint/metadata.py,sha256=ojSEmq0udFwdzIC5vkbF0yEdhMaJ2iBrZCSFNDkeeGY,5578
 nshtrainer/_checkpoint/saver.py,sha256=65UDrz3KuhkgVfco-RkWuoa1wzTZoXxunlC769yJaMc,1639
 nshtrainer/_directory.py,sha256=TJR9ccyuzRlAVfVjGyeQ3E2AFAcz-XbBCxWfiXo2SlY,3191
 nshtrainer/_experimental/__init__.py,sha256=U4S_2y3zgLZVfMenHRaJFBW8yqh2mUBuI291LGQVOJ8,35
@@ -11,22 +11,22 @@ nshtrainer/callbacks/actsave.py,sha256=NSXIIu62MNYe5gz479SMW33bdoKYoYtWtd_iTWFpK
 nshtrainer/callbacks/base.py,sha256=Alaou1IHAIlMEM7g58d_02ozY2xWlshBN7fsw5Ee21s,3683
 nshtrainer/callbacks/checkpoint/__init__.py,sha256=l8tkHc83_mLiU0-wT09SWdRzwpm2ulbkLzcuCmuTwzE,620
 nshtrainer/callbacks/checkpoint/_base.py,sha256=f7lpk8W4xqxk3PolBEU3AWt9VTIpoLW7wMUhC5DNm3c,6345
-nshtrainer/callbacks/checkpoint/best_checkpoint.py,sha256=2CQuhPJ3Fi7lDw7z-J8kXXXuDU8-4HcU48oZxR49apk,2667
+nshtrainer/callbacks/checkpoint/best_checkpoint.py,sha256=aCs3E1eucfDlUeW2Iq_Ke7hb96BxHanmvn7PCCbqq0E,2648
 nshtrainer/callbacks/checkpoint/last_checkpoint.py,sha256=vn-as3ex7kaTRcKsIurVtM6kUSHYNwHJeYG82j2dMcc,3554
 nshtrainer/callbacks/checkpoint/on_exception_checkpoint.py,sha256=nljzETqkHwA-4g8mxaeFK5HxA8My0dlIPzIUscSMWyk,3525
 nshtrainer/callbacks/debug_flag.py,sha256=96fuP0C7C6dSs1GiMeUYzzs0X3Q4Pjt9JVWg3b75fU4,1748
 nshtrainer/callbacks/directory_setup.py,sha256=wPas_Ren8ANejogmIdKhqqgj4ulxz9AS_8xVIAfRXa0,2565
-nshtrainer/callbacks/early_stopping.py,sha256=EjzN-gD_Xd4YHZLkXsbi00g_4ti3RTMJEdHJ8GMeaFM,4776
+nshtrainer/callbacks/early_stopping.py,sha256=rC_qYKCQWjRQJFo0ky46uG0aDJdYP8vsSlKunk0bUVI,4765
 nshtrainer/callbacks/ema.py,sha256=dBFiUXG0xmyCw8-ayuSzJMKqSbepl6Ii5VIbhFlT5ug,12255
 nshtrainer/callbacks/finite_checks.py,sha256=3lZ3kEIjmYQfqTF0DcrgZ9_98ZLQhQj8usH7SgWst3o,2185
 nshtrainer/callbacks/gradient_skipping.py,sha256=8g7oC7PF0LTAEzwiNoaS5tWOnkjk_EB0QG3JdHkQ8ek,3523
 nshtrainer/callbacks/interval.py,sha256=UCzUzt3XCFVyQyCWL9lOrStkkxesvduNOYk8yMrGTTk,8116
 nshtrainer/callbacks/log_epoch.py,sha256=B5Dm8XVZwCzKUhUWfT_5PDdDac993191OsbcxxuSVJE,1457
 nshtrainer/callbacks/lr_monitor.py,sha256=qy_C0R40J0hBAukzBwng5FI2jJUpWuXOi5N6FU6ym3I,1210
-nshtrainer/callbacks/metric_validation.py,sha256=tqUVS2n9QRT3v1_8jAGlYBFhLpA6Bm9pxOsfWhD3yZQ,2915
+nshtrainer/callbacks/metric_validation.py,sha256=4RDr1FuNKfro-6QEtmcFqT4iNf2twmJVNk9y-8nq9bg,2882
 nshtrainer/callbacks/norm_logging.py,sha256=nVIDWe-ASl5zN830-ODR8QMCqI1ma-QPCIwoy0Wb-Nk,6390
 nshtrainer/callbacks/print_table.py,sha256=VaS4JgI963do79laXK4lUkFQx8v6aRSy22W0zyal_LA,3035
-nshtrainer/callbacks/rlp_sanity_checks.py,sha256=74BZvV2HLO__ucQXsLXb8eJLUZgRFUNJZ6TL9efMp74,10051
+nshtrainer/callbacks/rlp_sanity_checks.py,sha256=Df9Prq2QKXnaeMBIvMQBhDhJTDeru5UbiuXJOJR16Gk,10050
 nshtrainer/callbacks/shared_parameters.py,sha256=s94jJTAIbDGukYJu6l247QonVOCudGClU4t5kLt8XrY,3076
 nshtrainer/callbacks/timer.py,sha256=gDcw_K_ikf0bkVgxQ0cDhvvNvz6GLZVLcatuKfh0ORU,4731
 nshtrainer/callbacks/wandb_upload_code.py,sha256=shV7UtnXgY2bUlXdVrXiaDs0PNLlIt7TzNJkJPkzvzI,2414
@@ -100,25 +100,25 @@ nshtrainer/configs/util/config/dtype/__init__.py,sha256=PmGF-O4r6SXqEaagVsQ5YxEq
 nshtrainer/configs/util/config/duration/__init__.py,sha256=44lS2irOIPVfgshMTfnZM2jC6l0Pjst9w2M_lJoS_MU,353
 nshtrainer/data/__init__.py,sha256=K4i3Tw4g9EOK2zlMMbidi99y0SyI4w8P7_XUf1n42Ts,260
 nshtrainer/data/balanced_batch_sampler.py,sha256=r1cBKRXKHD8E1Ax6tj-FUbE-z1qpbO58mQ9VrK9uLnc,5481
-nshtrainer/data/datamodule.py,sha256=lSOgH32nysJWa6Y7ba1QyOdUV0DVVdO98qokP8wigjk,4138
+nshtrainer/data/datamodule.py,sha256=0M-HjGZQkLG77HXn4ZgLSypnbSjkjTq6GEJwGWe_gbM,4136
 nshtrainer/data/transform.py,sha256=qd0lIocO59Fk_m90xyOHgFezbymd1mRwly8nbYIfHGc,2263
-nshtrainer/loggers/__init__.py,sha256=Ddd3JJXVzew_ZpwHA9kGnGmvq4OwhItwghDL5PzNhDc,614
+nshtrainer/loggers/__init__.py,sha256=fI0OHEltHP4tZI-KFB3npdzoxm_M2QsEYKxY3um05_s,592
 nshtrainer/loggers/actsave.py,sha256=wgNrpBB6wQM7qff8iLDb_sQnbiAcYHRmH56pcEJPB3o,1409
 nshtrainer/loggers/base.py,sha256=ON92XbwTSgadQOSyw5PiRRFzyH6uJ-xLtE0nB3cbgPc,1205
 nshtrainer/loggers/csv.py,sha256=xJ8mSmw4vJwinIfqhF6t2HWmh_1dXEYyLfGuXwL7WHo,1160
 nshtrainer/loggers/tensorboard.py,sha256=E7iO_fDt9bfH02hBL430bXPLljOo5iGgq2QyPqmx2gQ,2324
 nshtrainer/loggers/wandb.py,sha256=KZXAUWrrmdX_L8rqej77oUHaM0JxZRM8y9z6JP9PISw,6856
 nshtrainer/lr_scheduler/__init__.py,sha256=daMMK3erUcNXGGd_nZB8AWu3ZTYqfS1RSWeK4FV2udw,851
-nshtrainer/lr_scheduler/base.py,sha256=062fGcH5sYeEKwoY55RydCTvfPwTnyZHCi049a3nMbM,3805
+nshtrainer/lr_scheduler/base.py,sha256=LE53JRBTuAlA1fqbMgCZ7m39D1z0rGj2TizhJ62CPvE,3756
 nshtrainer/lr_scheduler/linear_warmup_cosine.py,sha256=MsoXgCcWTKsrkNZiGnKS6yC-slRuleuwFxeM_lmG_pQ,5560
-nshtrainer/lr_scheduler/reduce_lr_on_plateau.py,sha256=v9T0GpvOoHV30atFB0MwExHgHcTpMCYxbMRoPjPBjt8,2938
+nshtrainer/lr_scheduler/reduce_lr_on_plateau.py,sha256=zKO_4Cl28m3TopoNFmc5H6GSUuVUGYUoAlXpMh_EJIk,2931
 nshtrainer/metrics/__init__.py,sha256=Nqkn_jsDf3n5WtfMcnaaEftYjIIT2b-S7rmsB1MOMkU,86
-nshtrainer/metrics/_config.py,sha256=XIRokFM8PHrhBa3w2R6BM6a4es3ncsoBqE_LqXQFsFE,1223
+nshtrainer/metrics/_config.py,sha256=ox_ScK6V0J9nzIMhEB0qpToNKpt83VVgOVSRFCV-wBc,595
 nshtrainer/model/__init__.py,sha256=3G-bwPPSRStWdsdwG9-rn0bXcRpEiP1BiQpF_qavtls,97
-nshtrainer/model/base.py,sha256=JL3AmH17GQjQIoMrZl3O0vUI7dj5ZsO5iEJgoLPyzHw,10356
+nshtrainer/model/base.py,sha256=bZMNap0rkxRbAbu2BOHV_6YS2iZZnvy6wVSMOXGa_ZM,8680
 nshtrainer/model/mixins/callback.py,sha256=0LPgve4VszHbLipid4mpI1qnnmdGS2spivs0dXLvqHw,3154
-nshtrainer/model/mixins/debug.py,sha256=1LX9KzeFX9JDPs_a6YCdYDZXLhEk_5rBO2aCqlfBy7w,2087
-nshtrainer/model/mixins/logger.py,sha256=27H99FuLaxc6_dDLG2pid4E_5E0-eLGnc2Ifpt0HYIM,6066
+nshtrainer/model/mixins/debug.py,sha256=ydLuAAaa7M5bX0gougZ5gWuZnvn4Ra9assal3IZ9hq8,2086
+nshtrainer/model/mixins/logger.py,sha256=IYfyyW_1VAD_HiTsfX28P-XNgz_SMb07t5lwb5rjlZ0,6221
 nshtrainer/nn/__init__.py,sha256=5Gg3nieGSC5_dXaI9KUVUUbM13hHexH9831m4hcf6no,1475
 nshtrainer/nn/mlp.py,sha256=nYUgAISzuhC8sav6PloAdyz0PdEoikwppiXIuToEVdE,7550
 nshtrainer/nn/module_dict.py,sha256=9plb8aQUx5TUEPhX5jI9u8LrpTeKe7jZAHi8iIqcN8w,2365
@@ -142,7 +142,7 @@ nshtrainer/trainer/plugin/layer_sync.py,sha256=h-ydZwXepnsw5-paLgiDatqPyQ_8C0QEv
 nshtrainer/trainer/plugin/precision.py,sha256=I0QsB1bVxmsFmBOkgrAfGONsuYae_lD9Bz0PfJEQvH4,5598
 nshtrainer/trainer/signal_connector.py,sha256=GhfGcSzfaTNhnj2QFkBDq5aT7FqbLMA7eC8SYQs8_8w,10828
 nshtrainer/trainer/strategy.py,sha256=VPTn5z3zvXTydY8IJchjhjcOfpvtoejnvUkq5E4WTus,1368
-nshtrainer/trainer/trainer.py,sha256=8wMe0qArbDfStS4UdmuKSC2aiAImR3mhj14_kCJiNSM,20797
+nshtrainer/trainer/trainer.py,sha256=QEK-0bcw1y5Cconi99PYFXr0MElUGgGYMZ_SlcJUQ1k,20364
 nshtrainer/util/_environment_info.py,sha256=MT8mBe6ZolRfKiwU-les1P-lPNPqXpHQcfADrh_A3uY,24629
 nshtrainer/util/bf16.py,sha256=9QhHZCkYSfYpIcxwAMoXyuh2yTSHBzT-EdLQB297jEs,762
 nshtrainer/util/config/__init__.py,sha256=Z39JJufSb61Lhn2GfVcv3eFW_eorOrN9-9llDWlnZZM,272
@@ -154,6 +154,6 @@ nshtrainer/util/seed.py,sha256=diMV8iwBKN7Xxt5pELmui-gyqyT80_CZzomrWhNss0k,316
 nshtrainer/util/slurm.py,sha256=HflkP5iI_r4UHMyPjw9R4dD5AHsJUpcfJw5PLvGYBRM,1603
 nshtrainer/util/typed.py,sha256=Xt5fUU6zwLKSTLUdenovnKK0N8qUq89Kddz2_XeykVQ,164
 nshtrainer/util/typing_utils.py,sha256=MjY-CUX9R5Tzat-BlFnQjwl1PQ_W2yZQoXhkYHlJ_VA,442
-nshtrainer-1.0.0b46.dist-info/METADATA,sha256=L6-5RyLlIcoFyURkoCuHsAgItT0gSl6Ip0l4iDKvs4o,988
-nshtrainer-1.0.0b46.dist-info/WHEEL,sha256=XbeZDeTWKc1w7CSIyre5aMDU_-PohRwTQceYnisIYYY,88
-nshtrainer-1.0.0b46.dist-info/RECORD,,
+nshtrainer-1.0.0b48.dist-info/METADATA,sha256=b26a0GYVQcEszYiodjGF34N7gvEKONBVuB1bXTv35U4,988
+nshtrainer-1.0.0b48.dist-info/WHEEL,sha256=XbeZDeTWKc1w7CSIyre5aMDU_-PohRwTQceYnisIYYY,88
+nshtrainer-1.0.0b48.dist-info/RECORD,,