Deprecate LightningModule.on_pretrain_routine_{start/end}

CHANGELOG.md

@@ -442,6 +442,9 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Deprecated `on_pretrain_routine_start` and `on_pretrain_routine_end` callback hooks in favor of `on_fit_start` ([#11794](https://github.com/PyTorchLightning/pytorch-lightning/pull/11794))
 
 
+- Deprecated `LightningModule.on_pretrain_routine_start` and `LightningModule.on_pretrain_routine_end` hooks in favor of `on_fit_start` ([#12122](https://github.com/PyTorchLightning/pytorch-lightning/pull/12122))
+
+
 - Deprecated `agg_key_funcs` and `agg_default_func` parameters from `LightningLoggerBase` ([#11871](https://github.com/PyTorchLightning/pytorch-lightning/pull/11871))
 
 

docs/source/common/lightning_module.rst

@@ -1225,9 +1225,6 @@ for more information.
         setup("fit")
         configure_optimizers()
 
-        on_pretrain_routine_start()
-        on_pretrain_routine_end()
-
         # the sanity check runs here
 
         on_train_start()
@@ -1391,18 +1388,6 @@ on_validation_end
 .. automethod:: pytorch_lightning.core.lightning.LightningModule.on_validation_end
     :noindex:
 
-on_pretrain_routine_start
-~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. automethod:: pytorch_lightning.core.lightning.LightningModule.on_pretrain_routine_start
-    :noindex:
-
-on_pretrain_routine_end
-~~~~~~~~~~~~~~~~~~~~~~~
-
-.. automethod:: pytorch_lightning.core.lightning.LightningModule.on_pretrain_routine_end
-    :noindex:
-
 on_test_batch_start
 ~~~~~~~~~~~~~~~~~~~
 

pytorch_lightning/core/hooks.py

@@ -68,6 +68,10 @@ def on_pretrain_routine_start(self) -> None:
         - pretrain_routine start
         - pretrain_routine end
         - training_start
+
+        .. deprecated:: v1.6
+            :meth:`on_pretrain_routine_start` has been deprecated in v1.6 and will be removed in v1.8.
+            Use ``on_fit_start`` instead.
         """
 
     def on_pretrain_routine_end(self) -> None:
@@ -77,6 +81,10 @@ def on_pretrain_routine_end(self) -> None:
         - pretrain_routine start
         - pretrain_routine end
         - training_start
+
+        .. deprecated:: v1.6
+            :meth:`on_pretrain_routine_end` has been deprecated in v1.6 and will be removed in v1.8.
+            Use ``on_fit_start`` instead.
         """
 
     def on_train_batch_start(self, batch: Any, batch_idx: int, unused: int = 0) -> Optional[int]:

pytorch_lightning/trainer/configuration_validator.py

@@ -60,6 +60,8 @@ def verify_loop_configurations(trainer: "pl.Trainer") -> None:
     _check_on_epoch_start_end(model)
     # TODO: Delete CheckpointHooks off PrecisionPlugin in v1.8
     _check_precision_plugin_checkpoint_hooks(trainer)
+    # TODO: Delete on_pretrain_routine_start/end hooks in v1.8
+    _check_on_pretrain_routine(model)
 
 
 def __verify_train_val_loop_configuration(trainer: "pl.Trainer", model: "pl.LightningModule") -> None:
@@ -306,6 +308,16 @@ def _check_on_epoch_start_end(model: "pl.LightningModule") -> None:
             )
 
 
+def _check_on_pretrain_routine(model: "pl.LightningModule") -> None:
+    hooks = (("on_pretrain_routine_start", "on_fit_start"), ("on_pretrain_routine_end", "on_fit_start"))
+    for hook, alternative_hook in hooks:
+        if is_overridden(hook, model):
+            rank_zero_deprecation(
+                f"The `LightningModule.{hook}` hook was deprecated in v1.6 and"
+                f" will be removed in v1.8. Please use `LightningModule.{alternative_hook}` instead."
+            )
+
+
 def _check_dl_idx_in_on_train_batch_hooks(model: "pl.LightningModule") -> None:
     for hook in ("on_train_batch_start", "on_train_batch_end"):
         if is_param_in_hook_signature(getattr(model, hook), "dataloader_idx", explicit=True):

tests/deprecated_api/test_remove_1-8.py

@@ -451,6 +451,37 @@ def on_epoch_end(self, *args, **kwargs):
         trainer.fit(model)
 
 
+def test_v1_8_0_remove_on_pretrain_routine_start_end_lightning_module(tmpdir):
+    class CustomModel(BoringModel):
+        def on_pretrain_routine_start(self, *args, **kwargs):
+            print("foo")
+
+    model = CustomModel()
+    trainer = Trainer(
+        fast_dev_run=True,
+        default_root_dir=tmpdir,
+    )
+    with pytest.deprecated_call(
+        match="The `LightningModule.on_pretrain_routine_start` hook was deprecated in v1.6 and will be removed in v1.8"
+    ):
+        trainer.fit(model)
+
+    class CustomModel(BoringModel):
+        def on_pretrain_routine_end(self, *args, **kwargs):
+            print("foo")
+
+    trainer = Trainer(
+        fast_dev_run=True,
+        default_root_dir=tmpdir,
+    )
+
+    model = CustomModel()
+    with pytest.deprecated_call(
+        match="The `LightningModule.on_pretrain_routine_end` hook was deprecated in v1.6 and will be removed in v1.8"
+    ):
+        trainer.fit(model)
+
+
 def test_v1_8_0_rank_zero_imports():
 
     import warnings

tests/models/test_restore.py

@@ -28,7 +28,7 @@
 import tests.helpers.utils as tutils
 from pytorch_lightning import Callback, Trainer
 from pytorch_lightning.callbacks import ModelCheckpoint
-from pytorch_lightning.trainer.states import RunningStage, TrainerFn
+from pytorch_lightning.trainer.states import TrainerFn
 from tests.helpers import BoringModel
 from tests.helpers.datamodules import ClassifDataModule
 from tests.helpers.runif import RunIf
@@ -253,7 +253,7 @@ def test_correct_step_and_epoch(tmpdir):
     assert trainer.global_step == 0
 
     class TestModel(BoringModel):
-        def on_pretrain_routine_end(self) -> None:
+        def on_train_start(self) -> None:
             assert self.trainer.current_epoch == first_max_epochs
             # TODO(@carmocca): should not need `+1`
             assert self.trainer.global_step == first_max_epochs * train_batches + 1
@@ -610,26 +610,18 @@ def test_dp_resume(tmpdir):
     class CustomModel(CustomClassificationModelDP):
         def __init__(self):
             super().__init__()
-            self.on_pretrain_routine_end_called = False
+            self.on_train_start_called = False
 
-        # set the epoch start hook so we can predict before the model does the full training
-        def on_pretrain_routine_end(self):
+        def on_validation_start(self):
             assert self.trainer.current_epoch == real_global_epoch and self.trainer.current_epoch > 0
-
-            # if model and state loaded correctly, predictions will be good even though we
-            # haven't trained with the new loaded model
-            new_trainer.state.stage = RunningStage.VALIDATING
-
-            dataloader = dm.train_dataloader()
+            dataloader = dm.val_dataloader()
             tpipes.run_model_prediction(self.trainer.lightning_module, dataloader=dataloader)
-            self.on_pretrain_routine_end_called = True
 
     # new model
     model = CustomModel()
 
-    # fit new model which should load hpc weights
-    new_trainer.fit(model, datamodule=dm)
-    assert model.on_pretrain_routine_end_called
+    # validate new model which should load hpc weights
+    new_trainer.validate(model, datamodule=dm, ckpt_path=hpc_save_path)
 
     # test freeze on gpu
     model.freeze()

tests/trainer/test_data_loading.py

@@ -104,7 +104,7 @@ def __init__(self, num_workers):
     def train_dataloader(self):
         return DataLoader(RandomDataset(32, 64), num_workers=self.num_workers)
 
-    def on_pretrain_routine_start(self):
+    def on_fit_start(self):
         self._resout = StringIO()
         self.ctx = redirect_stderr(self._resout)
         self.ctx.__enter__()