Remove deprecated max_steps=None (#13591)

* Remove max_steps=None

* Update changelog

* Update docs

* Unused import

Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>

docs/source-pytorch/common/trainer.rst

@@ -954,7 +954,7 @@ Training will stop if max_steps or max_epochs have reached (earliest).
 .. testcode::
 
     # Default (disabled)
-    trainer = Trainer(max_steps=None)
+    trainer = Trainer(max_steps=-1)
 
     # Stop after 100 steps
     trainer = Trainer(max_steps=100)

src/pytorch_lightning/CHANGELOG.md

@@ -264,6 +264,9 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Removed the deprecated `test_transforms` argument from the `LightningDataModule` constructor ([#12773](https://github.com/PyTorchLightning/pytorch-lightning/pull/12773))
 
 
+- Removed deprecated `Trainer(max_steps=None)` ([#13591](https://github.com/Lightning-AI/lightning/pull/13591))
+
+
 - Removed deprecated `dataloader_idx` argument from `on_train_batch_start/end` hooks `Callback` and `LightningModule` ([#12769](https://github.com/PyTorchLightning/pytorch-lightning/pull/12769), [#12977](https://github.com/PyTorchLightning/pytorch-lightning/pull/12977))
 
 

src/pytorch_lightning/loops/epoch/training_epoch_loop.py

@@ -48,13 +48,7 @@ class TrainingEpochLoop(loops.Loop[_OUTPUTS_TYPE]):
 
     def __init__(self, min_steps: Optional[int] = None, max_steps: int = -1) -> None:
         super().__init__()
-        if max_steps is None:
-            rank_zero_deprecation(
-                "Setting `max_steps = None` is deprecated in v1.5 and will no longer be supported in v1.7."
-                " Use `max_steps = -1` instead."
-            )
-            max_steps = -1
-        elif max_steps < -1:
+        if max_steps < -1:
             raise MisconfigurationException(
                 f"`max_steps` must be a non-negative integer or -1 (infinite steps). You passed in {max_steps}."
             )

src/pytorch_lightning/loops/fit_loop.py

@@ -33,7 +33,7 @@
     InterBatchParallelDataFetcher,
 )
 from pytorch_lightning.utilities.model_helpers import is_overridden
-from pytorch_lightning.utilities.rank_zero import rank_zero_deprecation, rank_zero_warn
+from pytorch_lightning.utilities.rank_zero import rank_zero_warn
 from pytorch_lightning.utilities.signature_utils import is_param_in_hook_signature
 
 log = logging.getLogger(__name__)
@@ -104,13 +104,7 @@ def max_steps(self) -> int:
     def max_steps(self, value: int) -> None:
         """Sets the maximum number of steps (forwards to epoch_loop)"""
         # TODO(@awaelchli): This setter is required by debugging connector (fast dev run), should be avoided
-        if value is None:
-            rank_zero_deprecation(
-                "Setting `max_steps = None` is deprecated in v1.5 and will no longer be supported in v1.7."
-                " Use `max_steps = -1` instead."
-            )
-            value = -1
-        elif value < -1:
+        if value < -1:
             raise MisconfigurationException(
                 f"`max_steps` must be a non-negative integer or -1 (infinite steps). You passed in {value}."
             )

tests/tests_pytorch/deprecated_api/test_remove_1-7.py

@@ -17,19 +17,9 @@
 import pytest
 import torch
 
-from pytorch_lightning import Trainer
 from pytorch_lightning.strategies import SingleDeviceStrategy
 
 
-def test_v1_7_0_deprecated_max_steps_none(tmpdir):
-    with pytest.deprecated_call(match="`max_steps = None` is deprecated in v1.5"):
-        _ = Trainer(max_steps=None)
-
-    trainer = Trainer()
-    with pytest.deprecated_call(match="`max_steps = None` is deprecated in v1.5"):
-        trainer.fit_loop.max_steps = None
-
-
 def test_v1_7_0_post_dispatch_hook():
     class CustomPlugin(SingleDeviceStrategy):
         def post_dispatch(self, trainer):