Add Fabric internal hooks (#17759)

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: 2 people

src/lightning/fabric/fabric.py

@@ -212,14 +212,19 @@ def setup(
         # Update the _DeviceDtypeModuleMixin's device parameter
         module.to(self.device if move_to_device else next(module.parameters(), torch.tensor(0)).device)
 
-        optimizers = [_FabricOptimizer(optimizer=optimizer, strategy=self._strategy) for optimizer in optimizers]
+        optimizers = [
+            _FabricOptimizer(optimizer=optimizer, strategy=self._strategy, callbacks=self._callbacks)
+            for optimizer in optimizers
+        ]
 
         self._models_setup += 1
 
         if hasattr(original_module, "_fabric"):  # this is probably a LightningModule
             original_module._fabric = self  # type: ignore[assignment]
             original_module._fabric_optimizers = optimizers  # type: ignore[assignment]
 
+        self.call("on_after_setup", fabric=self, module=module)
+
         if optimizers:
             # join both types in a tuple for API convenience
             return (module, *optimizers)
@@ -276,7 +281,10 @@ def setup_optimizers(self, *optimizers: Optimizer) -> Union[_FabricOptimizer, Tu
         """
         self._validate_setup_optimizers(optimizers)
         optimizers = [self._strategy.setup_optimizer(optimizer) for optimizer in optimizers]
-        optimizers = [_FabricOptimizer(optimizer=optimizer, strategy=self._strategy) for optimizer in optimizers]
+        optimizers = [
+            _FabricOptimizer(optimizer=optimizer, strategy=self._strategy, callbacks=self._callbacks)
+            for optimizer in optimizers
+        ]
         return optimizers[0] if len(optimizers) == 1 else tuple(optimizers)
 
     def setup_dataloaders(

src/lightning/fabric/wrappers.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import inspect
-from typing import Any, Callable, Dict, Generator, Iterator, Mapping, Optional, overload, TypeVar, Union
+from typing import Any, Callable, Dict, Generator, Iterator, List, Mapping, Optional, overload, TypeVar, Union
 
 import torch
 from lightning_utilities import WarningCache
@@ -38,7 +38,7 @@
 
 
 class _FabricOptimizer:
-    def __init__(self, optimizer: Optimizer, strategy: Strategy) -> None:
+    def __init__(self, optimizer: Optimizer, strategy: Strategy, callbacks: Optional[List[Callable]] = None) -> None:
         """FabricOptimizer is a thin wrapper around the :class:`~torch.optim.Optimizer` that delegates the
         optimizer step calls to the strategy plugin.
 
@@ -54,6 +54,7 @@ def __init__(self, optimizer: Optimizer, strategy: Strategy) -> None:
         self.__class__ = type("Fabric" + optimizer.__class__.__name__, (self.__class__, optimizer.__class__), {})
         self._optimizer = optimizer
         self._strategy = strategy
+        self._callbacks = callbacks or []
 
     @property
     def optimizer(self) -> Optimizer:
@@ -69,10 +70,15 @@ def step(self, closure: Optional[Callable] = None) -> Any:
             optimizer = self._strategy.model
         else:
             optimizer = self.optimizer
-        return self._strategy.optimizer_step(
+        output = self._strategy.optimizer_step(
             optimizer,
             **kwargs,
         )
+        for callback in self._callbacks:
+            hook = getattr(callback, "on_after_optimizer_step", None)
+            if callable(hook):
+                hook(strategy=self._strategy, optimizer=optimizer)
+        return output
 
 
 class _FabricModule(_DeviceDtypeModuleMixin):

tests/tests_fabric/test_fabric.py

@@ -753,6 +753,29 @@ def test_call():
     assert not callback1.mock_calls
 
 
+def test_special_callbacks():
+    """Tests special callbacks that have hooks for internal Fabric events."""
+
+    class SpecialCallback:
+        def on_after_optimizer_step(self, strategy, optimizer):
+            pass
+
+        def on_after_setup(self, fabric, module):
+            pass
+
+    callback = Mock(wraps=SpecialCallback())
+    fabric = Fabric(accelerator="cpu", callbacks=[callback])
+
+    model = torch.nn.Linear(2, 2)
+    optimizer = torch.optim.SGD(model.parameters(), lr=0.1)
+    fabric_model, fabric_optimizer = fabric.setup(model, optimizer)
+    callback.on_after_setup.assert_called_once_with(fabric=fabric, module=fabric_model)
+
+    model(torch.randn(2, 2)).sum().backward()
+    fabric_optimizer.step()
+    callback.on_after_optimizer_step.assert_called_once_with(strategy=fabric._strategy, optimizer=optimizer)
+
+
 def test_loggers_input():
     """Test the various ways in which loggers can be registered with Fabric."""
     logger0 = Mock()