[Feature] Add additional callbacks

docs/tutorials/qml/ml_tools/callbacks.md

@@ -146,6 +146,85 @@ config = TrainConfig(
 )
 ```
 
+### 1.9. `LRSchedulerStepDecay`
+
+Reduces the learning rate by a factor at regular intervals.
+
+```python exec="on" source="material-block" html="1"
+from qadence.ml_tools import TrainConfig
+from qadence.ml_tools.callbacks import LRSchedulerStepDecay
+
+lr_step_decay = LRSchedulerStepDecay(on="train_epoch_end", called_every=100, gamma=0.5)
+
+config = TrainConfig(
+    max_iter=10000,
+    callbacks=[lr_step_decay]
+)
+```
+
+### 1.10. `LRSchedulerCyclic`
+
+Applies a cyclic learning rate schedule during training.
+
+```python exec="on" source="material-block" html="1"
+from qadence.ml_tools import TrainConfig
+from qadence.ml_tools.callbacks import LRSchedulerCyclic
+
+lr_cyclic = LRSchedulerCyclic(on="train_batch_end", called_every=1, base_lr=0.001, max_lr=0.01, step_size=2000)
+
+config = TrainConfig(
+    max_iter=10000,
+    callbacks=[lr_cyclic]
+)
+```
+
+### 1.11. `LRSchedulerCosineAnnealing`
+
+Applies cosine annealing to the learning rate during training.
+
+```python exec="on" source="material-block" html="1"
+from qadence.ml_tools import TrainConfig
+from qadence.ml_tools.callbacks import LRSchedulerCosineAnnealing
+
+lr_cosine = LRSchedulerCosineAnnealing(on="train_batch_end", called_every=1, t_max=5000, min_lr=1e-6)
+
+config = TrainConfig(
+    max_iter=10000,
+    callbacks=[lr_cosine]
+)
+```
+
+### 1.12. `EarlyStopping`
+
+Stops training when a monitored metric has not improved for a specified number of epochs.
+
+```python exec="on" source="material-block" html="1"
+from qadence.ml_tools import TrainConfig
+from qadence.ml_tools.callbacks import EarlyStopping
+
+early_stopping = EarlyStopping(on="val_epoch_end", called_every=1, monitor="val_loss", patience=5, mode="min")
+
+config = TrainConfig(
+    max_iter=10000,
+    callbacks=[early_stopping]
+)
+```
+
+### 1.13. `GradientMonitoring`
+
+Logs gradient statistics (e.g., mean, standard deviation, max) during training.
+
+```python exec="on" source="material-block" html="1"
+from qadence.ml_tools import TrainConfig
+from qadence.ml_tools.callbacks import GradientMonitoring
+
+gradient_monitoring = GradientMonitoring(on="train_batch_end", called_every=10)
+
+config = TrainConfig(
+    max_iter=10000,
+    callbacks=[gradient_monitoring]
+)
+```
 
 ## 2. Custom Callbacks
 

docs/tutorials/qml/ml_tools/trainer.md

@@ -531,7 +531,7 @@ def train(
             writer.print_metrics(OptimizeResult(iteration, model, optimizer, loss, metrics))
 
         if iteration % config.write_every == 0:
-            writer.write(OptimizeResult(iteration, model, optimizer, loss, metrics))
+            writer.write(iteration, metrics)
 
         if config.log_folder:
             if iteration % config.checkpoint_every == 0:
@@ -540,7 +540,7 @@ def train(
     # Final writing and checkpointing
     if config.log_folder:
         write_checkpoint(config.log_folder, model, optimizer, iteration)
-    writer.write(OptimizeResult(iteration, model, optimizer, loss, metrics))
+    writer.write(iteration,metrics)
     writer.close()
 
     return model, optimizer

qadence/ml_tools/callbacks/__init__.py

@@ -2,9 +2,14 @@
 
 from .callback import (
     Callback,
+    EarlyStopping,
+    GradientMonitoring,
     LoadCheckpoint,
     LogHyperparameters,
     LogModelTracker,
+    LRSchedulerCosineAnnealing,
+    LRSchedulerCyclic,
+    LRSchedulerStepDecay,
     PlotMetrics,
     PrintMetrics,
     SaveBestCheckpoint,
@@ -26,5 +31,10 @@
     "SaveBestCheckpoint",
     "SaveCheckpoint",
     "WriteMetrics",
+    "GradientMonitoring",
+    "LRSchedulerStepDecay",
+    "LRSchedulerCyclic",
+    "LRSchedulerCosineAnnealing",
+    "EarlyStopping",
     "get_writer",
 ]