Addressed PR comments

alibi_detect/cd/tensorflow/preprocess.py

@@ -4,7 +4,7 @@
 import tensorflow as tf
 
 from alibi_detect.utils.tensorflow.prediction import (
-    predict_batch, predict_batch_transformer, get_named_arg
+    predict_batch, predict_batch_transformer, get_call_arg_mapping
 )
 from tensorflow.keras.layers import Dense, Flatten, Input, Lambda
 from tensorflow.keras.models import Model
@@ -37,7 +37,7 @@ def __init__(
 
     def call(self, x: Union[np.ndarray, tf.Tensor, Dict[str, tf.Tensor]]) -> tf.Tensor:
         if not isinstance(x, (np.ndarray, tf.Tensor)):
-            x = get_named_arg(self.input_layer, x)
+            x = get_call_arg_mapping(self.input_layer, x)
             x = self.input_layer(**x)
         else:
             x = self.input_layer(x)
@@ -68,7 +68,7 @@ def __init__(
 
     def call(self, x: Union[np.ndarray, tf.Tensor, Dict[str, tf.Tensor]]) -> tf.Tensor:
         if not isinstance(x, (np.ndarray, tf.Tensor)):
-            x = get_named_arg(self.encoder, x)
+            x = get_call_arg_mapping(self.encoder, x)
             return self.encoder(**x)
         else:
             return self.encoder(x)

alibi_detect/saving/_tensorflow/tests/test_saving_tf.py

@@ -15,9 +15,6 @@
 backend = param_fixture("backend", ['tensorflow'])
 
 
-# Note: The full save/load functionality of optimizers (inc. validation) is tested in test_save_classifierdrift.
-@pytest.mark.skipif(version.parse(tf.__version__) < version.parse('2.16.0'),
-                    reason="Skipping since tensorflow < 2.16.0")
 def test_load_optimizer_object_tf2pt11(backend):
     """
     Test the _load_optimizer_config with a tensorflow optimizer config. Only run if tensorflow>=2.16.

alibi_detect/utils/tensorflow/prediction.py

@@ -7,10 +7,14 @@
 from alibi_detect.utils.prediction import tokenize_transformer
 
 
-def get_named_arg(model: tf.keras.Model, x: Any) -> Dict[str, Any]:
-    """ Extract argument names from the model call function
-    because keras3 does not accept other types of input
-    as a positional argument.
+def get_call_arg_mapping(model: tf.keras.Model, x: Any) -> Dict[str, Any]:
+    """ Generates a dictionary mapping the first argument name of the
+    `call` method of a Keras model to the provided input value.
+
+    This function is particularly useful when working with Keras 3,
+    which enforces stricter input handling and requires named arguments
+    for certain operations. It extracts the argument names from the
+    `call` method of the provided model and maps the first argument to `x`.
 
     Parameters
     ----------
@@ -67,7 +71,7 @@ def predict_batch(
             x_batch = preprocess_fn(x_batch)
 
         if not isinstance(x_batch, (np.ndarray, tf.Tensor)):
-            x_batch = get_named_arg(model, x_batch)
+            x_batch = get_call_arg_mapping(model, x_batch)
             preds_tmp = model(**x_batch)
         else:
             preds_tmp = model(x_batch)

alibi_detect/utils/tests/test_saving_legacy.py

@@ -262,8 +262,6 @@ def test_save_load(select_detector):
 
 
 # Note: The full save/load functionality of optimizers (inc. validation) is tested in test_save_classifierdrift.
-@pytest.mark.skipif(version.parse(tf.__version__) < version.parse('2.16.0'),
-                    reason="Skipping since tensorflow < 2.16.0")
 @parametrize('legacy', [True, False])
 def test_load_optimizer_object_tf2pt11(legacy, backend):
     """