Add experimental.SyncBatchNormalization to Prune Registry

PiperOrigin-RevId: 423694994

Author: rino20

tensorflow_model_optimization/python/core/sparsity/keras/prune_registry.py

@@ -98,6 +98,7 @@ class PruneRegistry(object):
           '_query_dense.kernel', '_key_dense.kernel', '_value_dense.kernel',
           '_output_dense.kernel'
       ],
+      layers.experimental.SyncBatchNormalization: [],
       layers.experimental.preprocessing.Rescaling.__class__: [],
       TensorFlowOpLayer: [],
       layers_compat_v1.BatchNormalization: [],

tensorflow_model_optimization/python/core/sparsity/keras/prune_registry_test.py

@@ -72,7 +72,8 @@ class PruneRegistryTest(tf.test.TestCase, parameterized.TestCase):
       layers.Conv2D(10, 5),
       layers.Dropout(0.5),
       # Supports specific layers from experimental or compat_v1.
-      tf.keras.layers.experimental.preprocessing.Rescaling,
+      layers.experimental.SyncBatchNormalization(),
+      layers.experimental.preprocessing.Rescaling,
       tf.compat.v1.keras.layers.BatchNormalization(),
       # Supports Keras RNN Layers with prunable cells.
       layers.LSTM(10),