Update TFLite concrete function conversion codes to have the associated

trackable object.

Providing trackable objects is now recommended for the users. This makes the
concrete function conversion API be based on the new SavedModel importer, which
will enable new TensorFlow Lite features including variable support, resources
and variant tensor, and signature concept.

PiperOrigin-RevId: 410120653

Author: Googler

fix_conversion_issues_ptq_tf2.ipynb

@@ -451,7 +451,7 @@
         "imported = tf.saved_model.load(save_path)\n",
         "concrete_func = imported.signatures[\"serving_default\"]\n",
         "concrete_func.inputs[0].set_shape([1, 224, 224, 3])\n",
-        "converter = tf.lite.TFLiteConverter.from_concrete_functions([concrete_func])"
+        "converter = tf.lite.TFLiteConverter.from_concrete_functions([concrete_func], imported)"
       ]
     },
     {
@@ -576,7 +576,7 @@
         "imported = tf.saved_model.load(\"imagenet_inception_v2_classification_4\")\n",
         "concrete_func = imported.__call__.get_concrete_function(\n",
         "          tf.TensorSpec([1, 224, 224, 3]))\n",
-        "converter = tf.lite.TFLiteConverter.from_concrete_functions([concrete_func])"
+        "converter = tf.lite.TFLiteConverter.from_concrete_functions([concrete_func], imported)"
       ]
     },
     {