Update dependency on official/vision/image_classification.

PiperOrigin-RevId: 420895912

research/object_detection/g3doc/using_your_own_dataset.md

@@ -3,7 +3,7 @@
 [TOC]
 
 To use your own dataset in TensorFlow Object Detection API, you must convert it
-into the [TFRecord file format](https://www.tensorflow.org/tutorials/load_data/tfrecord).
+into the [TFRecord file format](https://www.tensorflow.org/api_guides/python/python_io#tfrecords_format_details).
 This document outlines how to write a script to generate the TFRecord file.
 
 ## Label Maps

research/object_detection/models/ssd_efficientnet_bifpn_feature_extractor.py

@@ -31,6 +31,9 @@ from object_detection.utils import shape_utils
 from object_detection.utils import tf_version
 # pylint: disable=g-import-not-at-top
 if tf_version.is_tf2():
+  try:
+    from official.legacy.image_classification.efficientnet import efficientnet_model
+  except ModuleNotFoundError:
     from official.vision.image_classification.efficientnet import efficientnet_model
 
 _EFFICIENTNET_LEVEL_ENDPOINTS = {