A strange thing

When I trained with tensorflow object detection API, I found a strange thing. If there is a pre_training model in the configuration file, I can continue training after interruption; if there is no pre_training model in the configuration file, I cannot train after interruption and report an error:Object was never used (type <class 'tensorflow.python.framework.ops.Tensor'>): <tf.Tensor 'init_ops/report_uninitialized_variables/boolean_mask/GatherV2:0' shape=(?,) dtype=string> If you want to mark it as used call its "mark_used()" method. It was originally created here: File "train.py", line 195, in tf.app.run() File "C:\Users\dw\Anaconda3\lib\site-packages\tensorflow\python\platform\app.py", line 125, in run _sys.exit(main(argv)) File "C:\Users\dw\Anaconda3\lib\site-packages\tensorflow\python\util\deprecation.py", line 324, in new_func return func(*args, **kwargs) File "train.py", line 191, in main graph_hook_fn=graph_rewriter_fn) File "D:\tf_train\models\research\object_detection\legacy\trainer.py", line 416, in train saver=saver) File "C:\Users\dw\Anaconda3\lib\site-packages\tensorflow\contrib\slim\python\slim\learning.py", line 791, in train should_retry = True File "C:\Users\dw\Anaconda3\lib\site-packages\tensorflow\python\util\tf_should_use.py", line 193, in wrapped return _add_should_use_warning(fn(*args, **kwargs)) ==================================
System information:
tensorflow-gpu=1.13.1
train.py
ssd_inception_v2_coco.config