TensorFlow_Tut_3_Object_Detection_Walk-through - 3_Train.py

[Setareh123]

Hi; I have followed the steps from the beginning but in running 3_train.py I get several errors! of course, by commenting # tf.app.run() the code is running successfully but this is not the deal!

The errors are lited below:

C:\Users\Setareh\AppData\Local\Programs\Python\Python36\python.exe C:/Users/Setareh/Documents/TensorFlow-examples/TensorFlow_Tut_3_Object_Detection_Walk-through/3_train.py starting program . . . Traceback (most recent call last): File "C:/Users/Setareh/Documents/TensorFlow-examples/TensorFlow_Tut_3_Object_Detection_Walk-through/3_train.py", line 153, in tf.app.run() File "C:\Users\Setareh\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\platform\app.py", line 126, in run _sys.exit(main(argv)) File "C:/Users/Setareh/Documents/TensorFlow-examples/TensorFlow_Tut_3_Object_Detection_Walk-through/3_train.py", line 115, in main CLONE_ON_CPU, ps_tasks, worker_job_name, is_chief, TRAINING_DATA_DIR) File "C:\TensorFlow\models\research\object_detection\trainer.py", line 240, in train detection_model = create_model_fn() File "C:\TensorFlow\models\research\object_detection\builders\model_builder.py", line 98, in build add_background_class) File "C:\TensorFlow\models\research\object_detection\builders\model_builder.py", line 166, in _build_ssd_model is_training=is_training) File "C:\TensorFlow\models\research\object_detection\builders\model_builder.py", line 138, in _build_ssd_feature_extractor override_base_feature_extractor_hyperparams) File "C:\TensorFlow\models\research\object_detection\models\ssd_inception_v2_feature_extractor.py", line 68, in init raise ValueError('SSD Inception V2 feature extractor always uses' ValueError: SSD Inception V2 feature extractor always usesscope returned by conv_hyperparams_fn for both the base feature extractor and the additional layers added since there is no arg_scope defined for the base feature extractor.

Process finished with exit code 1

[MicrocontrollersAndMore]

Did you try the fix from this issue:

https://github.com/MicrocontrollersAndMore/TensorFlow_Tut_3_Object_Detection_Walk-through/issues/3

Due to some change Google has made recently which I haven't yet had the time to properly work out, you have to comment out some lines in ssd_inception_v2_feature_extractor.py. I also included a note to this effect in the video comments. I'm going to close this at this time since it's the same error and I suspect this will resolve the concern.

[AekkaratKongsuk]

I have followed the steps from the beginning but in running 3_train.py I get several errors! How to fix ?

[sellinge]

In line 68 i changed the dataset_util to dataset_builder train.py runs now def get_next(config): return dataset_builder.make_initializable_iterator(dataset_builder.build(config)).get_next()

[AekkaratKongsuk]

How to fix ? please...

[roshankumarbhuyan]

This Commit solves the Issue