ValueError: 'image' (shape (?, 26, 26, 256)) must be three-dimensional.

[arjunkedlaya]

Traceback (most recent call last):
File "train.py", line 113, in
tf.app.run()
File "/home/tendulkar/anaconda3/envs/deeplearn/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 48, in run
_sys.exit(main(_sys.argv[:1] + flags_passthrough))
File "train.py", line 31, in main
image_placeholder, groundtruth_text_placeholder, FLAGS.single_seq)
File "/home/tendulkar/Desktop/aon/AON-master/model_aon.py", line 313, in inference
aon_core_output_dict = _arbitrary_orientation_network(base_features)
File "/home/tendulkar/Desktop/aon/AON-master/model_aon.py", line 174, in _arbitrary_orientation_network
featute_vertical = get_feature_sequence(inputs=tf.image.rot90(inputs), reuse=True)
File "/home/tendulkar/anaconda3/envs/deeplearn/lib/python3.6/site-packages/tensorflow/python/ops/image_ops_impl.py", line 322, in rot90
_Check3DImage(image, require_static=False), image)
File "/home/tendulkar/anaconda3/envs/deeplearn/lib/python3.6/site-packages/tensorflow/python/ops/image_ops_impl.py", line 135, in _Check3DImage
image.shape)
ValueError: 'image' (shape (?, 26, 26, 256)) must be three-dimensional.

I have already created train.tfrecord using the answer given for issue #4 .Encountering this issue when i try to run

Once the tfrecord file has been created, run the train.py file with the following arguments:
python3 train.py --tfrecord_file_path path_to_tfrecord_file