search

DrSleep / tensorflow-deeplab-resnet

DeepLab-ResNet rebuilt in TensorFlow
MIT License
1.25k stars 429 forks source link

error #53

Closed weigq closed 7 years ago

weigq commented 7 years ago

I have some thing wrong :

Traceback (most recent call last):
  File "train.py", line 187, in <module>
    main()
  File "train.py", line 106, in main
    image_batch, label_batch = reader.dequeue(args.batch_size)
  File "/data6/weigq/deeplab/tensorflow-deeplab-resnet/deeplab_resnet/image_reader.py", line 121, in dequeue
    image_batch, label_batch = tf.train.batch([self.image, self.label], num_elements)
  File "/home/liusen/local/anaconda2/envs/tensorflow/lib/python2.7/site-packages/tensorflow/python/training/input.py", line 586, in batch
    capacity=capacity, dtypes=types, shapes=shapes, shared_name=shared_name)
  File "/home/liusen/local/anaconda2/envs/tensorflow/lib/python2.7/site-packages/tensorflow/python/ops/data_flow_ops.py", line 638, in __init__
    shapes = _as_shape_list(shapes, dtypes)
  File "/home/liusen/local/anaconda2/envs/tensorflow/lib/python2.7/site-packages/tensorflow/python/ops/data_flow_ops.py", line 72, in _as_shape_list
    raise ValueError("All shapes must be fully defined: %s" % shapes)
ValueError: All shapes must be fully defined: [TensorShape([Dimension(321), Dimension(321), Dimension(3)]), TensorShape([Dimension(321), Dimension(321), Dimension(None)])]

what is the matter