vanhuyz / CycleGAN-TensorFlow

An implementation of CycleGan using TensorFlow
MIT License
1.19k stars 436 forks source link

Cannot convert a tensor of type float32 to an input of type float32_ref #93

Open doantientai opened 6 years ago

doantientai commented 6 years ago

Thank you @vanhuyz for your great implementation. I have finished training with my own data and so excited to run the inference. However, after exporting graph, I run inference.py and receiving the follow error. I suggest the was something wrong when the model is exported, wasn't it?

Thank you in advance!

Traceback (most recent call last):
  File "/home/tai/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/framework/importer.py", line 667, in import_graph_def
    op._add_input(source_tensor, dtype=input_type)
  File "/home/tai/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1898, in _add_input
    (tensor.dtype.name, dtype.name))
TypeError: Cannot convert a tensor of type float32 to an input of type float32_ref

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/media/tai/6TB/Projects/CycleGAN-TensorFlow/inference.py", line 49, in <module>
    tf.app.run()
  File "/home/tai/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/platform/app.py", line 126, in run
    _sys.exit(main(argv))
  File "/media/tai/6TB/Projects/CycleGAN-TensorFlow/inference.py", line 46, in main
    inference()
  File "/media/tai/6TB/Projects/CycleGAN-TensorFlow/inference.py", line 38, in inference
    name='output')
  File "/home/tai/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/util/deprecation.py", line 432, in new_func
    return func(*args, **kwargs)
  File "/home/tai/anaconda3/envs/tf/lib/python3.5/site-packages/tensorflow/python/framework/importer.py", line 671, in import_graph_def
    node, 'Input tensor %r %s' % (input_name, te)))
ValueError: graph_def is invalid at node 'G_7/c7s1_32/batch_norm/BatchNorm/cond_1/AssignMovingAvg/Switch': Input tensor 'G/c7s1_32/batch_norm/BatchNorm/moving_mean:0' Cannot convert a tensor of type float32 to an input of type float32_ref.

Process finished with exit code 1