Skip to content

Conversation

@Uha661
Copy link

@Uha661 Uha661 commented Sep 3, 2019

No description provided.

Copy link
Author

@Uha661 Uha661 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

StopIteration Traceback (most recent call last)
in ()
9 generator_network, discriminator_network, Gs_network = pickle.load(f)
10
---> 11 generator = Generator(Gs_network, batch_size=1, randomize_noise=False)
12
13

/content/stylegan-encoder/encoder/generator_model.py in init(self, model, batch_size, randomize_noise)
30 self.graph = tf.get_default_graph()
31
---> 32 self.dlatent_variable = next(v for v in tf.global_variables() if 'learnable_dlatents' in v.name)
33 self.set_dlatents(self.initial_dlatents)
34

StopIteration:

@progmars
Copy link

The fix leads to error:

 File "encode_images.py", line 84, in <module>
    main()
  File "encode_images.py", line 57, in main
    generator = Generator(Gs_network, args.batch_size, randomize_noise=args.randomize_noise)
  File "D:\GAN\stylegan\encoder\generator_model.py", line 35, in __init__
    self.set_dlatents(self.initial_dlatents)
  File "D:\GAN\stylegan\encoder\generator_model.py", line 46, in set_dlatents
    self.sess.run(tf.assign(self.dlatent_variable, dlatents))
  File "D:\GAN\Anaconda\lib\site-packages\tensorflow\python\ops\state_ops.py", line 224, in assign
    if ref.dtype._is_ref_dtype:
AttributeError: 'generator' object has no attribute 'dtype'

And it seems to be unnecessary at all. I found that StopIteration error occurs only if you use Nvlabs dmlib instead of this repository dmlib because this one has modifications to pass custom arguments to the run method.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants