Apply to speak at TensorFlow World. Deadline April 23rd. Propose talk

tf.contrib.gan.StarGANModel

Class StarGANModel

Defined in tensorflow/contrib/gan/python/namedtuples.py.

A StarGANModel contains all the pieces needed for StarGAN training.

Args:

  • input_data: The real images that need to be transferred by the generator.
  • input_data_domain_label: The real domain labels associated with the real images.
  • generated_data: The generated images produced by the generator. It has the same shape as the input_data.
  • generated_data_domain_target: The target domain that the generated images belong to. It has the same shape as the input_data_domain_label.
  • reconstructed_data: The reconstructed images produced by the G(enerator). reconstructed_data = G(G(input_data, generated_data_domain_target), input_data_domain_label).
  • discriminator_input_data_source: The discriminator's output for predicting the source (real/generated) of input_data.
  • discriminator_generated_data_source: The discriminator's output for predicting the source (real/generated) of generated_data.
  • discriminator_input_data_domain_predication: The discriminator's output for predicting the domain_label for the input_data.
  • discriminator_generated_data_domain_predication: The discriminatorr's output for predicting the domain_target for the generated_data.
  • generator_variables: A list of all generator variables.
  • generator_scope: Variable scope all generator variables live in.
  • generator_fn: The generator function.
  • discriminator_variables: A list of all discriminator variables.
  • discriminator_scope: Variable scope all discriminator variables live in.
  • discriminator_fn: The discriminator function.

__new__

__new__(
    _cls,
    input_data,
    input_data_domain_label,
    generated_data,
    generated_data_domain_target,
    reconstructed_data,
    discriminator_input_data_source_predication,
    discriminator_generated_data_source_predication,
    discriminator_input_data_domain_predication,
    discriminator_generated_data_domain_predication,
    generator_variables,
    generator_scope,
    generator_fn,
    discriminator_variables,
    discriminator_scope,
    discriminator_fn
)

Create new instance of StarGANModel(input_data, input_data_domain_label, generated_data, generated_data_domain_target, reconstructed_data, discriminator_input_data_source_predication, discriminator_generated_data_source_predication, discriminator_input_data_domain_predication, discriminator_generated_data_domain_predication, generator_variables, generator_scope, generator_fn, discriminator_variables, discriminator_scope, discriminator_fn)

Properties

input_data

input_data_domain_label

generated_data

generated_data_domain_target

reconstructed_data

discriminator_input_data_source_predication

discriminator_generated_data_source_predication

discriminator_input_data_domain_predication

discriminator_generated_data_domain_predication

generator_variables

generator_scope

generator_fn

discriminator_variables

discriminator_scope

discriminator_fn