Class CycleGANModel
Defined in tensorflow/contrib/gan/python/namedtuples.py
.
An CycleGANModel contains all the pieces needed for CycleGAN training.
The model model_x2y
generator F maps data set X to Y, while the model
model_y2x
generator G maps data set Y to X.
See https://arxiv.org/abs/1703.10593 for more details.
Args:
model_x2y
: AGANModel
namedtuple whose generator maps data set X to Y.model_y2x
: AGANModel
namedtuple whose generator maps data set Y to X.reconstructed_x
: ATensor
of reconstructed data X which is G(F(X)).reconstructed_y
: ATensor
of reconstructed data Y which is F(G(Y)).
__new__
__new__(
_cls,
model_x2y,
model_y2x,
reconstructed_x,
reconstructed_y
)
Create new instance of CycleGANModel(model_x2y, model_y2x, reconstructed_x, reconstructed_y)