tf.contrib.gan.gan_train(
train_ops,
logdir,
get_hooks_fn=get_sequential_train_hooks(),
master='',
is_chief=True,
scaffold=None,
hooks=None,
chief_only_hooks=None,
save_checkpoint_secs=600,
save_summaries_steps=100,
config=None
)
Defined in tensorflow/contrib/gan/python/train.py
.
A wrapper around contrib.training.train
that uses GAN hooks.
Args:
train_ops
: A GANTrainOps named tuple.logdir
: The directory where the graph and checkpoints are saved.get_hooks_fn
: A function that takes a GANTrainOps tuple and returns a list of hooks.master
: The URL of the master.is_chief
: Specifies whether or not the training is being run by the primary replica during replica training.scaffold
: An tf.train.Scaffold instance.hooks
: List oftf.train.SessionRunHook
callbacks which are run inside the training loop.chief_only_hooks
: List oftf.train.SessionRunHook
instances which are run inside the training loop for the chief trainer only.save_checkpoint_secs
: The frequency, in seconds, that a checkpoint is saved using a default checkpoint saver. Ifsave_checkpoint_secs
is set toNone
, then the default checkpoint saver isn't used.save_summaries_steps
: The frequency, in number of global steps, that the summaries are written to disk using a default summary saver. Ifsave_summaries_steps
is set toNone
, then the default summary saver isn't used.config
: An instance oftf.ConfigProto
.
Returns:
Output of the call to training.train
.