View source on GitHub
|
Creates a tf.Variable suitable for use as a global step counter.
orbit.utils.create_global_step() -> tf.Variable
Creating and managing a global step variable may be necessary for
AbstractTrainer subclasses that perform multiple parameter updates per
Controller "step", or use different optimizers on different steps.
In these cases, an optimizer.iterations property generally can't be used
directly, since it would correspond to parameter updates instead of iterations
in the Controller's training loop. Such use cases should simply call
step.assign_add(1) at the end of each step.
Returns | |
|---|---|
A non-trainable scalar tf.Variable of dtype tf.int64, with only the
first replica's value retained when synchronizing across replicas in
a distributed setting.
|