Adafactor
classtf.keras.optimizers.experimental.Adafactor(
learning_rate=0.001,
beta_2_decay=-0.8,
epsilon_1=1e-30,
epsilon_2=0.001,
clip_threshold=1.0,
relative_step=True,
weight_decay=None,
clipnorm=None,
clipvalue=None,
global_clipnorm=None,
use_ema=False,
ema_momentum=0.99,
ema_overwrite_frequency=None,
jit_compile=True,
name="Adafactor",
**kwargs
)
Optimizer that implements the Adafactor algorithm.
Adafactor is commonly used in NLP tasks, and has the advantage of taking less memory because it only saves partial information of previous gradients.
The default argument setup is based on the original paper (see reference). When gradients are of dimension > 2, Adafactor optimizer will delete the last 2 dimensions separately in its accumulator variables.
Arguments
tf.keras.optimizers.schedules.LearningRateSchedule
instance.
Defaults to 0.001.beta_2
.clipnorm
, clipvalue
and
global_clipnorm
.learning_rate
is a
constant and relative_step=True
, learning rate will be adjusted
based on current iterations. This is a default learning rate decay
in Adafactor.use_ema=True
. This is # noqa: E501
the momentum to use when computing the EMA of the model's weights:
new_average = ema_momentum * old_average + (1 - ema_momentum) *
current_variable_value
.use_ema=True
. Every ema_overwrite_frequency
steps of iterations, we
overwrite the model variable by its moving average. If None, the optimizer # noqa: E501
does not overwrite model variables in the middle of training, and you
need to explicitly overwrite the variables at the end of training
by calling optimizer.finalize_variable_values()
(which updates the model # noqa: E501
variables in-place). When using the built-in fit()
training loop, this
happens automatically after the last epoch, and you don't need to do
anything.Reference