Type NoisyLinearCosineDecay
Namespace tensorflow.keras.experimental
Parent LearningRateSchedule
Interfaces INoisyLinearCosineDecay
A LearningRateSchedule that uses a noisy linear cosine decay schedule.
Methods
Properties
Public static methods
NoisyLinearCosineDecay NewDyn(object initial_learning_rate, object decay_steps, ImplicitContainer<T> initial_variance, ImplicitContainer<T> variance_decay, ImplicitContainer<T> num_periods, ImplicitContainer<T> alpha, ImplicitContainer<T> beta, object name)
Applies noisy linear cosine decay to the learning rate. See [Bello et al., ICML2017] Neural Optimizer Search with RL.
https://arxiv.org/abs/1709.07417 For the idea of warm starts here controlled by `num_periods`,
see [Loshchilov & Hutter, ICLR2016] SGDR: Stochastic Gradient Descent
with Warm Restarts. https://arxiv.org/abs/1608.03983 Note that linear cosine decay is more aggressive than cosine decay and
larger initial learning rates can typically be used. When training a model, it is often recommended to lower the learning rate as
the training progresses. This schedule applies a noisy linear cosine decay
function to an optimizer step, given a provided initial learning rate.
It requires a `step` value to compute the decayed learning rate. You can
just pass a TensorFlow variable that you increment at each training step. The schedule a 1-arg callable that produces a decayed learning
rate when passed the current optimizer step. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
It is computed as:
where eps_t is 0-centered gaussian noise with variance
initial_variance / (1 + global_step) ** variance_decay Example usage:
You can pass this schedule directly into a
tf.keras.optimizers.Optimizer
as the learning rate. The learning rate schedule is also serializable and
deserializable using tf.keras.optimizers.schedules.serialize and
tf.keras.optimizers.schedules.deserialize.
Parameters
-
objectinitial_learning_rate - A scalar `float32` or `float64` Tensor or a Python number. The initial learning rate.
-
objectdecay_steps - A scalar `int32` or `int64` `Tensor` or a Python number. Number of steps to decay over.
-
ImplicitContainer<T>initial_variance - initial variance for the noise. See computation above.
-
ImplicitContainer<T>variance_decay - decay for the noise's variance. See computation above.
-
ImplicitContainer<T>num_periods - Number of periods in the cosine part of the decay. See computation above.
-
ImplicitContainer<T>alpha - See computation above.
-
ImplicitContainer<T>beta - See computation above.
-
objectname - String. Optional name of the operation. Defaults to 'NoisyLinearCosineDecay'.
Returns
-
NoisyLinearCosineDecay - A 1-arg callable learning rate schedule that takes the current optimizer step and outputs the decayed learning rate, a scalar `Tensor` of the same type as `initial_learning_rate`.
Show Example
def decayed_learning_rate(step):
step = min(step, decay_steps)
linear_decay = (decay_steps - step) / decay_steps)
cosine_decay = 0.5 * (
1 + cos(pi * 2 * num_periods * step / decay_steps))
decayed = (alpha + linear_decay + eps_t) * cosine_decay + beta
return initial_learning_rate * decayed