| learning_rate_schedule_inverse_time_decay {keras} | R Documentation |
A LearningRateSchedule that uses an inverse time decay schedule
Description
A LearningRateSchedule that uses an inverse time decay schedule
Usage
learning_rate_schedule_inverse_time_decay(
initial_learning_rate,
decay_steps,
decay_rate,
staircase = FALSE,
...,
name = NULL
)
Arguments
initial_learning_rate |
A scalar |
decay_steps |
A scalar |
decay_rate |
An R number. The decay rate. |
staircase |
Boolean. Whether to apply decay in a discrete staircase, as opposed to continuous, fashion. |
... |
For backwards and forwards compatibility |
name |
String. Optional name of the operation. Defaults to 'InverseTimeDecay'. |
Details
When training a model, it is often useful to lower the learning rate as
the training progresses. This schedule applies the inverse 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 is 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:
decayed_learning_rate <- function(step) {
initial_learning_rate / (1 + decay_rate * step / decay_step)
}
or, if staircase is TRUE, as:
decayed_learning_rate function(step) {
initial_learning_rate / (1 + decay_rate * floor(step / decay_step))
}
You can pass this schedule directly into a keras Optimizer
as the learning_rate.
Example: Fit a Keras model when decaying 1/t with a rate of 0.5:
...
initial_learning_rate <- 0.1
decay_steps <- 1.0
decay_rate <- 0.5
learning_rate_fn <- learning_rate_schedule_inverse_time_decay(
initial_learning_rate, decay_steps, decay_rate)
model %>%
compile(optimizer = optimizer_sgd(learning_rate = learning_rate_fn),
loss = 'sparse_categorical_crossentropy',
metrics = 'accuracy')
model %>% fit(data, labels, epochs = 5)