Class SGD

Stochastic gradient descent and momentum optimizer.

Inherits From: Optimizer

Aliases: tf.optimizers.SGD

• Train and evaluate with Keras
• Distributed training with TensorFlow
• Eager execution
• Use a GPU
• Keras overview

• Custom training: walkthrough
• Multi-worker training with Keras

theta(t+1) = theta(t) - learning_rate * gradient
gradient is evaluated at theta(t).

or Computes (if nesterov = False):

v(t+1) = momentum * v(t) - learning_rate * gradient
theta(t+1) = theta(t) + v(t+1)
if `nesterov` is False, gradient is evaluated at theta(t).
if `nesterov` is True, gradient is evaluated at theta(t) + momentum * v(t),
  and the variables always store theta + m v instead of theta

Some of the args below are hyperparameters, where a hyperparameter is defined as a scalar Tensor, a regular Python value, or a callable (which will be evaluated when apply_gradients is called) returning a scalar Tensor or a Python value.

nesterov = True, See [Sutskever et al., 2013](
  http://jmlr.org/proceedings/papers/v28/sutskever13.pdf).

When eager execution is enabled, learning_rate can be a callable that takes no arguments and returns the actual value to use. This can be useful for changing these values across different invocations of optimizer functions.

View source

__init__(
    learning_rate=0.01,
    momentum=0.0,
    nesterov=False,
    name='SGD',
    **kwargs
)

Construct a new Stochastic Gradient Descent or Momentum optimizer.

• learning_rate: float hyperparameter >= 0. Learning rate.
• momentum: float hyperparameter >= 0 that accelerates SGD in the relevant direction and dampens oscillations.
• nesterov: boolean. Whether to apply Nesterov momentum.
• name: Optional name prefix for the operations created when applying gradients. Defaults to 'SGD'.
• **kwargs: keyword arguments. Allowed to be {clipnorm, clipvalue, lr, decay}. clipnorm is clip gradients by norm; clipvalue is clip gradients by value, decay is included for backward compatibility to allow time inverse decay of learning rate. lr is included for backward compatibility, recommended to use learning_rate instead.

Variable. The number of training steps this Optimizer has run.

Returns variables of this Optimizer based on the order created.

View source

add_slot(
    var,
    slot_name,
    initializer='zeros'
)

Add a new slot variable for var.

View source

add_weight(
    name,
    shape,
    dtype=None,
    initializer='zeros',
    trainable=None,
    synchronization=tf.VariableSynchronization.AUTO,
    aggregation=tf.compat.v1.VariableAggregation.NONE
)

View source

apply_gradients(
    grads_and_vars,
    name=None
)

Apply gradients to variables.

This is the second part of minimize(). It returns an Operation that applies gradients.

• grads_and_vars: List of (gradient, variable) pairs.
• name: Optional name for the returned operation. Default to the name passed to the Optimizer constructor.

An Operation that applies the specified gradients. The iterations will be automatically increased by 1.

• TypeError: If grads_and_vars is malformed.
• ValueError: If none of the variables have gradients.

View source

from_config(
    cls,
    config,
    custom_objects=None
)

Creates an optimizer from its config.

This method is the reverse of get_config, capable of instantiating the same optimizer from the config dictionary.

• config: A Python dictionary, typically the output of get_config.
• custom_objects: A Python dictionary mapping names to additional Python objects used to create this optimizer, such as a function used for a hyperparameter.

An optimizer instance.

View source

get_config()

Returns the config of the optimimizer.

An optimizer config is a Python dictionary (serializable) containing the configuration of an optimizer. The same optimizer can be reinstantiated later (without any saved state) from this configuration.

Python dictionary.

View source

get_gradients(
    loss,
    params
)

Returns gradients of loss with respect to params.

• loss: Loss tensor.
• params: List of variables.

List of gradient tensors.

• ValueError: In case any gradient cannot be computed (e.g. if gradient function not implemented).

View source

get_slot(
    var,
    slot_name
)

View source

get_slot_names()

A list of names for this optimizer's slots.

View source

get_updates(
    loss,
    params
)

View source

get_weights()

View source

minimize(
    loss,
    var_list,
    grad_loss=None,
    name=None
)

Minimize loss by updating var_list.

This method simply computes gradient using tf.GradientTape and calls apply_gradients(). If you want to process the gradient before applying then call tf.GradientTape and apply_gradients() explicitly instead of using this function.

• loss: A callable taking no arguments which returns the value to minimize.
• var_list: list or tuple of Variable objects to update to minimize loss, or a callable returning the list or tuple of Variable objects. Use callable when the variable list would otherwise be incomplete before minimize since the variables are created at the first time loss is called.
• grad_loss: Optional. A Tensor holding the gradient computed for loss.
• name: Optional name for the returned operation.

An Operation that updates the variables in var_list. The iterations will be automatically increased by 1.

• ValueError: If some of the variables are not Variable objects.

View source

set_weights(weights)

View source

variables()

Returns variables of this Optimizer based on the order created.

• tf.compat.v1.keras.optimizers.SGD
• tf.compat.v2.keras.optimizers.SGD
• tf.compat.v2.optimizers.SGD