mxnet.optimizer¶
Optimizer API of MXNet.
Optimization methods¶
AdaDelta([rho, epsilon]) |
The AdaDelta optimizer. |
AdaGrad([eps]) |
AdaGrad optimizer. |
Adam([learning_rate, beta1, beta2, epsilon, …]) |
The Adam optimizer. |
Adamax([learning_rate, beta1, beta2]) |
The AdaMax optimizer. |
DCASGD([momentum, lamda]) |
The DCASGD optimizer. |
FTML([beta1, beta2, epsilon]) |
The FTML optimizer. |
Ftrl([lamda1, learning_rate, beta]) |
The Ftrl optimizer. |
LBSGD([momentum, multi_precision, …]) |
The Large Batch SGD optimizer with momentum and weight decay. |
NAG([momentum]) |
Nesterov accelerated gradient. |
Nadam([learning_rate, beta1, beta2, …]) |
The Nesterov Adam optimizer. |
Optimizer([rescale_grad, param_idx2name, …]) |
The base class inherited by all optimizers. |
RMSProp([learning_rate, gamma1, gamma2, …]) |
The RMSProp optimizer. |
SGD([momentum, lazy_update]) |
The SGD optimizer with momentum and weight decay. |
SGLD(**kwargs) |
Stochastic Gradient Riemannian Langevin Dynamics. |
Signum([learning_rate, momentum, wd_lh]) |
The Signum optimizer that takes the sign of gradient or momentum. |
Updater(optimizer) |
Updater for kvstore. |
Helper functions¶
create(name, \*\*kwargs) |
Instantiates an optimizer with a given name and kwargs. |
get_updater(optimizer) |
Returns a closure of the updater needed for kvstore. |
register(klass) |
Registers a new optimizer. |