Optimizers¶
chainer.optimizers.AdaDelta |
Zeiler’s ADADELTA. |
chainer.optimizers.AdaGrad |
AdaGrad optimizer. |
chainer.optimizers.Adam |
Adam optimizer. |
chainer.optimizers.CorrectedMomentumSGD |
Momentum SGD optimizer. |
chainer.optimizers.MomentumSGD |
Momentum SGD optimizer. |
chainer.optimizers.NesterovAG |
Nesterov’s Accelerated Gradient. |
chainer.optimizers.RMSprop |
RMSprop optimizer. |
chainer.optimizers.RMSpropGraves |
Alex Graves’s RMSprop. |
chainer.optimizers.SGD |
Vanilla Stochastic Gradient Descent. |
chainer.optimizers.SMORMS3 |
Simon Funk’s SMORMS3. |
Optimizer base classes¶
chainer.Optimizer |
Base class of all numerical optimizers. |
chainer.UpdateRule |
Base class of all update rules. |
chainer.optimizer.Hyperparameter |
Set of hyperparameter entries of an optimizer. |
chainer.GradientMethod |
Base class of all single gradient-based optimizers. |
Hook functions¶
chainer.optimizer_hooks.WeightDecay |
Optimizer/UpdateRule hook function for weight decay regularization. |
chainer.optimizer_hooks.Lasso |
Optimizer/UpdateRule hook function for Lasso regularization. |
chainer.optimizer_hooks.GradientClipping |
Optimizer hook function for gradient clipping. |
chainer.optimizer_hooks.GradientHardClipping |
Optimizer/UpdateRule hook function for gradient clipping. |
chainer.optimizer_hooks.GradientNoise |
Optimizer/UpdateRule hook function for adding gradient noise. |
chainer.optimizer_hooks.GradientLARS |
Optimizer/UpdateRule hook function for layer wise adaptive rate scaling. |