mindspore.ops.AdamWeightDecay

class mindspore.ops.AdamWeightDecay(use_locking=False)

Updates gradients by the Adaptive Moment Estimation algorithm with weight decay (AdamWeightDecay).

The Adam algorithm is proposed in Adam: A Method for Stochastic Optimization. The AdamWeightDecay variant was proposed in Decoupled Weight Decay Regularization.

The updating formulas are as follows,

$$\begin{array}{r}\begin{array}{l}\\ m={\beta }_1\ast m+(1-{\beta }_1)\ast g\\ v={\beta }_2\ast v+(1-{\beta }_2)\ast g\ast g\\ update=\frac{m}{\sqrt{v}+ϵ}\\ update=\{\begin{array}{ll}update+weight\mathrm{\_}decay\ast w& \text{ if }weight\mathrm{\_}decay>0\\ update& \text{ otherwise }\end{array}\\ w=w-lr\ast update\end{array}\end{array}$$

$m$ represents the 1st moment vector, $v$ represents the 2nd moment vector, $g$ represents gradient, ${\beta }_1,{\beta }_2$ represent beta1 and beta2, $lr$ represents learning_rate, $w$ represents var, $decay$ represents weight_decay, $ϵ$ represents epsilon.

Parameters

use_locking (bool) – Whether to enable a lock to protect variable tensors from being updated. If True , updates of the var, m, and v tensors will be protected by a lock. If False , the result is unpredictable. Default: False .

Inputs:

• var (Union[Parameter, Tensor]) - Weights to be updated. The shape is $(N,\ast )$ where $\ast$ means any number of additional dimensions. The data type can be float16 or float32.

• m (Union[Parameter, Tensor]) - The 1st moment vector in the updating formula, it should have the the shape as var. The data type can be float16 or float32.

• v (Union[Parameter, Tensor]) - The 2nd moment vector in the updating formula, it should have the same shape as m.

• lr (float) - $lr$ in the updating formula. The paper suggested value is ${10}^{-8}$, the data type should be float32.

• beta1 (float) - The exponential decay rate for the 1st moment estimations, the data type should be float32. The paper suggested value is $0.9$

• beta2 (float) - The exponential decay rate for the 2nd moment estimations, the data type should be float32. The paper suggested value is $0.999$

• epsilon (float) - Term added to the denominator to improve numerical stability, the data type should be float32.

• decay (float) - The weight decay value, must be a scalar tensor with float32 data type. Default: 0.0 .

• gradient (Tensor) - Gradient, has the same shape as var.

Outputs:

Tuple of 3 Tensor, the updated parameters.

• var (Tensor) - The same shape and data type as var.

• m (Tensor) - The same shape and data type as m.

• v (Tensor) - The same shape and data type as v.

Raises

• TypeError – If use_locking is not a bool.

• TypeError – If lr, beta1, beta2, epsilon or decay is not a float32.

• TypeError – If var, m or v is neither float16 nor float32.

• TypeError – If gradient is not a Tensor.

• ValueError – If epsilon <= 0.

• ValueError – If beta1, beta2 is not in range (0.0,1.0).

• ValueError – If decay < 0.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import numpy as np
>>> import mindspore.nn as nn
>>> from mindspore import Tensor, Parameter, ops
>>> class Net(nn.Cell):
...     def __init__(self):
...         super(Net, self).__init__()
...         self.adam_weight_decay = ops.AdamWeightDecay()
...         self.var = Parameter(Tensor(np.ones([2, 2]).astype(np.float32)), name="var")
...         self.m = Parameter(Tensor(np.ones([2, 2]).astype(np.float32)), name="m")
...         self.v = Parameter(Tensor(np.ones([2, 2]).astype(np.float32)), name="v")
...     def construct(self, lr, beta1, beta2, epsilon, decay, grad):
...         out = self.adam_weight_decay(self.var, self.m, self.v, lr, beta1, beta2,
...                               epsilon, decay, grad)
...         return out
>>> net = Net()
>>> gradient = Tensor(np.ones([2, 2]).astype(np.float32))
>>> output = net(0.001, 0.9, 0.999, 1e-8, 0.0, gradient)
>>> print(net.var.asnumpy())
[[0.999 0.999]
[0.999 0.999]]