mindspore.ops.ApplyAddSign

class mindspore.ops.ApplyAddSign

Updates relevant entries according to the AddSign algorithm.

\[\begin{split}\begin{array}{ll} \\ m_{t+1} = \beta * m_{t} + (1 - \beta) * g \\ \text{update} = (\alpha + \text{sign_decay} * sign(g) * sign(m)) * g \\ var = var - lr_{t+1} * \text{update} \end{array}\end{split}\]

\(t\) represents updating step while \(m\) represents the 1st moment vector, \(m_{t}\) is the last moment of \(m_{t+1}\), \(lr\) represents scaling factor lr, \(g\) represents grad, \(\alpha\) represents alpha, \(\beta\) represents beta.

The data type of all inputs must be float16 or float32 on Ascend and float16, float32 or float64 on CPU and GPU.

Inputs of var, accum and grad , sign_decay and beta comply with the implicit type conversion rules to make the data types consistent. If they have different data types, the lower priority data type will be converted to the relatively highest priority data type.

Inputs:

• var (Union[Parameter, Tensor]) - Variable tensor to be updated. The shape is \((N, *)\) where \(*\) means, any number of additional dimensions.

• m (Union[Parameter, Tensor]) - Variable tensor to be updated, has the same data type as var.

• lr (Union[Number, Tensor]) - The learning rate value, must be a scalar.

• alpha (Union[Number, Tensor]) - Must be a scalar.

• sign_decay (Union[Number, Tensor]) - Must be a scalar.

• beta (Union[Number, Tensor]) - The exponential decay rate, must be a scalar.

• grad (Tensor) - A tensor of the same shape as var, for the gradient.

Outputs:

Tuple of 2 Tensors, the updated parameters or tensors.

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

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

Raises

• TypeError – If dtype of var, lr and alpha is not float16, float32 or float64.

• TypeError – If dtype of sign_decay and beta are both not float16, float32 or float64.

• TypeError – If lr, alpha or sign_decay is neither a Number nor a Tensor.

• TypeError – If grad is not a Tensor.

• TypeError – If the data type of var, accum and grad conversion of Parameter is not supported.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import numpy as np
>>> from mindspore import Tensor, nn, ops, Parameter
>>> class Net(nn.Cell):
...     def __init__(self):
...         super(Net, self).__init__()
...         self.apply_add_sign = ops.ApplyAddSign()
...         self.var = Parameter(Tensor(np.array([[0.6, 0.4],
...                                               [0.1, 0.5]]).astype(np.float32)), name="var")
...         self.m = Parameter(Tensor(np.array([[0.6, 0.5],
...                                             [0.2, 0.6]]).astype(np.float32)), name="m")
...         self.lr = 0.001
...         self.alpha = 1.0
...         self.sign_decay = 0.99
...         self.beta = 0.9
...     def construct(self, grad):
...         out = self.apply_add_sign(self.var, self.m, self.lr, self.alpha, self.sign_decay, self.beta, grad)
...         return out
...
>>> net = Net()
>>> grad = Tensor(np.array([[0.3, 0.7], [0.1, 0.8]]).astype(np.float32))
>>> output = net(grad)
>>> print(output)
(Tensor(shape=[2, 2], dtype=Float32, value=
[[ 5.99403024e-01,  3.98607016e-01],
 [ 9.98010039e-02,  4.98407990e-01]]), Tensor(shape=[2, 2], dtype=Float32, value=
[[ 5.70000052e-01,  5.19999981e-01],
 [ 1.89999998e-01,  6.20000064e-01]]))