mindspore.ops.ApplyGradientDescent

class mindspore.ops.ApplyGradientDescent[source]

Updates var by subtracting alpha * delta from it.

\[var = var - \alpha * \delta\]

where \(\alpha\) represents alpha, \(\delta\) represents delta.

Inputs of var and delta 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 (Parameter) - Variable tensor to be updated. With float32 or float16 data type. The shape is \((N, *)\) where \(*\) means, any number of additional dimensions.

  • alpha (Union[Number, Tensor]) - Scaling factor, must be a scalar. With float32 or float16 data type.

  • delta (Tensor) - A tensor for the change, has the same shape as var.

Outputs:

Tensor, represents the updated var.

Raises

  • TypeError – If dtype of var or alpha is neither float16 nor float32.

  • TypeError – If delta is not a Tensor.

  • TypeError – If alpha is neither a Number nor a Tensor.

  • TypeError – If the data type of var and delta 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_gradient_descent = ops.ApplyGradientDescent()
...         self.var = Parameter(Tensor(np.ones([2, 2]).astype(np.float32)), name="var")
...         self.alpha = 0.001
...     def construct(self, delta):
...         out = self.apply_gradient_descent(self.var, self.alpha, delta)
...         return out
...
>>> net = Net()
>>> delta = Tensor(np.array([[0.1, 0.1], [0.1, 0.1]]).astype(np.float32))
>>> output = net(delta)
>>> print(output)
[[0.9999 0.9999]
 [0.9999 0.9999]]