mindspore.nn.FixedLossScaleUpdateCell

class mindspore.nn.FixedLossScaleUpdateCell(loss_scale_value)[source]

Static scale update cell, the loss scaling value will not be updated.

For usage, refer to DynamicLossScaleUpdateCell.

Parameters

loss_scale_value (float) – Initializes loss scale.

Inputs:
  • loss_scale (Tensor) - The loss scale value during training with shape \(()\), that will be ignored.

  • overflow (bool) - Whether the overflow occurs or not.

Outputs:

bool, the input overflow.

Supported Platforms:

Ascend GPU

Examples

>>> import numpy as np
>>> from mindspore import Tensor, Parameter, nn
>>> from mindspore.ops import operations as P
>>> from mindspore.nn.wrap.cell_wrapper import WithLossCell
>>>
>>> class Net(nn.Cell):
...     def __init__(self, in_features, out_features):
...         super(Net, self).__init__()
...         self.weight = Parameter(Tensor(np.ones([in_features, out_features]).astype(np.float32)),
...                                 name='weight')
...         self.matmul = P.MatMul()
...
...     def construct(self, x):
...         output = self.matmul(x, self.weight)
...         return output
...
>>> in_features, out_features = 16, 10
>>> net = Net(in_features, out_features)
>>> loss = nn.MSELoss()
>>> optimizer = nn.Momentum(net.trainable_params(), learning_rate=0.1, momentum=0.9)
>>> net_with_loss = WithLossCell(net, loss)
>>> manager = nn.FixedLossScaleUpdateCell(loss_scale_value=2**12)
>>> train_network = nn.TrainOneStepWithLossScaleCell(net_with_loss, optimizer, scale_sense=manager)
>>> input = Tensor(np.ones([out_features, in_features]), mindspore.float32)
>>> labels = Tensor(np.ones([out_features,]), mindspore.float32)
>>> output = train_network(input, labels)