mindspore.nn.ASGD

class mindspore.nn.ASGD(*args, **kwargs)

Implements Average Stochastic Gradient Descent.

Introduction to ASGD can be found at Acceleration of stochastic approximation by average.

The updating formulas are as follows:

\[\begin{split}\begin{gather*} w_{t} = w_{t-1} * (1 - \lambda * \eta_{t-1}) - \eta_{t-1} * g_{t} \\ ax_{t} = (w_t - ax_{t-1}) * \mu_{t-1} \\ \eta_{t} = \frac{1.}{(1 + \lambda * lr * t)^\alpha} \\ \mu_{t} = \frac{1}{\max(1, t - t0)} \end{gather*}\end{split}\]

\(\lambda\) represents the decay term, \(\mu\) and \(\eta\) are tracked to update \(ax\) and \(w\), \(t0\) represents the point of starting averaging, \(\alpha\) represents the power for eta update, \(ax\) represents the averaged parameter value, \(t\) represents the current step, \(g\) represents gradients, \(w\) represents params.

Note

If parameters are not grouped, the weight_decay in optimizer will be applied on the parameters without ‘beta’ or ‘gamma’ in their names. Users can group parameters to change the strategy of decaying weight. When parameters are grouped, each group can set weight_decay, if not, the weight_decay in optimizer will be applied.

Parameters

• params (Union[list[Parameter], list[dict]]) –

  Must be list of Parameter or list of dict. When the parameters is a list of dict, the “params”, “lr”, “weight_decay”, “grad_centralization” and “order_params” are the keys can be parsed.

  • params: Required. Parameters in current group. The value must be a list of Parameter.

  • lr: Optional. If “lr” in the keys, the value of corresponding learning rate will be used. If not, the learning_rate in optimizer will be used. Fixed and dynamic learning rate are supported.

  • weight_decay: Optional. If “weight_decay” in the keys, the value of corresponding weight decay will be used. If not, the weight_decay in the optimizer will be used. It should be noted that weight decay can be a constant value or a Cell. It is a Cell only when dynamic weight decay is applied. Dynamic weight decay is similar to dynamic learning rate, users need to customize a weight decay schedule only with global step as input, and during training, the optimizer calls the instance of WeightDecaySchedule to get the weight decay value of current step.

  • grad_centralization: Optional. Must be Boolean. If “grad_centralization” is in the keys, the set value will be used. If not, the grad_centralization is False by default. This configuration only works on the convolution layer.

  • order_params: Optional. When parameters is grouped, this usually is used to maintain the order of parameters that appeared in the network to improve performance. The value should be parameters whose order will be followed in optimizer. If order_params in the keys, other keys will be ignored and the element of ‘order_params’ must be in one group of params.

• learning_rate (Union[float, int, Tensor, Iterable, LearningRateSchedule]) –

  • float: The fixed learning rate value. Must be equal to or greater than 0.

  • int: The fixed learning rate value. Must be equal to or greater than 0. It will be converted to float.

  • Tensor: Its value should be a scalar or a 1-D vector. For scalar, fixed learning rate will be applied. For vector, learning rate is dynamic, then the i-th step will take the i-th value as the learning rate.

  • Iterable: Learning rate is dynamic. The i-th step will take the i-th value as the learning rate.

  • LearningRateSchedule: Learning rate is dynamic. During training, the optimizer calls the instance of LearningRateSchedule with step as the input to get the learning rate of current step.

• lambd (float) – The decay term. Default: 1e-4.

• alpha (float) – The power for eta update. Default: 0.75.

• t0 (float) – The point of starting averaging. Default: 1e6.

• weight_decay (Union[float, int, Cell]) –

  Weight decay (L2 penalty). Default: 0.0.

  • float: The fixed weight decay value. Must be equal to or greater than 0.

  • int: The fixed weight decay value. Must be equal to or greater than 0. It will be converted to float.

  • Cell: Weight decay is dynamic. During training, the optimizer calls the instance of the Cell with step as the input to get the weight decay value of current step.

Inputs:

• gradients (tuple[Tensor]) - The gradients of params, the shape is the same as params.

Outputs:

Tensor[bool], the value is True.

Raises

• TypeError – If learning_rate is not one of int, float, Tensor, Iterable, LearningRateSchedule.

• TypeError – If element of parameters is neither Parameter nor dict.

• TypeError – If lambd, alpha or t0 is not a float.

• TypeError – If weight_decay is neither float nor int.

• ValueError – If weight_decay is less than 0.

Supported Platforms:

Ascend GPU CPU

Examples

>>> from mindspore import nn, Model
>>>
>>> net = Net()
>>> #1) All parameters use the same learning rate and weight decay
>>> optim = nn.ASGD(params=net.trainable_params())
>>>
>>> #2) Use parameter groups and set different values
>>> conv_params = list(filter(lambda x: 'conv' in x.name, net.trainable_params()))
>>> no_conv_params = list(filter(lambda x: 'conv' not in x.name, net.trainable_params()))
>>> group_params = [{'params': conv_params,'grad_centralization':True},
...                 {'params': no_conv_params, 'lr': 0.01},
...                 {'order_params': net.trainable_params()}]
>>> optim = nn.ASGD(group_params, learning_rate=0.1, weight_decay=0.0)
>>> # The conv_params's parameters will use default learning rate of 0.1 default weight decay of 0.0 and grad
>>> # centralization of True.
>>> # The no_conv_params's parameters will use learning rate of 0.01 and default weight decay of 0.0 and grad
>>> # centralization of False.
>>> # The final parameters order in which the optimizer will be followed is the value of 'order_params'.
>>>
>>> loss = nn.SoftmaxCrossEntropyWithLogits()
>>> model = Model(net, loss_fn=loss, optimizer=optim)