mindspore.experimental.optim.lr_scheduler.LinearLR

View Source On Gitee
class mindspore.experimental.optim.lr_scheduler.LinearLR(optimizer, start_factor=1.0 / 3, end_factor=1.0, total_iters=5, last_epoch=- 1)[source]

Decays the learning rate of each parameter group by linearly changing small multiplicative factor until the number of epoch reaches a pre-defined milestone: total_iters. Notice that such decay can happen simultaneously with other changes to the learning rate from outside this scheduler.

Warning

This is an experimental lr scheduler module that is subject to change. This module must be used with optimizers in Experimental Optimizer .

Parameters

  • optimizer (mindspore.experimental.optim.Optimizer) – Wrapped optimizer.

  • start_factor (float, optional) – The number we multiply learning rate in the first epoch. The multiplication factor changes towards end_factor in the following epochs. Default: 1.0 /3.

  • end_factor (float, optional) – The number we multiply learning rate at the end of linear changing process. Default: 1.0.

  • total_iters (int, optional) – The number of iterations that multiplicative factor reaches to 1. Default: 5.

  • last_epoch (int, optional) – The index of the last epoch. Default: -1.

Raises

  • ValueError – If start_factor is not in the range of (0, 1].

  • ValueError – If end_factor is not in the range of [0, 1].

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> from mindspore import nn
>>> from mindspore.experimental import optim
>>> # Define the network structure of LeNet5. Refer to
>>> # https://gitee.com/mindspore/docs/blob/r2.2/docs/mindspore/code/lenet.py
>>> net = LeNet5()
>>> loss_fn = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
>>> optimizer = optim.Adam(net.trainable_params(), lr=0.05)
>>> # Assuming optimizer uses lr = 0.05 for all groups
>>> # lr = 0.025    if epoch == 0
>>> # lr = 0.03125  if epoch == 1
>>> # lr = 0.0375   if epoch == 2
>>> # lr = 0.04375  if epoch == 3
>>> # lr = 0.05    if epoch >= 4
>>> scheduler = optim.lr_scheduler.LinearLR(optimizer, start_factor=0.5, total_iters=4)
>>> def forward_fn(data, label):
...     logits = net(data)
...     loss = loss_fn(logits, label)
...     return loss, logits
>>> grad_fn = mindspore.value_and_grad(forward_fn, None, optimizer.parameters, has_aux=True)
>>> def train_step(data, label):
...     (loss, _), grads = grad_fn(data, label)
...     optimizer(grads)
...     return loss
>>> for epoch in range(5):
...     # Create the dataset taking MNIST as an example. Refer to
...     # https://gitee.com/mindspore/docs/blob/r2.2/docs/mindspore/code/mnist.py
...     for data, label in create_dataset():
...         train_step(data, label)
...     scheduler.step()
...     current_lr = scheduler.get_last_lr()