Shortcuts

MultiStepParamScheduler

class mmengine.optim.MultiStepParamScheduler(optimizer, param_name, milestones, gamma=0.1, last_step=- 1, begin=0, end=1000000000, by_epoch=True, verbose=False)[source]

Decays the specified parameter in each parameter group by gamma once the number of epoch reaches one of the milestones. Notice that such decay can happen simultaneously with other changes to the parameter from outside this scheduler.

Parameters
  • optimizer (BaseOptimWrapper or Optimizer) – Wrapped optimizer.

  • param_name (str) – Name of the parameter to be adjusted, such as lr, momentum.

  • milestones (list) – List of epoch indices. Must be increasing.

  • gamma (float) – Multiplicative factor of parameter value decay. Defaults to 0.1.

  • begin (int) – Step at which to start updating the parameters. Defaults to 0.

  • end (int) – Step at which to stop updating the parameters. Defaults to INF.

  • last_step (int) – The index of last step. Used for resume without state dict. Defaults to -1.

  • by_epoch (bool) – Whether the scheduled parameters are updated by epochs. Defaults to True.

  • verbose (bool) – Whether to print the value for each update. Defaults to False.

classmethod build_iter_from_epoch(*args, milestones, begin=0, end=1000000000, by_epoch=True, epoch_length=None, **kwargs)[source]

Build an iter-based instance of this scheduler from an epoch-based config.

Read the Docs v: v0.8.3
Versions
latest
stable
v0.8.3
v0.8.2
v0.8.1
v0.8.0
v0.7.4
v0.7.3
v0.7.2
v0.7.1
v0.7.0
v0.6.0
v0.5.0
v0.4.0
v0.3.0
v0.2.0
Downloads
On Read the Docs
Project Home
Builds

Free document hosting provided by Read the Docs.