ParamScheduler

class ignite.handlers.param_scheduler.ParamScheduler(optimizer, param_name, save_history=False, param_group_index=None)

An abstract class for updating an optimizer’s parameter value during training.

Parameters

• optimizer (Optimizer) – torch optimizer or any object with attribute param_groups as a sequence.

• param_name (str) – name of optimizer’s parameter to update.

• save_history (bool) – whether to log the parameter values to engine.state.param_history, (default=False).

• param_group_index (int | None) – optimizer’s parameters group to use

Note

Parameter scheduler works independently of the internal state of the attached optimizer. More precisely, whatever the state of the optimizer (newly created or used by another scheduler) the scheduler sets defined absolute values.

Methods

attach	Attach the handler to the engine.
simulate_values	Method to simulate scheduled values during num_events events.

attach(engine, event=Events.ITERATION_STARTED)

Attach the handler to the engine.

Parameters

• engine (Engine) – trainer to which the handler will be attached.

• event (str | ignite.engine.events.Events | ignite.engine.events.CallableEventWithFilter | ignite.engine.events.EventsList) – trigger event to update the param value.

Return type

None

New in version 0.5.4.

classmethod simulate_values(num_events, **scheduler_kwargs)

Method to simulate scheduled values during num_events events.

Parameters

• num_events (int) – number of events during the simulation.

• scheduler_kwargs (Any) – parameter scheduler configuration kwargs.

Returns

event_index, value

Return type

list[list[int]]

Examples:

lr_values = np.array(LinearCyclicalScheduler.simulate_values(num_events=50, param_name='lr',
                                                             start_value=1e-1, end_value=1e-3,
                                                             cycle_size=10))

plt.plot(lr_values[:, 0], lr_values[:, 1], label="learning rate")
plt.xlabel("events")
plt.ylabel("values")
plt.legend()