PiecewiseLinearStateScheduler

class ignite.handlers.state_param_scheduler.PiecewiseLinearStateScheduler(milestones_values, param_name, save_history=False, create_new=False)

Piecewise linear state parameter scheduler.

Parameters

• milestones_values (List[Tuple[int, float]]) – list of tuples (event index, parameter value) represents milestones and parameter values. Milestones should be increasing integers.

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

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

• create_new (bool) – whether to create param_name on engine.state taking into account whether param_name attribute already exists or not. Overrides existing attribute by default, (default=False).

Examples

param_scheduler = PiecewiseLinearStateScheduler(
    param_name="param",  milestones_values=[(5, 1.0), (10, 0.8), (15, 0.6)], create_new=True
)

# parameter is param, milestone (5, 1.0) sets param to 1.0
# milestone is (5, 1.0), param=1  for Epoch 1 to 5,
# next milestone is (10, 0.8), param linearly reduces from 1.0 to 0.8
# Epoch 10, param = 0.8
# next milestone is (15,0.6), param linearly reduces from 0.8 to 0.6
# Epoch 15, param = 0.6

param_scheduler.attach(default_trainer, Events.EPOCH_COMPLETED)

@default_trainer.on(Events.EPOCH_COMPLETED)
def print_param():
    print(default_trainer.state.param)

default_trainer.run([0], max_epochs=15)

1.0
1.0
1.0
1.0
1.0
0.96
0.92
0.88
0.8400...
0.8
0.76
0.72
0.68
0.64
0.6

New in version 0.4.7.

Methods

get_param

Method to get current parameter values

get_param()

Method to get current parameter values

Returns

list of params, or scalar param

Return type

Union[List[float], float]