tf_agents.train.triggers.ReverbCheckpointTrigger

Checkpoints data from Reverb replay buffer.

Inherits From: IntervalTrigger

tf_agents.train.triggers.ReverbCheckpointTrigger(
    train_step: tf.Variable,
    interval: int,
    reverb_client: tf_agents.typing.types.ReverbClient
)

train_step tf.Variable which keeps track of the number of train steps.
interval How often, in train_steps, the trigger will save. Note that as long as the >= interval number of steps have passed since the last trigger, the event gets triggered. The current value is not necessarily interval steps away from the last triggered value.
reverb_client the Reverb client required for checkpointing.

Methods

reset

View source

reset() -> None

Resets the trigger interval.

set_start

View source

set_start(
    start: int
) -> None

__call__

View source

__call__(
    value: int, force_trigger: bool = False
) -> None

Maybe trigger the event based on the interval.

Args
value the value for triggering.
force_trigger If True, the trigger will be forced triggered unless the last trigger value is equal to value.