ray.rllib.callbacks.callbacks.RLlibCallback#
- class ray.rllib.callbacks.callbacks.RLlibCallback[source]#
Abstract base class for RLlib callbacks (similar to Keras callbacks).
These callbacks can be used for custom metrics and custom postprocessing.
By default, all of these callbacks are no-ops. To configure custom training callbacks, subclass RLlibCallback and then set {“callbacks”: YourCallbacksClass} in the algo config.
Methods
Callback run when a new Algorithm instance has finished setup.
Callback run when an Algorithm has loaded a new state from a checkpoint.
Callback run whenever a new policy is added to an algorithm.
Callback run after one or more EnvRunner actors have been recreated.
Callback run when a new environment object has been created.
Callback run when a new episode is created (but has not started yet!).
Called when an episode is done (after terminated/truncated have been logged).
Callback run right after an Episode has been started.
Called on each episode step (after the action(s) has/have been logged).
Runs when the evaluation is done.
Runs when the offline evaluation is done.
Callback before offline evaluation starts.
Callback before evaluation starts.
Called at the beginning of Policy.learn_on_batch().
Callback run after one or more OfflineEvaluationRunner actors have been recreated.
Called immediately after a policy's postprocess_fn is called.
Called at the end of
EnvRunner.sample()
.Callback run when a new sub-environment has been created.
Called at the end of Algorithm.train().