ray.rllib.core.learner.learner.Learner.save_state#

Learner.save_state(path: Union[str, pathlib.Path]) → None[source]#

Save the state of the learner to path

NOTE: if path doesn’t exist, then a new directory will be created. otherwise, it will be appended to.

the state of the learner is saved in the following format:

checkpoint_dir/
    learner_state.json
    module_state/
        module_1/
            ...
    optimizer_state/
        optimizers_module_1/
            ...
Parameters

path – The path to the directory to save the state to.