ray.rllib.algorithms.algorithm.Algorithm.train

Algorithm.train()

Runs one logical iteration of training.

Calls step() internally. Subclasses should override step() instead to return results. This method automatically fills the following fields in the result:

done (bool): training is terminated. Filled only if not provided.

time_this_iter_s (float): Time in seconds this iteration took to run. This may be overridden in order to override the system-computed time difference.

time_total_s (float): Accumulated time in seconds for this entire experiment.

training_iteration (int): The index of this training iteration, e.g. call to train(). This is incremented after step() is called.

pid (str): The pid of the training process.

date (str): A formatted date of when the result was processed.

timestamp (str): A UNIX timestamp of when the result was processed. This may be overridden.

hostname (str): Hostname of the machine hosting the training process.

node_ip (str): Node ip of the machine hosting the training process.

Returns:

A dict that describes training progress.