cambrian.ml.trainer
===================

.. py:module:: cambrian.ml.trainer

.. autoapi-nested-parse::

   This module contains the trainer class for training and evaluating agents.



Classes
-------

.. autoapisummary::

   cambrian.ml.trainer.MjCambrianTrainerConfig
   cambrian.ml.trainer.MjCambrianTrainer


Module Contents
---------------

.. py:class:: MjCambrianTrainerConfig

   Bases: :py:obj:`hydra_config.HydraContainerConfig`


   Settings for the training process. Used for type hinting.

   :ivar total_timesteps: The total number of timesteps to train for.
   :vartype total_timesteps: int
   :ivar max_episode_steps: The maximum number of steps per episode.
   :vartype max_episode_steps: int
   :ivar n_envs: The number of parallel environments to use for training.

   :vartype n_envs: int
   :ivar model: The model to use for
                training.
   :vartype model: Callable[[MjCambrianEnv], MjCambrianModel]
   :ivar callbacks: The callbacks to use for training.
   :vartype callbacks: Dict[str, BaseCallback]
   :ivar wrappers: The wrappers to use for
                   training. If None, will ignore.

   :vartype wrappers: Dict[str, Callable[[VecEnv], VecEnv]] | None
   :ivar prune_fn: The function to use to
                   determine if an experiment should be pruned. If None, will ignore. If set,
                   this function will be called prior to training to check whether the config
                   is valid for training. This is the get around the fact that some sweepers
                   will evaluate configs that are invalid for training, which is a waste
                   computationally. The train method will return -inf if this function returns
                   True. NOTE: for nevergrad, it is recommended to use cheap_constraints.
   :vartype prune_fn: Optional[Callable[[MjCambrianConfig], bool]]
   :ivar fitness_fn: The function to use to
                     calculate the fitness of the agent after training.

   :vartype fitness_fn: Callable[[MjCambrianConfig, float]]


.. py:class:: MjCambrianTrainer(config)

   This is the trainer class for running training and evaluation.

   :Parameters: **config** (*MjCambrianConfig*) -- The config to use for training and evaluation.


   .. py:method:: train()

      Train the agent.