scripts.sb3

Full path: schola.scripts.sb3

schola.scripts.sb3

Scripts for training with Stable Baselines 3.

Modules

Item	Description
sb3_to_onnx	Script to convert a Stable Baselines3 model to an ONNX model for use in Unreal Engine.
settings	—
train	Script to train a Stable Baselines3 model using Schola.
utils	Collection of custom callback classes for training NPCRL through Schola with stable baselines3

• schola.scripts.sb3.sb3_to_onnx

• schola.scripts.sb3.sb3_to_onnx.export

• export()

• schola.scripts.sb3.settings

• schola.scripts.sb3.settings.BaseSb3AlgorithmArgs

• BaseSb3AlgorithmArgs

• BaseSb3AlgorithmArgs.init()

• BaseSb3AlgorithmArgs.batch_size

• BaseSb3AlgorithmArgs.learning_rate

• BaseSb3AlgorithmArgs.n_steps

• schola.scripts.sb3.settings.PPOSettings

• PPOSettings

• PPOSettings.init()

• PPOSettings.clip_range

• PPOSettings.constructor

• PPOSettings.critic_type

• PPOSettings.ent_coef

• PPOSettings.gae_lambda

• PPOSettings.gamma

• PPOSettings.max_grad_norm

• PPOSettings.n_epochs

• PPOSettings.name

• PPOSettings.normalize_advantage

• PPOSettings.sde_sample_freq

• PPOSettings.use_sde

• PPOSettings.vf_coef

• schola.scripts.sb3.settings.SACSettings

• SACSettings

• SACSettings.init()

• SACSettings.batch_size

• SACSettings.buffer_size

• SACSettings.constructor

• SACSettings.critic_type

• SACSettings.ent_coef

• SACSettings.gamma

• SACSettings.gradient_steps

• SACSettings.learning_rate

• SACSettings.learning_starts

• SACSettings.name

• SACSettings.optimize_memory_usage

• SACSettings.replay_buffer_kwargs

• SACSettings.sde_sample_freq

• SACSettings.target_entropy

• SACSettings.target_update_interval

• SACSettings.tau

• SACSettings.train_freq

• SACSettings.use_sde

• schola.scripts.sb3.settings.SB3ScriptArgs

• SB3ScriptArgs

• SB3ScriptArgs.init()

• SB3ScriptArgs.algorithm_settings

• SB3ScriptArgs.checkpoint_settings

• SB3ScriptArgs.disable_eval

• SB3ScriptArgs.environment_settings

• SB3ScriptArgs.logging_settings

• SB3ScriptArgs.name_prefix

• SB3ScriptArgs.network_architecture_settings

• SB3ScriptArgs.pbar

• SB3ScriptArgs.resume_settings

• SB3ScriptArgs.timesteps

• schola.scripts.sb3.settings.Sb3CheckpointArgs

• Sb3CheckpointArgs

• Sb3CheckpointArgs.init()

• Sb3CheckpointArgs.save_replay_buffer

• Sb3CheckpointArgs.save_vecnormalize

• schola.scripts.sb3.settings.Sb3LoggingArgs

• Sb3LoggingArgs

• Sb3LoggingArgs.init()

• Sb3LoggingArgs.callback_verbosity

• Sb3LoggingArgs.enable_tensorboard

• Sb3LoggingArgs.log_dir

• Sb3LoggingArgs.log_freq

• Sb3LoggingArgs.sb3_verbosity

• Sb3LoggingArgs.schola_verbosity

• schola.scripts.sb3.settings.Sb3NetworkArchitectureArgs

• Sb3NetworkArchitectureArgs

• Sb3NetworkArchitectureArgs.init()

• Sb3NetworkArchitectureArgs.activation

• Sb3NetworkArchitectureArgs.critic_parameters

• Sb3NetworkArchitectureArgs.policy_parameters

• schola.scripts.sb3.settings.Sb3ResumeArgs

• Sb3ResumeArgs

• Sb3ResumeArgs.init()

• Sb3ResumeArgs.load_replay_buffer

• Sb3ResumeArgs.load_vecnormalize

• Sb3ResumeArgs.reset_timestep

• Sb3ResumeArgs.resume_from

• schola.scripts.sb3.train

• schola.scripts.sb3.train.main

• main()

• schola.scripts.sb3.train.ppo

• ppo()

• schola.scripts.sb3.train.sac

• sac()

• schola.scripts.sb3.train.warn_if_small_image_observation

• warn_if_small_image_observation()

• schola.scripts.sb3.utils

• schola.scripts.sb3.utils.CustomProgressBarCallback

• CustomProgressBarCallback

• CustomProgressBarCallback.init()

• CustomProgressBarCallback.pbar

• schola.scripts.sb3.utils.RewardCallback

• RewardCallback

• RewardCallback.num_envs

• RewardCallback.callbacks

• RewardCallback.summarize_every

• RewardCallback.curr_logging_interval

• RewardCallback.start_time

• RewardCallback.init()

• RewardCallback.ready_to_log

• schola.scripts.sb3.utils.SingleEnvRewardCallback

• SingleEnvRewardCallback

• SingleEnvRewardCallback.episode_reward

• SingleEnvRewardCallback.episode_rewards

• SingleEnvRewardCallback.episode_steps

• SingleEnvRewardCallback.step_count

• SingleEnvRewardCallback.last_logging_interval

• SingleEnvRewardCallback.logging_interval_size

• SingleEnvRewardCallback.id

• SingleEnvRewardCallback.init()

• SingleEnvRewardCallback.get_reward_interval()

• SingleEnvRewardCallback.get_step_interval()

• SingleEnvRewardCallback.increment_logging_interval()

• SingleEnvRewardCallback.ready_to_log