Index

A

ActorCritic (class in fsrl.utils.net.common)

apply_grid_filter() (fsrl.data.TrajectoryBuffer method) auto_name() (in module fsrl.utils.exp_util)

B

BaseAgent (class in fsrl.agent) BaseLogger (class in fsrl.utils)

BasePolicy (class in fsrl.policy) BaseTrainer (class in fsrl.trainer) BasicCollector (class in fsrl.data)

C

collect() (fsrl.data.BasicCollector method) (fsrl.data.FastCollector method) compute_gae_returns() (fsrl.policy.BasePolicy method) compute_nstep_returns() (fsrl.policy.BasePolicy method)

CPO (class in fsrl.policy) CPOAgent (class in fsrl.agent) CVPO (class in fsrl.policy) CVPOAgent (class in fsrl.agent)

D

DDPGLagAgent (class in fsrl.agent) DDPGLagrangian (class in fsrl.policy)

display_tabular() (fsrl.utils.BaseLogger method) DoubleCritic (class in fsrl.utils.net.continuous) DummyLogger (class in fsrl.utils)

E

evaluate() (fsrl.agent.BaseAgent method) exploration_noise() (fsrl.policy.BasePolicy method) (fsrl.policy.DDPGLagrangian method) (fsrl.policy.SACLagrangian method)

F

FastCollector (class in fsrl.data) filter_points() (fsrl.data.TrajectoryBuffer static method) FOCOPS (class in fsrl.policy) FOCOPSAgent (class in fsrl.agent) forward() (fsrl.policy.BasePolicy method) (fsrl.utils.net.continuous.DoubleCritic method) (fsrl.utils.net.continuous.SingleCritic method) fsrl.agent module

fsrl.data module fsrl.trainer module fsrl.utils.exp_util module fsrl.utils.net.common module fsrl.utils.net.continuous module

G

gather_update_info() (fsrl.trainer.BaseTrainer method) gaussian_kl() (fsrl.policy.CVPO static method) gen_doc() (fsrl.trainer.BaseTrainer static method) get_all() (fsrl.data.TrajectoryBuffer method) get_extra_state() (fsrl.policy.CVPO method) (fsrl.policy.LagrangianPolicy method)

get_lag() (fsrl.utils.LagrangianOptimizer method) get_mean() (fsrl.utils.BaseLogger method) get_mean_dict() (fsrl.utils.BaseLogger method) get_mean_list() (fsrl.utils.BaseLogger method) get_metrics() (fsrl.policy.BasePolicy static method) get_std() (fsrl.utils.BaseLogger method)

L

LagrangianOptimizer (class in fsrl.utils) LagrangianPolicy (class in fsrl.policy) learn() (fsrl.agent.BaseAgent method) (fsrl.agent.CPOAgent method) (fsrl.agent.OffpolicyAgent method) (fsrl.agent.OnpolicyAgent method) (fsrl.agent.TRPOLagAgent method) (fsrl.policy.BasePolicy method)

load_config_and_model() (in module fsrl.utils.exp_util) load_state_dict() (fsrl.utils.LagrangianOptimizer method) logger_keys (fsrl.utils.BaseLogger property)

M

map_action() (fsrl.policy.BasePolicy method) map_action_inverse() (fsrl.policy.BasePolicy method) module fsrl.agent fsrl.data fsrl.trainer fsrl.utils.exp_util fsrl.utils.net.common fsrl.utils.net.continuous

N

name (fsrl.agent.BaseAgent attribute) (fsrl.agent.CPOAgent attribute) (fsrl.agent.CVPOAgent attribute) (fsrl.agent.DDPGLagAgent attribute) (fsrl.agent.FOCOPSAgent attribute) (fsrl.agent.OffpolicyAgent attribute) (fsrl.agent.OnpolicyAgent attribute) (fsrl.agent.PPOLagAgent attribute) (fsrl.agent.SACLagAgent attribute) (fsrl.agent.TRPOLagAgent attribute)

nu_loss() (fsrl.policy.FOCOPS method)

O

offpolicy_trainer() (in module fsrl.trainer) OffpolicyAgent (class in fsrl.agent) OffpolicyTrainer (class in fsrl.trainer)

onpolicy_trainer() (in module fsrl.trainer) OnpolicyAgent (class in fsrl.agent) OnpolicyTrainer (class in fsrl.trainer)

P

perf_is_better() (fsrl.trainer.BaseTrainer method) policy (fsrl.agent.CPOAgent attribute) (fsrl.agent.CVPOAgent attribute) (fsrl.agent.DDPGLagAgent attribute) (fsrl.agent.FOCOPSAgent attribute) (fsrl.agent.OffpolicyAgent attribute) (fsrl.agent.OnpolicyAgent attribute) (fsrl.agent.PPOLagAgent attribute) (fsrl.agent.SACLagAgent attribute) (fsrl.agent.TRPOLagAgent attribute) policy_update_fn() (fsrl.trainer.BaseTrainer method) (fsrl.trainer.OffpolicyTrainer method) (fsrl.trainer.OnpolicyTrainer method)

post_process_fn() (fsrl.policy.BasePolicy method) post_update_fn() (fsrl.policy.BasePolicy method) (fsrl.policy.CVPO method) PPOLagAgent (class in fsrl.agent) PPOLagrangian (class in fsrl.policy) pre_update_fn() (fsrl.policy.BasePolicy method) (fsrl.policy.CPO method) (fsrl.policy.CVPO method) (fsrl.policy.FOCOPS method) (fsrl.policy.LagrangianPolicy method) predict() (fsrl.utils.net.continuous.DoubleCritic method) (fsrl.utils.net.continuous.SingleCritic method) print() (fsrl.utils.BaseLogger method) process_fn() (fsrl.policy.BasePolicy method)

R

reset() (fsrl.data.BasicCollector method) (fsrl.data.FastCollector method) (fsrl.trainer.BaseTrainer method) reset_buffer() (fsrl.data.BasicCollector method) (fsrl.data.FastCollector method) reset_data() (fsrl.utils.BaseLogger method) reset_env() (fsrl.data.BasicCollector method) (fsrl.data.FastCollector method)

reset_stat() (fsrl.data.BasicCollector method) (fsrl.data.FastCollector method) restore_data() (fsrl.utils.BaseLogger method) (fsrl.utils.TensorboardLogger method) (fsrl.utils.WandbLogger method) run() (fsrl.trainer.BaseTrainer method)

S

SACLagAgent (class in fsrl.agent) SACLagrangian (class in fsrl.policy) safety_loss() (fsrl.policy.LagrangianPolicy method) sample() (fsrl.data.TrajectoryBuffer method) save() (fsrl.data.TrajectoryBuffer method) save_checkpoint() (fsrl.utils.BaseLogger method) save_config() (fsrl.utils.BaseLogger method) seed_all() (in module fsrl.utils.exp_util) set_exp_noise() (fsrl.policy.DDPGLagrangian method) (fsrl.policy.SACLagrangian method) set_extra_state() (fsrl.policy.CVPO method) (fsrl.policy.LagrangianPolicy method)

setup_checkpoint_fn() (fsrl.utils.BaseLogger method) SingleCritic (class in fsrl.utils.net.continuous) soft_update() (fsrl.policy.BasePolicy method) state_dict (fsrl.agent.BaseAgent property) state_dict() (fsrl.utils.LagrangianOptimizer method) stats_mean (fsrl.utils.BaseLogger property) step() (fsrl.utils.LagrangianOptimizer method) store() (fsrl.data.TrajectoryBuffer method) (fsrl.utils.BaseLogger method) sync_weight() (fsrl.policy.CVPO method) (fsrl.policy.DDPGLagrangian method) (fsrl.policy.SACLagrangian method)

T

TensorboardLogger (class in fsrl.utils) test_step() (fsrl.trainer.BaseTrainer method) to_string() (in module fsrl.utils.exp_util) train_step() (fsrl.trainer.BaseTrainer method)

training (fsrl.policy.BasePolicy attribute) TrajectoryBuffer (class in fsrl.data) TRPOLagAgent (class in fsrl.agent) TRPOLagrangian (class in fsrl.policy)

U

update() (fsrl.policy.BasePolicy method) update_cost_limit() (fsrl.policy.CPO method) (fsrl.policy.CVPO method) (fsrl.policy.FOCOPS method) (fsrl.policy.LagrangianPolicy method)

update_lagrangian() (fsrl.policy.LagrangianPolicy method)

V

value_mask() (fsrl.policy.BasePolicy static method)

W

WandbLogger (class in fsrl.utils) write() (fsrl.utils.BaseLogger method) (fsrl.utils.TensorboardLogger method) (fsrl.utils.WandbLogger method)

write_without_reset() (fsrl.utils.BaseLogger method) (fsrl.utils.TensorboardLogger method) (fsrl.utils.WandbLogger method)