ray/rllib/agents/ddpg/__init__.py

from ray.rllib.agents.ddpg.apex import ApexDDPGTrainer
from ray.rllib.agents.ddpg.ddpg import DDPGTrainer, DEFAULT_CONFIG
from ray.rllib.agents.ddpg.td3 import TD3Trainer
from ray.rllib.utils import renamed_agent

ApexDDPGAgent = renamed_agent(ApexDDPGTrainer)
DDPGAgent = renamed_agent(DDPGTrainer)

__all__ = [
    "DDPGAgent", "ApexDDPGAgent", "DDPGTrainer", "ApexDDPGTrainer",
    "TD3Trainer", "DEFAULT_CONFIG"
]
[rllib] Rename Agent to Trainer (#4556) 2019-04-07 00:36:18 -07:00			`from ray.rllib.agents.ddpg.apex import ApexDDPGTrainer`
			`from ray.rllib.agents.ddpg.ddpg import DDPGTrainer, DEFAULT_CONFIG`
[rllib] TD3/DDPG improvements and MuJoCo benchmarks (#4694) * [rllib] Separate optimisers for DDPG actor & crit. * [rllib] Better names for DDPG variables & options Config changes: - noise_scale -> exploration_ou_noise_scale - exploration_theta -> exploration_ou_theta - exploration_sigma -> exploration_ou_sigma - act_noise -> exploration_gaussian_sigma - noise_clip -> target_noise_clip * [rllib] Make DDPG less class-y Used functions to replace three classes with only an __init__ method & a handful of unrelated attributes. * [rllib] Refactor DDPG noise * [rllib] Unify DDPG exploration annealing Added option "exploration_should_anneal" to enable linear annealing of exploration noise. By default this is off, for consistency with DDPG & TD3 papers. Also renamed "exploration_final_eps" to "exploration_final_scale" (that name seems to have been carried over from DQN, and doesn't really make sense here). Finally, tried to rename "eps" to "noise_scale" wherever possible. 2019-04-26 17:49:53 -07:00			`from ray.rllib.agents.ddpg.td3 import TD3Trainer`
[rllib] Rename PolicyGraph => Policy, move from evaluation/ to policy/ (#4819) This implements some of the renames proposed in #4813 We leave behind backwards-compatibility aliases for *PolicyGraph and SampleBatch. 2019-05-20 16:46:05 -07:00			`from ray.rllib.utils import renamed_agent`
[RLLib] DDPG (#1685) 2018-04-11 15:08:39 -07:00
[rllib] Rename PolicyGraph => Policy, move from evaluation/ to policy/ (#4819) This implements some of the renames proposed in #4813 We leave behind backwards-compatibility aliases for *PolicyGraph and SampleBatch. 2019-05-20 16:46:05 -07:00			`ApexDDPGAgent = renamed_agent(ApexDDPGTrainer)`
			`DDPGAgent = renamed_agent(DDPGTrainer)`
[rllib] Rename Agent to Trainer (#4556) 2019-04-07 00:36:18 -07:00
			`__all__ = [`
			`"DDPGAgent", "ApexDDPGAgent", "DDPGTrainer", "ApexDDPGTrainer",`
[rllib] TD3/DDPG improvements and MuJoCo benchmarks (#4694) * [rllib] Separate optimisers for DDPG actor & crit. * [rllib] Better names for DDPG variables & options Config changes: - noise_scale -> exploration_ou_noise_scale - exploration_theta -> exploration_ou_theta - exploration_sigma -> exploration_ou_sigma - act_noise -> exploration_gaussian_sigma - noise_clip -> target_noise_clip * [rllib] Make DDPG less class-y Used functions to replace three classes with only an __init__ method & a handful of unrelated attributes. * [rllib] Refactor DDPG noise * [rllib] Unify DDPG exploration annealing Added option "exploration_should_anneal" to enable linear annealing of exploration noise. By default this is off, for consistency with DDPG & TD3 papers. Also renamed "exploration_final_eps" to "exploration_final_scale" (that name seems to have been carried over from DQN, and doesn't really make sense here). Finally, tried to rename "eps" to "noise_scale" wherever possible. 2019-04-26 17:49:53 -07:00			`"TD3Trainer", "DEFAULT_CONFIG"`
[rllib] Rename Agent to Trainer (#4556) 2019-04-07 00:36:18 -07:00			`]`