Skip to content

ding.example.d4pg

ding.example.d4pg

Full Source Code

../ding/example/d4pg.py

1import gym 2from ditk import logging 3from ding.model.template.qac_dist import QACDIST 4from ding.policy import D4PGPolicy 5from ding.envs import DingEnvWrapper, BaseEnvManagerV2 6from ding.data import DequeBuffer 7from ding.data.buffer.middleware import PriorityExperienceReplay 8from ding.config import compile_config 9from ding.framework import task 10from ding.framework.context import OnlineRLContext 11from ding.framework.middleware import OffPolicyLearner, StepCollector, interaction_evaluator, data_pusher, \ 12 CkptSaver, nstep_reward_enhancer 13from ding.utils import set_pkg_seed 14from dizoo.classic_control.pendulum.envs.pendulum_env import PendulumEnv 15from dizoo.classic_control.pendulum.config.pendulum_d4pg_config import main_config, create_config 16 17 18def main(): 19 logging.getLogger().setLevel(logging.INFO) 20 cfg = compile_config(main_config, create_cfg=create_config, auto=True) 21 with task.start(async_mode=False, ctx=OnlineRLContext()): 22 collector_env = BaseEnvManagerV2( 23 env_fn=[lambda: PendulumEnv(cfg.env) for _ in range(cfg.env.collector_env_num)], cfg=cfg.env.manager 24 ) 25 evaluator_env = BaseEnvManagerV2( 26 env_fn=[lambda: PendulumEnv(cfg.env) for _ in range(cfg.env.evaluator_env_num)], cfg=cfg.env.manager 27 ) 28 29 set_pkg_seed(cfg.seed, use_cuda=cfg.policy.cuda) 30 31 model = QACDIST(**cfg.policy.model) 32 buffer_ = DequeBuffer(size=cfg.policy.other.replay_buffer.replay_buffer_size) 33 buffer_.use(PriorityExperienceReplay(buffer_, IS_weight=True)) 34 policy = D4PGPolicy(cfg.policy, model=model) 35 36 task.use(interaction_evaluator(cfg, policy.eval_mode, evaluator_env)) 37 task.use( 38 StepCollector(cfg, policy.collect_mode, collector_env, random_collect_size=cfg.policy.random_collect_size) 39 ) 40 task.use(nstep_reward_enhancer(cfg)) 41 task.use(data_pusher(cfg, buffer_)) 42 task.use(OffPolicyLearner(cfg, policy.learn_mode, buffer_)) 43 task.use(CkptSaver(policy, cfg.exp_name, train_freq=100)) 44 task.run() 45 46 47if __name__ == "__main__": 48 main()