actor explores in env, then returns the trajectory_list (env transition) `object env` RL training environment. env.reset() env.step() `int target_step` explored target_step number of step in env `float reward_scale` scale reward, 'reward * reward_scale' `float g
(self, env, target_step, reward_scale, gamma)
source not stored for this graph (policy: none)
nothing calls this directly
no test coverage detected