32c84e8b245ce5bf8b299451ce1290df5d4fb0dc,slm_lab/experiment/control.py,Session,run_episode,#Session#,98

Before Change


        episode_coor = self.monitor.data_coor["episode"]
        // TODO generalize and make state to include observables
        state = self.env.reset()
        logger.debug(f"reset state {state}")

        self.agent.reset()
        // RL steps for SARS
        for t in range(self.env.max_timestep):

After Change


            action_space = self.agent_space.act(state_space)
            logger.debug(f"action_space {action_space}")
            (reward_space, state_space,
             done_space) = self.env_space.step(action_space)
            logger.debug(
                f"reward_space: {reward_space}, state_space: {state_space}, done_space: {done_space}")
            // completes cycle of full info for agent_space
            self.agent_space.update(reward_space, state_space, done_space)
Italian Trulli
In pattern: SUPERPATTERN

Frequency: 4

Non-data size: 3

Instances


Project Name: kengz/SLM-Lab
Commit Name: 32c84e8b245ce5bf8b299451ce1290df5d4fb0dc
Time: 2017-11-26
Author: kengzwl@gmail.com
File Name: slm_lab/experiment/control.py
Class Name: Session
Method Name: run_episode


Project Name: kengz/SLM-Lab
Commit Name: 0ac2b33e8c63304a50db7d2b484368299706b58b
Time: 2018-11-14
Author: kengzwl@gmail.com
File Name: slm_lab/agent/net/recurrent.py
Class Name: RecurrentNet
Method Name: training_step


Project Name: kengz/SLM-Lab
Commit Name: b05a4a145bd485c1b988942b033191ca57410f72
Time: 2018-11-14
Author: kengzwl@gmail.com
File Name: slm_lab/agent/net/mlp.py
Class Name: HydraMLPNet
Method Name: training_step


Project Name: kengz/SLM-Lab
Commit Name: 0ac2b33e8c63304a50db7d2b484368299706b58b
Time: 2018-11-14
Author: kengzwl@gmail.com
File Name: slm_lab/agent/net/mlp.py
Class Name: MLPNet
Method Name: training_step