ray/rllib/tuned_examples/dqn/atari-dist-dqn.yaml

atari-dist-dqn:
    env:
        grid_search:
            - BreakoutNoFrameskip-v4
            - BeamRiderNoFrameskip-v4
            - QbertNoFrameskip-v4
            - SpaceInvadersNoFrameskip-v4
    run: DQN
    config:
        double_q: false
        dueling: false
        num_atoms: 51
        noisy: false
        replay_buffer_config:
          type: MultiAgentReplayBuffer
          capacity: 1000000
        num_steps_sampled_before_learning_starts: 20000
        n_step: 1
        target_network_update_freq: 8000
        lr: .0000625
        adam_epsilon: .00015
        hiddens: [512]
        rollout_fragment_length: 4
        train_batch_size: 32
        exploration_config:
          epsilon_timesteps: 200000
          final_epsilon: 0.01
        num_gpus: 0.2
        min_sample_timesteps_per_iteration: 10000
[rllib] Port IMPALA to ModelV2/build_tf_policy (#5130) * port vtrace * fix vf * fix vs * fix the example * wip ddpg * fix tests * fix tests * remove ddpg model * comments * set vf share layers True by default * typo * fix test 2019-07-07 15:06:41 -07:00			`atari-dist-dqn:`
[rllib] Add some more tuned atari results to documentation (#2991) * dqn results ++ * add scale * hour * fix * small dqn table * update * steps * upd * apex * up * add apex results * tip 2018-09-29 23:13:36 -07:00			`env:`
			`grid_search:`
			`- BreakoutNoFrameskip-v4`
			`- BeamRiderNoFrameskip-v4`
			`- QbertNoFrameskip-v4`
			`- SpaceInvadersNoFrameskip-v4`
			`run: DQN`
			`config:`
			`double_q: false`
			`dueling: false`
			`num_atoms: 51`
			`noisy: false`
[RLlib] Replay Buffer API and Ape-X. (#24506) 2022-05-17 13:43:49 +02:00			`replay_buffer_config:`
			`type: MultiAgentReplayBuffer`
			`capacity: 1000000`
[RLlib] Move learning_starts logic from buffers into `training_step()`. (#26032) 2022-08-11 13:07:30 +02:00			`num_steps_sampled_before_learning_starts: 20000`
[rllib] Add some more tuned atari results to documentation (#2991) * dqn results ++ * add scale * hour * fix * small dqn table * update * steps * upd * apex * up * add apex results * tip 2018-09-29 23:13:36 -07:00			`n_step: 1`
			`target_network_update_freq: 8000`
			`lr: .0000625`
			`adam_epsilon: .00015`
			`hiddens: [512]`
[rllib] Rename sample_batch_size => rollout_fragment_length (#7503) * bulk rename * deprecation warn * update doc * update fig * line length * rename * make pytest comptaible * fix test * fi sys * rename * wip * fix more * lint * update svg * comments * lint * fix use of batch steps 2020-03-14 12:05:04 -07:00			`rollout_fragment_length: 4`
[rllib] Add some more tuned atari results to documentation (#2991) * dqn results ++ * add scale * hour * fix * small dqn table * update * steps * upd * apex * up * add apex results * tip 2018-09-29 23:13:36 -07:00			`train_batch_size: 32`
Fix old exploration configs. (#7240) 2020-02-20 17:39:16 +01:00			`exploration_config:`
[RLlib] Exploration API (+EpsilonGreedy sub-class). (#6974) 2020-02-11 00:22:07 +01:00			`epsilon_timesteps: 200000`
			`final_epsilon: 0.01`
[rllib] example and docs on how to use parametric actions with DQN / PG algorithms (#3384) 2018-11-27 23:35:19 -08:00			`num_gpus: 0.2`
[RLlib] Trainer.training_iteration -> Trainer.training_step; Iterations vs reportings: Clarification of terms. (#25076) 2022-06-10 17:09:18 +02:00			`min_sample_timesteps_per_iteration: 10000`