From 6043ce710dd9fedadf7a54e0f4003665c0057d35 Mon Sep 17 00:00:00 2001 From: Sven Mika Date: Thu, 20 Feb 2020 17:39:16 +0100 Subject: [PATCH] Fix old exploration configs. (#7240) --- rllib/agents/dqn/apex.py | 2 +- rllib/tests/test_ignore_worker_failure.py | 2 +- rllib/tuned_examples/atari-apex.yaml | 2 +- rllib/tuned_examples/atari-dist-dqn.yaml | 2 +- rllib/tuned_examples/atari-dqn.yaml | 2 +- rllib/tuned_examples/atari-duel-ddqn.yaml | 2 +- rllib/tuned_examples/compact-regression-test.yaml | 4 ++-- rllib/tuned_examples/pong-dqn.yaml | 2 +- rllib/tuned_examples/pong-rainbow.yaml | 2 +- 9 files changed, 10 insertions(+), 10 deletions(-) diff --git a/rllib/agents/dqn/apex.py b/rllib/agents/dqn/apex.py index ac2c91c21..fc5c3e677 100644 --- a/rllib/agents/dqn/apex.py +++ b/rllib/agents/dqn/apex.py @@ -22,7 +22,7 @@ APEX_DEFAULT_CONFIG = merge_dicts( "sample_batch_size": 50, "target_network_update_freq": 500000, "timesteps_per_iteration": 25000, - "exploration": {"type": "PerWorkerEpsilonGreedy"}, + "exploration_config": {"type": "PerWorkerEpsilonGreedy"}, "worker_side_prioritization": True, "min_iter_time_s": 30, }, diff --git a/rllib/tests/test_ignore_worker_failure.py b/rllib/tests/test_ignore_worker_failure.py index 807990653..7cc9edbd3 100644 --- a/rllib/tests/test_ignore_worker_failure.py +++ b/rllib/tests/test_ignore_worker_failure.py @@ -72,7 +72,7 @@ class IgnoresWorkerFailure(unittest.TestCase): "timesteps_per_iteration": 1000, "num_gpus": 0, "min_iter_time_s": 1, - "exploration": False, + "explore": False, "learning_starts": 1000, "target_network_update_freq": 100, "optimizer": { diff --git a/rllib/tuned_examples/atari-apex.yaml b/rllib/tuned_examples/atari-apex.yaml index 4785c4b30..1ca1613be 100644 --- a/rllib/tuned_examples/atari-apex.yaml +++ b/rllib/tuned_examples/atari-apex.yaml @@ -17,7 +17,7 @@ apex: adam_epsilon: .00015 hiddens: [512] buffer_size: 1000000 - exploration: + exploration_config: final_epsilon: 0.01 epsilon_timesteps: 200000 prioritized_replay_alpha: 0.5 diff --git a/rllib/tuned_examples/atari-dist-dqn.yaml b/rllib/tuned_examples/atari-dist-dqn.yaml index ab2480a04..05e8b8eca 100644 --- a/rllib/tuned_examples/atari-dist-dqn.yaml +++ b/rllib/tuned_examples/atari-dist-dqn.yaml @@ -21,7 +21,7 @@ atari-dist-dqn: buffer_size: 1000000 sample_batch_size: 4 train_batch_size: 32 - exploration: + exploration_config: epsilon_timesteps: 200000 final_epsilon: 0.01 prioritized_replay_alpha: 0.5 diff --git a/rllib/tuned_examples/atari-dqn.yaml b/rllib/tuned_examples/atari-dqn.yaml index 6ebaf4906..672e0d42d 100644 --- a/rllib/tuned_examples/atari-dqn.yaml +++ b/rllib/tuned_examples/atari-dqn.yaml @@ -23,7 +23,7 @@ atari-basic-dqn: buffer_size: 1000000 sample_batch_size: 4 train_batch_size: 32 - exploration: + exploration_config: epsilon_timesteps: 200000 final_epsilon: 0.01 prioritized_replay_alpha: 0.5 diff --git a/rllib/tuned_examples/atari-duel-ddqn.yaml b/rllib/tuned_examples/atari-duel-ddqn.yaml index 93cfd1e0d..6818c9477 100644 --- a/rllib/tuned_examples/atari-duel-ddqn.yaml +++ b/rllib/tuned_examples/atari-duel-ddqn.yaml @@ -23,7 +23,7 @@ dueling-ddqn: buffer_size: 1000000 sample_batch_size: 4 train_batch_size: 32 - exploration: + exploration_config: epsilon_timesteps: 200000 final_epsilon: 0.01 prioritized_replay_alpha: 0.5 diff --git a/rllib/tuned_examples/compact-regression-test.yaml b/rllib/tuned_examples/compact-regression-test.yaml index 1a10e24b8..2c5210548 100644 --- a/rllib/tuned_examples/compact-regression-test.yaml +++ b/rllib/tuned_examples/compact-regression-test.yaml @@ -85,7 +85,7 @@ apex: adam_epsilon: .00015 hiddens: [512] buffer_size: 1000000 - exploration: + exploration_config: epsilon_timesteps: 200000 final_epsilon: 0.01 prioritized_replay_alpha: 0.5 @@ -135,7 +135,7 @@ atari-basic-dqn: buffer_size: 1000000 sample_batch_size: 4 train_batch_size: 32 - exploration: + exploration_config: epsilon_timesteps: 200000 final_epsilon: 0.01 prioritized_replay_alpha: 0.5 diff --git a/rllib/tuned_examples/pong-dqn.yaml b/rllib/tuned_examples/pong-dqn.yaml index a9b709eed..7bbdd602d 100644 --- a/rllib/tuned_examples/pong-dqn.yaml +++ b/rllib/tuned_examples/pong-dqn.yaml @@ -13,7 +13,7 @@ pong-deterministic-dqn: buffer_size: 50000 sample_batch_size: 4 train_batch_size: 32 - exploration: + exploration_config: epsilon_timesteps: 200000 final_epsilon: .01 model: diff --git a/rllib/tuned_examples/pong-rainbow.yaml b/rllib/tuned_examples/pong-rainbow.yaml index e76694cd3..984043eac 100644 --- a/rllib/tuned_examples/pong-rainbow.yaml +++ b/rllib/tuned_examples/pong-rainbow.yaml @@ -13,7 +13,7 @@ pong-deterministic-rainbow: buffer_size: 50000 sample_batch_size: 4 train_batch_size: 32 - exploration: + exploration_config: epsilon_timesteps: 2 final_epsilon: 0.0 target_network_update_freq: 500