Skip to main content

Regression Report: ppo_atari_envpool_xla_jax_scan

[['?we=openrlbenchmark&wpn=sb3&ceik=env&cen=algo&metric=rollout/ep_rew_mean', 'ppo', 'ppo_lstm'], ['?we=tianshou&wpn=atari.benchmark&ceik=task&cen=algo_name&metric=test/reward', 'iqn', 'ppo', 'rainbow', 'fqf', 'c51', 'dqn', 'qrdqn'], ['?we=openrlbenchmark&wpn=baselines&ceik=env&cen=exp_name&metric=charts/episodic_return', 'baselines-ppo2-cnn'], ['?we=openrlbenchmark&wpn=cleanrl&ceik=env_id&cen=exp_name&metric=charts/avg_episodic_return', 'ppo_atari_envpool_xla_jax_scan?tag=pr-328']]
Created on January 2|Last edited on January 2

Computing group metrics from first 50 groups
2M4M6M8MSteps100200300400500Episodic Return
Computing group metrics from first 50 groups
510152025Time (hours)100200300400500600Episodic Return
openrlbenchmark/sb3/ppo ({})
10
openrlbenchmark/sb3/ppo_lstm ({})
14
tianshou/atari.benchmark/iqn ({})
10
tianshou/atari.benchmark/ppo ({})
10
tianshou/atari.benchmark/rainbow ({})
10
tianshou/atari.benchmark/fqf ({})
10
tianshou/atari.benchmark/c51 ({})
10
tianshou/atari.benchmark/dqn ({})
10
tianshou/atari.benchmark/qrdqn ({})
10
openrlbenchmark/baselines/baselines-ppo2-cnn ({})
5
openrlbenchmark/cleanrl/ppo_atari_envpool_xla_jax_scan ({'tag': ['pr-328']})
3



openrlbenchmark/sb3/ppo ({})
10
openrlbenchmark/sb3/ppo_lstm ({})
11
tianshou/atari.benchmark/iqn ({})
10
tianshou/atari.benchmark/ppo ({})
10
tianshou/atari.benchmark/rainbow ({})
10
tianshou/atari.benchmark/fqf ({})
10
tianshou/atari.benchmark/c51 ({})
10
tianshou/atari.benchmark/dqn ({})
10
tianshou/atari.benchmark/qrdqn ({})
10
openrlbenchmark/baselines/baselines-ppo2-cnn ({})
5
openrlbenchmark/cleanrl/ppo_atari_envpool_xla_jax_scan ({'tag': ['pr-328']})
3