5aecf5a163fca64841e89662f8e9d4939171ed7e,benchmarks/src/garage_benchmarks/experiments/policies/categorical_lstm_policy.py,,categorical_lstm_policy,#Any#Any#Any#,14

Before Change

            ),
        )

        trainer.setup(algo, env, sampler_args=dict(n_envs=12))
        trainer.train(n_epochs=488, batch_size=2048)

After Change

        baseline = LinearFeatureBaseline(env_spec=env.spec)

        sampler = RaySampler(agents=policy,
                             envs=env,
                             max_episode_length=env.spec.max_episode_length,
                             is_tf_worker=True)

        algo = PPO(
            env_spec=env.spec,
            policy=policy,
            baseline=baseline,
            sampler=sampler,

Instances