ray/rllib/tuned_examples/regression_tests/pendulum-sac.yaml

pendulum-sac:
    env: Pendulum-v0
    run: SAC
    stop:
        episode_reward_mean: -300  # note that evaluation perf is higher
        timesteps_total: 15000
    config:
        evaluation_interval: 1  # logged under evaluation/* metric keys
        soft_horizon: true
        metrics_smoothing_episodes: 10
[rllib] Feature/soft actor critic v2 (#5328) * Add base for Soft Actor-Critic * Pick changes from old SAC branch * Update sac.py * First implementation of sac model * Remove unnecessary SAC imports * Prune unnecessary noise and exploration code * Implement SAC model and use that in SAC policy * runs but doesn't learn * clear state * fix batch size * Add missing alpha grads and vars * -200 by 2k timesteps * doc * lazy squash * one file * ignore tfp * revert done 2019-08-01 23:37:36 -07:00			`pendulum-sac:`
			`env: Pendulum-v0`
			`run: SAC`
			`stop:`
			`episode_reward_mean: -300 # note that evaluation perf is higher`
			`timesteps_total: 15000`
			`config:`
			`evaluation_interval: 1 # logged under evaluation/* metric keys`
			`soft_horizon: true`
			`metrics_smoothing_episodes: 10`