Update config.yaml

dexteroushandenvs/cfg/macpo/config.yaml

@@ -4,7 +4,7 @@ experiment_name: check
 run_dir: ./logs
 use_centralized_V: True
 use_obs_instead_of_state: False
-num_env_steps: 50000000
+num_env_steps: 100000000
 episode_length: 8
 n_rollout_threads: 80
 n_eval_rollout_threads: 1
@@ -20,13 +20,13 @@ eval_interval: 25
 log_interval: 25
 eval_episodes: 10000
 
-safety_bound: 20
+safety_bound: 2.5
 EPS: 1.e-8
 safety_gamma: 0.09
 line_search_fraction: 0.5
 g_step_dir_coef: 0.1
 b_step_dir_coef: 0.1
-fraction_coef: 0.27
+fraction_coef: 0.29
 
 gamma: 0.96
 gae_lambda: 0.95
@@ -35,7 +35,7 @@ use_popart: True
 use_valuenorm: False
 use_proper_time_limits: False
 
-kl_threshold: 0.0065
+kl_threshold: 0.0067
 ls_step: 10
 accept_ratio: 0.5
 clip_param: 0.2
@@ -54,8 +54,8 @@ use_huber_loss: True
 use_value_active_masks: False
 use_policy_active_masks: False
 
-lr: 5.e-4
-critic_lr: 5.e-4
+lr: 3.e-5
+critic_lr: 1.e-3
 opti_eps: 1.e-5
 weight_decay: 0.0