andrewzhang505
commited on
Commit
•
5b0f9a3
1
Parent(s):
bb961c9
Upload cfg.json with huggingface_hub
Browse files
cfg.json
ADDED
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"help": false,
|
3 |
+
"algo": "APPO",
|
4 |
+
"env": "mujoco_humanoid",
|
5 |
+
"experiment": "10_mujoco_all_envs_see_2222_env_mujoco_humanoid",
|
6 |
+
"train_dir": "./train_dir",
|
7 |
+
"experiments_root": "mujoco_all_envs/mujoco_all_envs_crl_4",
|
8 |
+
"device": "gpu",
|
9 |
+
"seed": 2222,
|
10 |
+
"num_policies": 1,
|
11 |
+
"async_rl": false,
|
12 |
+
"serial_mode": false,
|
13 |
+
"batched_sampling": false,
|
14 |
+
"num_batches_to_accumulate": 2,
|
15 |
+
"worker_num_splits": 2,
|
16 |
+
"policy_workers_per_policy": 1,
|
17 |
+
"max_policy_lag": 10000,
|
18 |
+
"num_workers": 8,
|
19 |
+
"num_envs_per_worker": 8,
|
20 |
+
"batch_size": 1024,
|
21 |
+
"num_batches_per_epoch": 4,
|
22 |
+
"num_epochs": 2,
|
23 |
+
"rollout": 64,
|
24 |
+
"recurrence": 1,
|
25 |
+
"shuffle_minibatches": false,
|
26 |
+
"gamma": 0.99,
|
27 |
+
"reward_scale": 1,
|
28 |
+
"reward_clip": 1000.0,
|
29 |
+
"value_bootstrap": false,
|
30 |
+
"normalize_returns": true,
|
31 |
+
"exploration_loss_coeff": 0.0,
|
32 |
+
"value_loss_coeff": 1.3,
|
33 |
+
"kl_loss_coeff": 0.1,
|
34 |
+
"exploration_loss": "entropy",
|
35 |
+
"gae_lambda": 0.95,
|
36 |
+
"ppo_clip_ratio": 0.2,
|
37 |
+
"ppo_clip_value": 1.0,
|
38 |
+
"with_vtrace": false,
|
39 |
+
"vtrace_rho": 1.0,
|
40 |
+
"vtrace_c": 1.0,
|
41 |
+
"optimizer": "adam",
|
42 |
+
"adam_eps": 1e-06,
|
43 |
+
"adam_beta1": 0.9,
|
44 |
+
"adam_beta2": 0.999,
|
45 |
+
"max_grad_norm": 3.5,
|
46 |
+
"learning_rate": 0.00295,
|
47 |
+
"lr_schedule": "linear_decay",
|
48 |
+
"lr_schedule_kl_threshold": 0.008,
|
49 |
+
"obs_subtract_mean": 0.0,
|
50 |
+
"obs_scale": 1.0,
|
51 |
+
"normalize_input": true,
|
52 |
+
"decorrelate_experience_max_seconds": 10,
|
53 |
+
"decorrelate_envs_on_one_worker": true,
|
54 |
+
"actor_worker_gpus": [],
|
55 |
+
"set_workers_cpu_affinity": true,
|
56 |
+
"force_envs_single_thread": true,
|
57 |
+
"default_niceness": 0,
|
58 |
+
"experiment_summaries_interval": 3,
|
59 |
+
"stats_avg": 100,
|
60 |
+
"train_for_env_steps": 10000000,
|
61 |
+
"train_for_seconds": 10000000000,
|
62 |
+
"save_every_sec": 15,
|
63 |
+
"keep_checkpoints": 3,
|
64 |
+
"load_checkpoint_kind": "latest",
|
65 |
+
"save_milestones_sec": -1,
|
66 |
+
"save_best_every_sec": 5,
|
67 |
+
"save_best_metric": "reward",
|
68 |
+
"save_best_after": 100000,
|
69 |
+
"benchmark": false,
|
70 |
+
"encoder_type": "mlp",
|
71 |
+
"encoder_subtype": "mlp_mujoco",
|
72 |
+
"encoder_custom": null,
|
73 |
+
"encoder_extra_fc_layers": 0,
|
74 |
+
"hidden_size": 64,
|
75 |
+
"nonlinearity": "tanh",
|
76 |
+
"policy_initialization": "torch_default",
|
77 |
+
"policy_init_gain": 1.0,
|
78 |
+
"actor_critic_share_weights": true,
|
79 |
+
"adaptive_stddev": false,
|
80 |
+
"initial_stddev": 1.0,
|
81 |
+
"use_rnn": false,
|
82 |
+
"rnn_type": "gru",
|
83 |
+
"rnn_num_layers": 1,
|
84 |
+
"env_gpu_actions": false,
|
85 |
+
"env_frameskip": 1,
|
86 |
+
"env_framestack": 4,
|
87 |
+
"pixel_format": "CHW",
|
88 |
+
"with_wandb": true,
|
89 |
+
"wandb_user": null,
|
90 |
+
"wandb_project": "sample_factory",
|
91 |
+
"wandb_group": null,
|
92 |
+
"wandb_job_type": "SF",
|
93 |
+
"wandb_tags": [
|
94 |
+
"mujoco",
|
95 |
+
"runner_crl_4"
|
96 |
+
],
|
97 |
+
"command_line": "--algo=APPO --with_wandb=True --wandb_tags mujoco runner_crl_4 --seed=2222 --env=mujoco_humanoid --experiment=10_mujoco_all_envs_see_2222_env_mujoco_humanoid --train_dir=./train_dir --experiments_root=mujoco_all_envs/mujoco_all_envs_crl_4",
|
98 |
+
"cli_args": {
|
99 |
+
"algo": "APPO",
|
100 |
+
"env": "mujoco_humanoid",
|
101 |
+
"experiment": "10_mujoco_all_envs_see_2222_env_mujoco_humanoid",
|
102 |
+
"train_dir": "./train_dir",
|
103 |
+
"experiments_root": "mujoco_all_envs/mujoco_all_envs_crl_4",
|
104 |
+
"seed": 2222,
|
105 |
+
"with_wandb": true,
|
106 |
+
"wandb_tags": [
|
107 |
+
"mujoco",
|
108 |
+
"runner_crl_4"
|
109 |
+
]
|
110 |
+
},
|
111 |
+
"git_hash": "4e69a02b57fff18dbbf96054866d0d759f70c5fa",
|
112 |
+
"git_repo_name": "https://github.com/andrewzhang505/sample-factory.git"
|
113 |
+
}
|