Commit ·
2d74621
1
Parent(s): 32d8bdd
Upload config.json with huggingface_hub
Browse files- config.json +37 -22
config.json
CHANGED
|
@@ -2,11 +2,11 @@
|
|
| 2 |
"help": false,
|
| 3 |
"algo": "APPO",
|
| 4 |
"env": "doom_health_gathering_supreme",
|
| 5 |
-
"experiment": "
|
| 6 |
-
"train_dir": "/
|
| 7 |
"restart_behavior": "resume",
|
| 8 |
"device": "gpu",
|
| 9 |
-
"seed":
|
| 10 |
"num_policies": 1,
|
| 11 |
"async_rl": true,
|
| 12 |
"serial_mode": false,
|
|
@@ -15,9 +15,9 @@
|
|
| 15 |
"worker_num_splits": 2,
|
| 16 |
"policy_workers_per_policy": 1,
|
| 17 |
"max_policy_lag": 1000,
|
| 18 |
-
"num_workers":
|
| 19 |
-
"num_envs_per_worker":
|
| 20 |
-
"batch_size":
|
| 21 |
"num_batches_per_epoch": 1,
|
| 22 |
"num_epochs": 1,
|
| 23 |
"rollout": 32,
|
|
@@ -42,7 +42,7 @@
|
|
| 42 |
"adam_eps": 1e-06,
|
| 43 |
"adam_beta1": 0.9,
|
| 44 |
"adam_beta2": 0.999,
|
| 45 |
-
"max_grad_norm":
|
| 46 |
"learning_rate": 0.0001,
|
| 47 |
"lr_schedule": "constant",
|
| 48 |
"lr_schedule_kl_threshold": 0.008,
|
|
@@ -50,7 +50,7 @@
|
|
| 50 |
"obs_scale": 255.0,
|
| 51 |
"normalize_input": true,
|
| 52 |
"normalize_input_keys": null,
|
| 53 |
-
"decorrelate_experience_max_seconds":
|
| 54 |
"decorrelate_envs_on_one_worker": true,
|
| 55 |
"actor_worker_gpus": [],
|
| 56 |
"set_workers_cpu_affinity": true,
|
|
@@ -61,10 +61,10 @@
|
|
| 61 |
"flush_summaries_interval": 30,
|
| 62 |
"stats_avg": 100,
|
| 63 |
"summaries_use_frameskip": true,
|
| 64 |
-
"heartbeat_interval":
|
| 65 |
-
"heartbeat_reporting_interval":
|
| 66 |
-
"train_for_env_steps":
|
| 67 |
-
"train_for_seconds":
|
| 68 |
"save_every_sec": 120,
|
| 69 |
"keep_checkpoints": 2,
|
| 70 |
"load_checkpoint_kind": "latest",
|
|
@@ -83,10 +83,10 @@
|
|
| 83 |
],
|
| 84 |
"use_rnn": true,
|
| 85 |
"rnn_size": 512,
|
| 86 |
-
"rnn_type": "
|
| 87 |
"rnn_num_layers": 1,
|
| 88 |
"decoder_mlp_layers": [],
|
| 89 |
-
"nonlinearity": "
|
| 90 |
"policy_initialization": "orthogonal",
|
| 91 |
"policy_init_gain": 1.0,
|
| 92 |
"actor_critic_share_weights": true,
|
|
@@ -95,7 +95,6 @@
|
|
| 95 |
"initial_stddev": 1.0,
|
| 96 |
"use_env_info_cache": false,
|
| 97 |
"env_gpu_actions": false,
|
| 98 |
-
"env_gpu_observations": true,
|
| 99 |
"env_frameskip": 4,
|
| 100 |
"env_framestack": 1,
|
| 101 |
"pixel_format": "CHW",
|
|
@@ -128,14 +127,30 @@
|
|
| 128 |
"wide_aspect_ratio": false,
|
| 129 |
"eval_env_frameskip": 1,
|
| 130 |
"fps": 35,
|
| 131 |
-
"command_line": "--
|
| 132 |
"cli_args": {
|
|
|
|
| 133 |
"env": "doom_health_gathering_supreme",
|
| 134 |
-
"
|
| 135 |
-
"
|
| 136 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 137 |
},
|
| 138 |
-
"git_hash": "
|
| 139 |
-
"git_repo_name": "
|
| 140 |
-
"train_script": ".usr.lib.python3.dist-packages.ipykernel_launcher"
|
| 141 |
}
|
|
|
|
| 2 |
"help": false,
|
| 3 |
"algo": "APPO",
|
| 4 |
"env": "doom_health_gathering_supreme",
|
| 5 |
+
"experiment": "vizdoom_doom_health_gathering_supreme_2222",
|
| 6 |
+
"train_dir": "/scratch/sample_factory/train_dir/vizdoom",
|
| 7 |
"restart_behavior": "resume",
|
| 8 |
"device": "gpu",
|
| 9 |
+
"seed": 2222,
|
| 10 |
"num_policies": 1,
|
| 11 |
"async_rl": true,
|
| 12 |
"serial_mode": false,
|
|
|
|
| 15 |
"worker_num_splits": 2,
|
| 16 |
"policy_workers_per_policy": 1,
|
| 17 |
"max_policy_lag": 1000,
|
| 18 |
+
"num_workers": 20,
|
| 19 |
+
"num_envs_per_worker": 12,
|
| 20 |
+
"batch_size": 2048,
|
| 21 |
"num_batches_per_epoch": 1,
|
| 22 |
"num_epochs": 1,
|
| 23 |
"rollout": 32,
|
|
|
|
| 42 |
"adam_eps": 1e-06,
|
| 43 |
"adam_beta1": 0.9,
|
| 44 |
"adam_beta2": 0.999,
|
| 45 |
+
"max_grad_norm": 0.0,
|
| 46 |
"learning_rate": 0.0001,
|
| 47 |
"lr_schedule": "constant",
|
| 48 |
"lr_schedule_kl_threshold": 0.008,
|
|
|
|
| 50 |
"obs_scale": 255.0,
|
| 51 |
"normalize_input": true,
|
| 52 |
"normalize_input_keys": null,
|
| 53 |
+
"decorrelate_experience_max_seconds": 1,
|
| 54 |
"decorrelate_envs_on_one_worker": true,
|
| 55 |
"actor_worker_gpus": [],
|
| 56 |
"set_workers_cpu_affinity": true,
|
|
|
|
| 61 |
"flush_summaries_interval": 30,
|
| 62 |
"stats_avg": 100,
|
| 63 |
"summaries_use_frameskip": true,
|
| 64 |
+
"heartbeat_interval": 10,
|
| 65 |
+
"heartbeat_reporting_interval": 300,
|
| 66 |
+
"train_for_env_steps": 10000000000,
|
| 67 |
+
"train_for_seconds": 3600000,
|
| 68 |
"save_every_sec": 120,
|
| 69 |
"keep_checkpoints": 2,
|
| 70 |
"load_checkpoint_kind": "latest",
|
|
|
|
| 83 |
],
|
| 84 |
"use_rnn": true,
|
| 85 |
"rnn_size": 512,
|
| 86 |
+
"rnn_type": "lstm",
|
| 87 |
"rnn_num_layers": 1,
|
| 88 |
"decoder_mlp_layers": [],
|
| 89 |
+
"nonlinearity": "relu",
|
| 90 |
"policy_initialization": "orthogonal",
|
| 91 |
"policy_init_gain": 1.0,
|
| 92 |
"actor_critic_share_weights": true,
|
|
|
|
| 95 |
"initial_stddev": 1.0,
|
| 96 |
"use_env_info_cache": false,
|
| 97 |
"env_gpu_actions": false,
|
|
|
|
| 98 |
"env_frameskip": 4,
|
| 99 |
"env_framestack": 1,
|
| 100 |
"pixel_format": "CHW",
|
|
|
|
| 127 |
"wide_aspect_ratio": false,
|
| 128 |
"eval_env_frameskip": 1,
|
| 129 |
"fps": 35,
|
| 130 |
+
"command_line": "--train_for_seconds=3600000 --algo=APPO --gamma=0.99 --use_rnn=True --num_workers=20 --num_envs_per_worker=12 --num_epochs=1 --rollout=32 --recurrence=32 --batch_size=2048 --benchmark=False --max_grad_norm=0.0 --decorrelate_experience_max_seconds=1 --nonlinearity=relu --rnn_type=lstm --num_policies=1 --heartbeat_reporting_interval=300 --train_dir=/scratch/sample_factory/train_dir/vizdoom --seed=2222 --experiment=vizdoom_doom_health_gathering_supreme_2222 --env=doom_health_gathering_supreme",
|
| 131 |
"cli_args": {
|
| 132 |
+
"algo": "APPO",
|
| 133 |
"env": "doom_health_gathering_supreme",
|
| 134 |
+
"experiment": "vizdoom_doom_health_gathering_supreme_2222",
|
| 135 |
+
"train_dir": "/scratch/sample_factory/train_dir/vizdoom",
|
| 136 |
+
"seed": 2222,
|
| 137 |
+
"num_policies": 1,
|
| 138 |
+
"num_workers": 20,
|
| 139 |
+
"num_envs_per_worker": 12,
|
| 140 |
+
"batch_size": 2048,
|
| 141 |
+
"num_epochs": 1,
|
| 142 |
+
"rollout": 32,
|
| 143 |
+
"recurrence": 32,
|
| 144 |
+
"gamma": 0.99,
|
| 145 |
+
"max_grad_norm": 0.0,
|
| 146 |
+
"decorrelate_experience_max_seconds": 1,
|
| 147 |
+
"heartbeat_reporting_interval": 300,
|
| 148 |
+
"train_for_seconds": 3600000,
|
| 149 |
+
"benchmark": false,
|
| 150 |
+
"use_rnn": true,
|
| 151 |
+
"rnn_type": "lstm",
|
| 152 |
+
"nonlinearity": "relu"
|
| 153 |
},
|
| 154 |
+
"git_hash": "9da68b57eecd73c3c884c1be2d938b46aa7a7f49",
|
| 155 |
+
"git_repo_name": "https://github.com/alex-petrenko/sample-factory.git"
|
|
|
|
| 156 |
}
|