dmlab_30_1111 / cfg.json
edbeeching's picture
edbeeching HF staff
Upload cfg.json with huggingface_hub
e15deaf
{
"help": false,
"algo": "APPO",
"env": "dmlab_30",
"experiment": "dmlab30_1111",
"train_dir": "/code/train_dir",
"restart_behavior": "resume",
"device": "gpu",
"seed": 1111,
"num_policies": 1,
"async_rl": true,
"serial_mode": false,
"batched_sampling": false,
"num_batches_to_accumulate": 2,
"worker_num_splits": 2,
"policy_workers_per_policy": 1,
"max_policy_lag": 1000,
"num_workers": 60,
"num_envs_per_worker": 12,
"batch_size": 2048,
"num_batches_per_epoch": 1,
"num_epochs": 1,
"rollout": 32,
"recurrence": 32,
"shuffle_minibatches": false,
"gamma": 0.99,
"reward_scale": 1.0,
"reward_clip": 1000.0,
"value_bootstrap": false,
"normalize_returns": true,
"exploration_loss_coeff": 0.003,
"value_loss_coeff": 0.5,
"kl_loss_coeff": 0.0,
"exploration_loss": "entropy",
"gae_lambda": 0.95,
"ppo_clip_ratio": 0.1,
"ppo_clip_value": 1.0,
"with_vtrace": false,
"vtrace_rho": 1.0,
"vtrace_c": 1.0,
"optimizer": "adam",
"adam_eps": 1e-06,
"adam_beta1": 0.9,
"adam_beta2": 0.999,
"max_grad_norm": 0.0,
"learning_rate": 0.0001,
"lr_schedule": "constant",
"lr_schedule_kl_threshold": 0.008,
"obs_subtract_mean": 0.0,
"obs_scale": 255.0,
"normalize_input": true,
"normalize_input_keys": [
"obs"
],
"decorrelate_experience_max_seconds": 1,
"decorrelate_envs_on_one_worker": true,
"actor_worker_gpus": [],
"set_workers_cpu_affinity": true,
"force_envs_single_thread": false,
"default_niceness": 0,
"log_to_file": true,
"experiment_summaries_interval": 10,
"flush_summaries_interval": 30,
"stats_avg": 100,
"summaries_use_frameskip": true,
"heartbeat_interval": 10,
"heartbeat_reporting_interval": 300,
"train_for_env_steps": 10000000000,
"train_for_seconds": 3600000,
"save_every_sec": 120,
"keep_checkpoints": 2,
"load_checkpoint_kind": "latest",
"save_milestones_sec": 3600,
"save_best_every_sec": 5,
"save_best_metric": "reward",
"save_best_after": 100000,
"benchmark": false,
"encoder_mlp_layers": [
512,
512
],
"encoder_conv_architecture": "resnet_impala",
"encoder_conv_mlp_layers": [
512
],
"use_rnn": true,
"rnn_size": 256,
"rnn_type": "lstm",
"rnn_num_layers": 1,
"decoder_mlp_layers": [],
"nonlinearity": "relu",
"policy_initialization": "orthogonal",
"policy_init_gain": 1.0,
"actor_critic_share_weights": true,
"adaptive_stddev": true,
"continuous_tanh_scale": 0.0,
"initial_stddev": 1.0,
"use_env_info_cache": false,
"env_gpu_actions": false,
"env_frameskip": 4,
"env_framestack": 1,
"pixel_format": "CHW",
"use_record_episode_statistics": false,
"with_wandb": false,
"wandb_user": null,
"wandb_project": "sample_factory",
"wandb_group": null,
"wandb_job_type": "SF",
"wandb_tags": [],
"with_pbt": false,
"pbt_mix_policies_in_one_env": true,
"pbt_period_env_steps": 5000000,
"pbt_start_mutation": 20000000,
"pbt_replace_fraction": 0.3,
"pbt_mutation_rate": 0.15,
"pbt_replace_reward_gap": 0.1,
"pbt_replace_reward_gap_absolute": 1e-06,
"pbt_optimize_batch_size": false,
"pbt_optimize_gamma": false,
"pbt_target_objective": "true_objective",
"pbt_perturb_min": 1.05,
"pbt_perturb_max": 1.5,
"res_w": 96,
"res_h": 72,
"dmlab_throughput_benchmark": false,
"dmlab_renderer": "software",
"dmlab_gpus": [
0
],
"dmlab30_dataset": "/datasets/dmlab/brady_konkle_oliva2008",
"dmlab_with_instructions": true,
"dmlab_extended_action_set": true,
"dmlab_use_level_cache": true,
"dmlab_level_cache_path": "/datasets/dmlab/.dmcache",
"dmlab_one_task_per_worker": true,
"command_line": "--train_for_seconds=3600000 --algo=APPO --gamma=0.99 --use_rnn=True --num_workers=60 --num_envs_per_worker=12 --num_epochs=1 --rollout=32 --recurrence=32 --batch_size=2048 --benchmark=False --max_grad_norm=0.0 --dmlab_renderer=software --decorrelate_experience_max_seconds=1 --encoder_conv_architecture=resnet_impala --encoder_conv_mlp_layers 512 --nonlinearity=relu --rnn_type=lstm --dmlab_extended_action_set=True --num_policies=1 --dmlab30_dataset=/datasets/dmlab/brady_konkle_oliva2008 --dmlab_use_level_cache=True --dmlab_level_cache_path=/datasets/dmlab/.dmcache --env=dmlab_30 --normalize_input=True --normalize_input_keys obs --normalize_returns=True --with_vtrace=False --dmlab_one_task_per_worker=True --save_milestones_sec=3600 --heartbeat_reporting_interval=300 --seed=1111 --experiment=dmlab30_1111",
"cli_args": {
"algo": "APPO",
"env": "dmlab_30",
"experiment": "dmlab30_1111",
"seed": 1111,
"num_policies": 1,
"num_workers": 60,
"num_envs_per_worker": 12,
"batch_size": 2048,
"num_epochs": 1,
"rollout": 32,
"recurrence": 32,
"gamma": 0.99,
"normalize_returns": true,
"with_vtrace": false,
"max_grad_norm": 0.0,
"normalize_input": true,
"normalize_input_keys": [
"obs"
],
"decorrelate_experience_max_seconds": 1,
"heartbeat_reporting_interval": 300,
"train_for_seconds": 3600000,
"save_milestones_sec": 3600,
"benchmark": false,
"encoder_conv_architecture": "resnet_impala",
"encoder_conv_mlp_layers": [
512
],
"use_rnn": true,
"rnn_type": "lstm",
"nonlinearity": "relu",
"dmlab_renderer": "software",
"dmlab30_dataset": "/datasets/dmlab/brady_konkle_oliva2008",
"dmlab_extended_action_set": true,
"dmlab_use_level_cache": true,
"dmlab_level_cache_path": "/datasets/dmlab/.dmcache",
"dmlab_one_task_per_worker": true
},
"git_hash": "e6123b3720d9aec139ae7d01970ddc109c198669",
"git_repo_name": "https://github.com/alex-petrenko/sample-factory.git"
}