andrewzhang505

Andrewzhang505's group workspace

Group: mujoco_pendulum

1-1

of 1

Tags

async

mujoco

Author

andrewzhang505

State

Finished

Start time

July 13th, 2022 10:54:48 PM

Runtime

8m 7s

Tracked hours

7m 57s

Run path

andrewzhang505/sample_factory/83_mujoco_all_envs_see_4444_env_mujoco_pendulum_a.rl_True_20220713_155446_990242

Linux-5.13.0-35-generic-x86_64-with-glibc2.31

Python version

3.10.5

Command

-m sf_examples.mujoco_examples.train_mujoco --algo=APPO --with_wandb=True --wandb_tags mujoco async --seed=4444 --env=mujoco_pendulum --async_rl=True --experiment=83_mujoco_all_envs_see_4444_env_mujoco_pendulum_a.rl_True --train_dir=./train_dir/mujoco_all_envs/mujoco_all_envs_slurm

System Hardware

CPU count	72
GPU count	1
GPU type	NVIDIA GeForce RTX 2080 Ti

W&B CLI Version

0.12.21

Group

mujoco_pendulum

Job Type

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 110 keys
- actor_critic_share_weights:
  true
- actor_worker_gpus:[] 0 items
- adam_beta1:
  0.9
- adam_beta2:
  0.999
- adam_eps:
  0.000001
- adaptive_stddev:
  false
- algo:
  "APPO"
- async_rl:
  true
- batch_size:
  1,024
- batched_sampling:
  false
- benchmark:
  false
- ▶
  cli_args:{} 8 keys
- command_line:
  "--algo=APPO --with_wandb=True --wandb_tags mujoco async --seed=4444 --env=mujoco_pendulum --async_rl=True --experiment=83_mujoco_all_envs_see_4444_env_mujoco_pendulum_a.rl_True --train_dir=./train_dir/mujoco_all_envs/mujoco_all_envs_slurm"
- decorrelate_envs_on_one_worker:
  true
- decorrelate_experience_max_seconds:
  10
- default_niceness:
  0
- device:
  "gpu"
- encoder_custom:
  null
- encoder_extra_fc_layers:
  0
- encoder_subtype:
  "mlp_mujoco"
- encoder_type:
  "mlp"
- env:
  "mujoco_pendulum"
- env_frameskip:
  1
- env_framestack:
  4
- env_gpu_actions:
  false
- experiment:
  "83_mujoco_all_envs_see_4444_env_mujoco_pendulum_a.rl_True"
- experiment_summaries_interval:
  3
- exploration_loss:
  "entropy"
- exploration_loss_coeff:
  0
- force_envs_single_thread:
  true
- gae_lambda:
  0.95
- gamma:
  0.99
- git_hash:
  "unknown"
- git_repo_name:
  "not a git repository"
- help:
  false
- hidden_size:
  64
- initial_stddev:
  1
- keep_checkpoints:
  3
- kl_loss_coeff:
  0.1
- learning_rate:
  0.00295
- load_checkpoint_kind:
  "latest"
- lr_schedule:
  "linear_decay"
- lr_schedule_kl_threshold:
  0.008
- max_grad_norm:
  3.5
- max_policy_lag:
  10,000
- nonlinearity:
  "tanh"
- with_pbt:
  false
- with_vtrace:
  false
- with_wandb:
  true
- worker_num_splits:
  2

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 63 keys
- global_step:
  9,973,760
- len/len:
  1,000
- len/len_max:
  1,000
- len/len_min:
  1,000
- perf/_fps:
  20,889.328125
- perf/_sample_throughput:
  42,933.88671875
- policy_stats/avg_true_objective:
  1,000
- policy_stats/avg_true_objective_max:
  1,000
- policy_stats/avg_true_objective_min:
  1,000
- reward/reward:
  1,000
- reward/reward_max:
  1,000
- reward/reward_min:
  1,000
- stats/avg_request_count:
  7.980000019073486
- stats/gpu_cache_learner:
  4.1943039894104
- stats/gpu_cache_policy_worker:
  2.0971519947052
- stats/gpu_mem_learner:
  0.809984028339386
- stats/gpu_mem_policy_worker:
  0.025087999179959297
- stats/master_process_memory_mb:
  329.65234375
- stats/memory_learner:
  2,792
- stats/memory_policy_worker:
  2,798.43359375
- stats/step_policy:
  0.001359682064503431
- stats/wait_policy:
  0.00003721714165294543
- train/action_log_std_max:
  -3.5278549194335938
- train/action_log_std_mean:
  -3.5278549194335938
- train/action_log_std_min:
  -3.5278549194335938
- train/action_mean:
  -0.0006495509296655655
- train/action_mean_max:
  0.07486190646886826
- train/action_mean_min:
  -0.06963923573493958
- train/action_stddev_max:
  0.02936784364283085
- train/action_stddev_mean:
  0.02936784364283085
- train/action_stddev_min:
  0.02936784364283085
- train/adam_max_second_moment:
  0.73076331615448
- train/adv_max:
  0.18843193352222443
- train/adv_min:
  -12.895159721374512
- train/adv_std:
  7.072385787963867
- train/entropy:
  -2.1089162826538086
- train/exploration_loss:
  0
- train/fraction_clipped:
  0.005859375
- train/grad_norm:
  3.4999990463256836
- train/kl_divergence:
  0.001021054806187749
- train/kl_divergence_max:
  0.008571564219892025
- train/kl_loss:
  0.00010210548498434946
- train/loss:
  0.347989022731781
- train/lr:
  0.00014651817036792636
- train/max_abs_logprob:
  3.5273101329803467
- train/num_sgd_steps:
  8
- train/value_loss:
  0.34880176186561584
- train/version_diff_avg:
  11.1875
- train/version_diff_max:
  15
- train/version_diff_min:
  7

This run produced these artifacts as outputs. Total: 1. Learn more

wandb-history

run-83_mujoco_all_envs_see_4444_env_mujoco_pendulum_a.rl_True_20220713_155446_990242-history:v0