single-shot-robot

Josselin's group workspace

Group: CARE

1-4

of 4

Tags

Notes

Author

josselin

State

Finished

Start time

September 13th, 2023 11:54:32 AM

Runtime

5h 34m 24s

Tracked hours

5h 33m 38s

Run path

single-shot-robot/PTSL_MTRL50-1M/ugw9wvu7

Linux-5.15.0-83-generic-x86_64-with-glibc2.10

Python version

3.8.0

Git repository

git clone https://github.com/JosselinSomervilleRoberts/mtrl.git

Git state

git checkout -b "CARE_4" ea3e61c6ee556dec14f1592437e0c08362d94677

Command

main.py setup=metaworld env=metaworld-mt50 agent=state_sac agent.actor.hidden_dim=400 agent.actor.num_layers=3 agent.multitask.num_envs=50 experiment.num_eval_episodes=10 experiment.num_train_steps=1000000 setup.seed_ref=4 setup.num_seeds=1 setup.name=CARE replay_buffer.batch_size=1280 agent.encoder.type_to_select=moe agent.encoder.moe.task_id_to_encoder_id_cfg.mode=attention agent.encoder.moe.num_experts=4 agent.multitask.num_envs=50 agent.multitask.should_use_disentangled_alpha=True agent.multitask.should_use_task_encoder=True agent.multitask.should_use_multi_head_policy=False agent.multitask.task_encoder_cfg.model_cfg.pretrained_embedding_cfg.should_use=True

System Hardware

CPU count	8
Logical CPU count	16
GPU count	1
GPU type	NVIDIA RTX A3000 Laptop GPU

W&B CLI Version

0.15.3

Group

CARE

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 21 keys
- actor/hidden_dim:
  400
- actor/num_layer:
  3
- agent:
  "state_sac"
- agent/encoder_feature_dim:
  50
- agent/num_filters:
  0
- agent/num_layers:
  0
- batch_size:
  1,280
- encoder:
  "moe"
- eval_freq:
  10,000
- lr/actor:
  0.0003
- lr/alpha:
  0.0003
- lr/critic:
  0.0003
- lr/decoder:
  0.0003
- lr/encoder:
  0.0003
- num_eval_episodes:
  10
- num_tasks:
  50
- num_train_steps:
  1,000,000
- pal/dim:
  50
- pal/residual:
  "none"
- pal/shared:
  true
- seed:
  4

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 112 keys
- Env 0/episode_reward_env_index_:
  448.1989997090191
- Env 0/success_env_index_:
  1
- Env 1/episode_reward_env_index_:
  -1.940430194826674
- Env 1/success_env_index_:
  0
- Env 2/episode_reward_env_index_:
  -6.028767529685825
- Env 2/success_env_index_:
  0
- Env 3/episode_reward_env_index_:
  -622.2697141335199
- Env 3/success_env_index_:
  0
- Env 4/episode_reward_env_index_:
  159.18324483708076
- Env 4/success_env_index_:
  0
- Env 5/episode_reward_env_index_:
  0.1097927119013668
- Env 5/success_env_index_:
  1
- Env 6/episode_reward_env_index_:
  251.34189190809008
- Env 6/success_env_index_:
  1
- Env 7/episode_reward_env_index_:
  0.04219922179629798
- Env 7/success_env_index_:
  0
- Env 8/episode_reward_env_index_:
  153.7867999696153
- Env 8/success_env_index_:
  0
- Env 9/episode_reward_env_index_:
  12.6689757522676
- Env 9/success_env_index_:
  0
- Env 10/episode_reward_env_index_:
  -0.04345298353550219
- Env 10/success_env_index_:
  0
- Env 11/episode_reward_env_index_:
  352.73940835061245
- Env 11/success_env_index_:
  1
- Env 12/episode_reward_env_index_:
  0.3932956852807752
- Env 12/success_env_index_:
  0
- Env 13/episode_reward_env_index_:
  111.82213566997856
- Env 13/success_env_index_:
  0
- Env 14/episode_reward_env_index_:
  165.42605025718487
- Env 14/success_env_index_:
  0
- Env 15/episode_reward_env_index_:
  207.56484915818072
- Env 15/success_env_index_:
  1
- Env 16/episode_reward_env_index_:
  214.54398145134675
- Env 16/success_env_index_:
  0
- Env 17/episode_reward_env_index_:
  190.05528942705513
- Env 17/success_env_index_:
  1
- Env 18/episode_reward_env_index_:
  -272.34441023077505
- Env 18/success_env_index_:
  0
- Env 19/episode_reward_env_index_:
  205.1645253216106
- Env 19/success_env_index_:
  0
- Env 20/episode_reward_env_index_:
  164.91375548769824
- Env 20/success_env_index_:
  1
- Env 21/episode_reward_env_index_:
  248.91196323245163
- Env 21/success_env_index_:
  1
- Env 22/episode_reward_env_index_:
  13.2190132248149
- Env 22/success_env_index_:
  0
- Train/critic_loss:
  50.365334281921385
- Train/duration:
  3.236506462097168
- Train/episode:
  6,666
- Train/success:
  0.3

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

wandb-history

run-ugw9wvu7-history:v0