state-machine

State-machine's group workspace

Group: debug

1-7

of 7

Tags

Notes

Author

state-machine

State

Finished

Start time

July 31st, 2024 10:06:46 PM

Runtime

1h 6m 4s

Tracked hours

1h 5m 59s

Run path

state-machine/xminigrid/a0wo41ip

Linux-5.15.0-1044-nvidia-x86_64-with-glibc2.31

Python version

3.10.13

Git repository

git clone git@github.com:corl-team/xland-minigrid.git

Git state

git checkout -b "meta-task-ppo" 2864a45afbeb97cf14b77832066461a6850ec013

Command

/home/jovyan/tmp/a5e18fdb/training/train_meta_task.py --config_path=configs/wip/config_best.yaml --enable_bf16=True --group=debug --total_timesteps=5000000000

System Hardware

CPU count	112
Logical CPU count	224
GPU count	1
GPU type	NVIDIA H100 80GB HBM3

W&B CLI Version

0.17.0

Group

debug

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 30 keys
- action_emb_dim:
  16
- benchmark_id:
  "trivial-1m"
- checkpoint_path:
  null
- clip_eps:
  0.2
- enable_bf16:
  true
- ent_coef:
  0.01
- env_id:
  "XLand-MiniGrid-R4-13x13"
- eval_num_envs:
  16,384
- eval_num_episodes:
  25
- eval_seed:
  42
- gae_lambda:
  0.95
- gamma:
  0.99
- group:
  "debug"
- head_hidden_dim:
  256
- img_obs:
  false
- lr:
  0.001
- max_grad_norm:
  0.5
- name:
  "meta-task-ppo"
- num_envs:
  16,384
- num_minibatches:
  16
- num_steps_per_env:
  12,800
- num_steps_per_update:
  256
- obs_emb_dim:
  16
- project:
  "xminigrid"
- rnn_hidden_dim:
  1,024
- rnn_num_layers:
  1
- total_timesteps:
  5,000,000,000
- train_seed:
  42
- update_epochs:
  1
- vf_coef:
  0.5

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 13 keys
- actor_loss:
  0.00005536484241019934
- entropy:
  0.256958931684494
- eval/lengths:
  2,436.29931640625
- eval/lengths_20percentile:
  768
- eval/returns_20percentile:
  16.95716094970703
- eval/returns_mean:
  20.512128829956055
- eval/returns_median:
  22.454437255859375
- lr:
  0.00004166662984061986
- steps_per_second:
  1,275,122.6884818554
- total_loss:
  0.017981838434934616
- training_time:
  3,921.191305875778
- transitions:
  5,033,164,800
- value_loss:
  0.041015625

This run consumed these artifacts as inputs. Learn more

job

job-git_github.com_corl-team_xland-minigrid.git_training_train_meta_task.py:v16

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

wandb-history

run-a0wo41ip-history:v0