state-machine

State-machine's group workspace

Group: xland-multi-task-test-v1

1-100

of 119

Tags

Notes

Author

state-machine

State

Finished

Start time

May 5th, 2024 11:51:21 AM

Runtime

1h 20m 49s

Tracked hours

1h 20m 43s

Run path

state-machine/xminigrid/6964495

Linux-5.4.0-171-generic-x86_64-with-glibc2.35

Python version

3.10.12

Git repository

git clone git@github.com:Howuhh/xland-minigrid-datasets.git

Git state

git checkout -b "ppo-rnn-multi-task-medium-finetune-XLand-MiniGrid-Extended-R4-13x13-medium-1m-60" 42b8e5b8106bc75371ee83d98733135e4ac35b29

Command

/workspace/xminigrid-datasets/src/xminigrid_datasets/collection/train.py --group=xland-multi-task-test-v1 --name=ppo-rnn-multi-task-medium-finetune --env_id=XLand-MiniGrid-Extended-R4-13x13 --benchmark_id=medium-1m --pretrain_multitask=False --ruleset_id=60 --total_timesteps=5_000_000_000 --seed=0 --num_envs=65536 --num_steps=256 --num_minibatches=64 --gae_lambda=0.999 --gamma=0.995 --context_hidden_dim=64 --context_dropout=1.0 --head_hidden_dim=64 --rnn_hidden_dim=512 --lr=0.0006788652208039325 --ent_coef=0.0001 --vf_coef=0.005 --use_bf16=True --decay_lr=False --eval_episodes=256 --pretrained_checkpoint_path=/workspace/xminigrid-datasets/checkpoints/medium-1m-R4-pretrained-extended-long-v12-XLand-MiniGrid-Extended-R4-13x13

System Hardware

CPU count	64
Logical CPU count	128
GPU count	1
GPU type	NVIDIA A100-PCIE-40GB

W&B CLI Version

0.16.6

Group

xland-multi-task-test-v1

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 39 keys
- action_emb_dim:
  16
- benchmark_id:
  "medium-1m"
- checkpoint_path:
  null
- clip_eps:
  0.2
- context_dropout:
  1
- context_emb_dim:
  16
- context_hidden_dim:
  64
- dataset_name:
  null
- dataset_num_histories:
  1
- dataset_path:
  null
- decay_lr:
  false
- ent_coef:
  0.0001
- env_id:
  "XLand-MiniGrid-Extended-R4-13x13"
- eval_episodes:
  256
- eval_seed:
  42
- gae_lambda:
  0.999
- gamma:
  0.995
- group:
  "xland-multi-task-test-v1"
- head_hidden_dim:
  64
- lr:
  0.0006788652208039325
- max_grad_norm:
  0.5
- name:
  "ppo-rnn-multi-task-medium-finetune-XLand-MiniGrid-Extended-R4-13x13-medium-1m-60"
- num_envs:
  65,536
- num_minibatches:
  64
- num_steps:
  256
- obs_emb_dim:
  16
- pretrain_multitask:
  false
- pretrained_checkpoint_path:
  "/workspace/xminigrid-datasets/checkpoints/medium-1m-R4-pretrained-extended-long-v12-XLand-MiniGrid-Extended-R4-13x13"
- project:
  "xminigrid"
- rnn_hidden_dim:
  512
- rnn_num_layers:
  1
- ruleset_id:
  60
- seed:
  0
- total_timesteps:
  5,000,000,000
- update_epochs:
  1
- upload_to_hf_repo:
  null
- use_bf16:
  true
- vf_coef:
  0.005
- wandb_logging:
  true

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 12 keys
- actor_loss:
  0.00005038356175646186
- entropy:
  0.0011915687937289476
- eval_dropout/lengths:
  676
- eval_dropout/returns:
  0
- eval/lengths:
  676
- eval/returns:
  0
- lr:
  0.0006788652390241623
- steps_per_second:
  1,042,024.6350243118
- total_loss:
  0.0000502986804349348
- training_time:
  4,798.3510484695435
- transitions:
  4,982,833,152
- value_loss:
  0.00000685495524521684

This run consumed these artifacts as inputs. Learn more

job

job-git_github.com_Howuhh_xland-minigrid-datasets.git_src_xminigrid_datasets_collection_train.py:v160

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

wandb-history

run-6964495-history:v0