Skip to content

Commit

Permalink
polish(wzl): polish config notation
Browse files Browse the repository at this point in the history
  • Loading branch information
zerlinwang committed Dec 9, 2022
1 parent 45e364b commit 3807b77
Show file tree
Hide file tree
Showing 2 changed files with 3 additions and 1 deletion.
2 changes: 2 additions & 0 deletions mario_dqn/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -191,3 +191,5 @@ python3 -u evaluate.py -ckpt <CHECKPOINT_PATH> -v <VERSION> -a <ACTION SET> -o <
- 修复了evaluate.py以及mario_dqn_main.py中,预设动作维度不正确的bug,该bug曾经导致无法使用COMPLEX_MOVEMENT。感谢邹岷强同学的反馈。
## 12.08
- 修复了因为DI-engine更新导致的FinalEvalRewardEnv wrapper不可用的bug,感谢吴天鹤同学的反馈。
## 12.09
- 润色了一下注释,不影响程序运行。
2 changes: 1 addition & 1 deletion mario_dqn/mario_dqn_config.py
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@
evaluator_env_num=8,
# 评估轮次
n_evaluator_episode=8,
# 训练停止的分数(这里设置了一个不可能达到的分数
# 训练停止的分数(3000分可以认为通关1-1,停止训练以节省计算资源
stop_value=3000
),
policy=dict(
Expand Down

0 comments on commit 3807b77

Please sign in to comment.