google-deepmind / google-deepmind/dm_control

set_state() not working in walker

未关闭
#408 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看
主要语言
Python
星标
4.7k
派生
764
PR 合并指标
30 天内没有已合并 PR

描述

Hello,

I believe the `set_state()` function does not work for the walker environment.
The walker environment (https://github.com/deepmind/dm_control/blob/main/dm_control/suite/walker.py) defines its own Physics class, that extends the mujoco.Physics class, but does not extend the set_state() function.
I tried implementing the extra functions needed on my own but I believe I would need to build the repo, and that is not possible.
Am I missing something?
If not, can the extra functionality be added?

贡献指南

打开贡献指南

调研方向

Start with dm_control/suite/walker.py and compare its Physics class with mujoco.Physics, focusing on the inherited set_state() behavior. Confirm whether walker-specific state needs handling and identify the repository's existing build or verification path; done means set_state() works for the walker environment without breaking its physics state.

由索引模型根据 Issue 内容生成。

评估

技术栈
python
领域
machine-learning
Issue 类型
缺陷
难度
3/5
预计耗时
1-2 天
活跃度
停滞
描述清晰度
基本清楚
新手友好度
35/100

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。